A recent preprint paper, Forecasting Future Language: Context Design for Mention Markets, authored by a cross-institutional team that includes researchers from MIT, UC Berkeley, Seoul National University, and Kalshi, examined how input context should be designed to support accurate prediction in mention markets.

The researchers introduce a technique they refer to as Market-Conditioned Prompting (MCP). MCP treats “the market-implied probability as a prior and instructs the LLM to update this prior using textual evidence.” In other words, they use the current market price of a contract as a baseline and -- after giving the LLM context -- they ask the LLM to determine if that price is too high, too low, or fair.

The primary variables in their study were:

1. the contextual information that was provided to the LLM, which, in this case, consisted of news and/or prior earnings-call transcripts, and

2. how market probability (the contract price) was used.

They tested three variants:

1. market probability with context, 

2. market probability framed as a prior with context, and 

3. a blend of the market price and the MCP output (70% weight market, 30% on MCP result)

…on 856 Kalshi earnings-call mention market contracts that spanned 50 companies and 70 earnings events from April through December 2025. The forecast cutoff was set at 7 days before each earnings call, and the LLM (GPT 5.1) was given up to 100 company-related news articles and the prior quarters’ earnings call transcripts as context.

One of their primary findings was that “richer context consistently improves forecasting performance.” 

This makes sense, both when it comes to prediction more generally and when providing context to LLMs. The quality of your inputs matters. If you’re making dinner how your meal ends up will ultimately be constrained by the quality of your ingredients. It’s the same case here. Better context leads to better results.

Now, regarding the type of inputs, the researchers found that providing relevant news to the LLMs helped, and that including prior transcripts yielded larger gains than news alone.

They also found that MCP improves over naive use of market probability. When they simply asked the LLM to predict the probability a particular word would be mentioned, they used the following prompt:

As you’ll see in a table below, this underperformed the market. This is interesting because even with transcripts, news, and market probability as context, without instruction to assess or revise the market signal, the LLM underperformed the market baseline.

Researchers also tested with prompting (i.e., their MCP method), where they framed the market probability as a prior to be evaluated and updated. Here they used this prompt:

They found that their MCP method yields better-calibrated forecasts than just giving the LLM access to the probability itself. They also found that MCP does best when market signals are uncertain:

You can see their comparison below:

At an intuitive level, this makes sense. In order to identify whether or not a market is mispriced, you need to identify information that (presumably) the market hasn’t priced in. But as discussed above, just having this information isn’t enough to beat the market baseline. You need to properly instruct the model about what you want it to do with it.

Just having the information isn’t particularly useful if you don’t know what to do with it.

They also used what they call MixMCP. The idea of MixMCP is to correct for potential overreactions LLMs might make to weak or noisy cues. To account for this, researchers used a convex mixture that anchors on the market, as seen below:

Where:

• The p_i^MKT is the market price, and

• p_i^MCP is the market probability on a 0-100 scale that the LLM provided as an output

They chose to use: α = 0.7

They found that MixMCP consistently outperforms the market baseline, and that “by dampening the LLM’s posterior update with the market prior, MixMCP yields more robust predictions than either the market or the LLM alone.”

However, the improvement in Brier score from Market Probability to Mix MCP is 0.1402 to 0.1392. This improvement of 0.001 is extremely small and doesn’t have much statistical significance. The accuracy improvement is 79.8% to 80.3%, which is about 4 additional correct predictions out of 856.

Now, there are some limitations at play here. 

The researchers explicitly point out that this work doesn’t necessarily generalize to other contract types. The researchers are right when they point out that LLMs are well suited for earnings call mention markets because the job here is predicting what might be said based on past context, which is -- to put it simply -- essentially what LLMs do.

The impact of the MixMCP might also be overstated. When you take a convex combination of two imperfectly correlated forecasters, the blend almost always produces a lower Brier score than the worse of the two, which happens even if the LLM has zero real informational edge.

For future studies, it would also be interesting to consider:

• how other LLMs would perform;

• to look more closely at how factors like trading volume, liquidity, and number of traders impact the outcome; and

• how this would compare to a human baseline (essentially asking the question: How would an LLM compare to an individual forecaster reading the same material?)

One thing this preprint leaves me considering is the question of how much of the edge here actual coming from the LLM.

Is the LLM actually extracting any signal that couldn’t otherwise be found from just reading through the material provided to it? Again, the information and the LLM alone didn’t result in a better outcome. The LLM is missing “signal” when it isn’t prompted to consider a prior. This would indicate that the LLM doesn’t necessarily know to ask what it doesn’t know to ask. I’m left wondering what other opportunities for edge are missed if someone overrelies on an LLM for their analysis.

The market composition matters a lot. Whether the market participants are highly informed or taking positions that are more aligned with gambling will greatly impact the initial market price. If the participant pool is small and not deeply informed, the “market baseline” is weaker than it sounds, and any improvement may just reflect beating a thin market rather than beating the crowd’s wisdom.

But I still think there are some concepts from this that are helpful to keep in mind when approaching any market…

Context quality drives everything. This is true in so much of life, and it’s true here. Whether you’re feeding the information to an LLM or yourself, quality information matters. (And there’s probably a point here about the importance of taste that I’ll likely make in some other piece.)  

So, invest in context quality over model sophistication. The gap between no-context and full-context was far larger than the gap between prompting strategies. Be patient with yourself at this step, and remember that identifying quality information and insights takes time. There’s a reason it hasn’t been priced into the market: It takes work.

Instead of asking an LLM to predict an outcome, ask it to adjust based on the current market price. Give LLMs more explicit instructions. If you want an LLM to “reason” or do something that appears to be a facsimile of that, you need to walk it step-by-step through what you’re doing. Imagine you’re coding. Think in if-then statements. What are the caveats to the caveats? 

And even if the gains are marginal, marginal gains can add up. As Bob Mercer of Renaissance Technologies has famously been quoted saying, “We’re right 50.75% of the time, but we’re 100% right 50.75% of the time. You can make billions that way.”

Subscribe to OddChain to get prediction market insights and potential edge sent straight to your inbox.

Today’s the day. The Oscars are nearly upon us. To make your life easy, I rounded up a bunch of research and analyses, and put together a spreadsheet of how some anonymous Oscar voters voted to help shed light on how they think. We also explore some interesting positions you can take that seem likely to end up in your favor.

One thing to keep in mind when you approach these markets is that you’ve got to separate the social media sentiment from the actual votes. What you think or Twitter thinks or your friend thinks doesn’t really matter because the Academy is made up of just over 10,000 members and their votes decide who wins.

There’s not a lot of demographic data available on the Academy, and if history continues to repeat itself, they’re not about to release more data any time soon. In 2012, the Los Angeles Times reported that there were:

• 5,765 voting members,

• 94% were Caucasian,

• 77% were male,

• Black people made up about 2% of the academy, 

• Latinos made up less than 2%

• the median age was 62, and 

• people younger than 50 made up just 14% of the membership.

However, since then, the Academy’s demographics have changed. Following the #OscarsSoWhite virality in the wake of the 2015 and 2016 nominations, the Academy launched an initiative to incorporate more diversity into the Academy. 

Perhaps the most “up to date” data that I could find was this overview of updated membership from 2020. And according to USA Today, in 2024, the Academy was made up of “10,894 members (9,905 voting members) and consists of 35% members identifying as women and 20% from underrepresented ethnic and racial communities.” And there has been an increase in international membership over the past years as well. For example, 55% of the new members in 2025 were international.

So, when you’re predicting how the Oscars will turn out, you’re predicting how these people will vote. 

Yet another complicating factor is that members don’t necessarily even watch all the movies before they vote. That said, the Academy is trying to remedy this. This year, there’s a new rule that “requires” members to watch all the nominees in a category before they can vote. Now, they can just check a box and say that they did, so this is essentially the honor system, but – anecdotally – according to some industry interviews I’ve listened to, it does seem to have made a bit of a difference.

Now, in an attempt to capture how some of the Oscar voters approached this year’s votes, I went through some “anonymous Oscar voter” articles that were floating around on the internet. 

Here is a spreadsheet of the results, but I also want to share some quotes that I found particularly interesting, such as:

And:

Then there’s this exchange:

This is an incredibly small sample size of the Oscar voting base, but it does provide some insight into how they approach their votes. And, at the very least, I find it interesting.

Then there’s what the pundits are saying…

What the Pundits Are Saying

If you Google Oscar predictions, you’ll find dozens of articles, but here are some that I found worth sharing…

One thing you’ll probably notice looking at all of this is that there is remarkable consistency with a lot of the predictions people are making, and that the categories that are still up in the air are actually still up in the air. And while this makes for a particularly exciting Oscars tonight, it makes for a less than ideal environment for predictors as unless you’re a select employee of  PWC who’s about to commit insider trading, it’s very unlikely that you’re going to come across genuine edge.

So, where do things currently stand?

There are some near guarantees…

Best Picture is a two-horse race between One Battle After Another and Sinners. I still think One Battle is the favorite to win, though I also wouldn’t be surprised if Sinners won. Talking with people in Los Angeles, however, it does seem like One Battle is still the favorite to win.  

Best Original Screenplay and Best Adapted Screenplay will almost certainly go to Sinners and One Battle After Another respectively. Not much to go into there.

Animated Feature will probably go to KPop.

Best Director is probably going to PTA. 

For Best Actress, Jessie Buckley is basically a lock. And it’s widely believed to be essentially the only sure thing of the night. 

Best Visual Effects is most likely going to go to Avatar.

Best Actor is largely a fight between Michael B. Jordan and Timothée Chalamet. While Timothée might not win, if he loses it won’t be because of his comments on ballet. The fateful event with Timothée and Matthew McConaughey didn’t occur until after Oscar voting closed. Of course, there's a chance of insiders, but far far far more likely is that the sentiment of the overreaction from the news swung this. So there’s definitely some mispricing here, but it’s still going to be a close race.

Regarding Best Supporting Actor, even if he decides not to show up, Sean Penn is still a favorite. He won SAG and the BAFTA, and the last actor to win both and lose the Oscar was Christopher Walken back in 2002.

Amy Madigan is likely underpriced for Best Supporting Actress at around 50%. 

Casting will likely go to Francine Maisler for Sinners. It’s the first year this category has ever been around, so there’s not much to go off of. She has quite the resume. 

Production and Makeup and Hairstyling will likely go to Frankenstein.

At around 76% One Battle is likely underpriced for Best Cinematography. 

There are some others that I think are underpriced as well…

One Battle for Best Editing is likely underpriced.

Best International Feature will probably go to Sentimental Value. This is likely underpriced at around 70%. Though, The Secret Agent could still eke out a win.

The rest are pretty aligned with many of the predictions above, and there isn’t really a lot of edge to find, but I want to spend a bit of time on the best live action short. 

First of all, to improve your accuracy here, you can watch some of the movies. Fewer people have probably watched these than, say, the nominees for best picture.

A Friend of Dorothy is no longer online, but you can watch Two People Exchanging Saliva, The Singers, and Jane Austen’s Period Drama.

I think Two People is far more likely to win than the odds currently indicate. The Singers were really enjoyable, but it was more of a vibe. And, at least right now, if you think it’s a coin flip between these two you can do a bit of arbitrage and put $100 on each and still come out in the green.

Subscribe to OddChain to get prediction market insights and potential edge sent straight to your inbox.

This afternoon (or early evening depending on where you are) Adobe will have their earnings call at 2 pm PT / 5 pm ET (3/12/26).

Kalshi has a market for it, so we wanted to break down some of the mentions in the market, provide some context, and identify some potential mispricing.

Coming into this call, Adobe has beaten analyst estimates for revenue in 12 straight quarters. And analysts expect that today will represent a new quarterly record, projecting Adobe will report first-quarter earnings per share of $5.46, up from $5.08 a year earlier. Polymarket currently has the odds for this at 92%, which aligns with analyst sentiment. 

Chart from Synthesis.

Now, looking to today’s call, we used our earnings analysis tool to break down Adobe earnings calls going back to 2020. 

Please note that some of the above mentions might include those of analysts. Per Kalshi rules, these wouldn't count toward resolution. And it goes without saying that past mentions don’t guarantee something will be said, but they do provide a good jumping off point for deeper analysis. If you want to take a look at the transcripts themselves, you can do so here.

At the time of writing, here are where prices on Kalshi currently stand:

Generative AI / Gen AI is pretty much a shoe-in, and is priced appropriately. Gen AI is a core strategic narrative of Adobe, and it’s appeared in earnings calls since Q4 2022 without fail. There’s no real edge here, but you can probably pick up some pennies if you want.

Now, Semrush is where we think there might be some mispricing. Current odds sit at 89% and we reckon they’re a bit higher. 

Last November, Adobe announced an agreement for Adobe to acquire Semrush in a $1.9 billion all-cash transaction. This would be the largest deal since the now-aborted Figma acquisition, which was mutually terminated due to insurmountable regulatory roadblocks. And, while that’s good for competition, it doesn’t have much to do with this particular market, so we’ll be moving on…

The deal received HSR (Hart-Scott-Rodino Act) clearance in January, and shareholder approval early last month. The deal is expected to close this year.

In the Q4 FY2025 Adobe earnings call transcript, "Semrush" appeared 9 times (1 was an analyst mention), mentioned by both Shantanu Narayen and Anil Chakravarthy: 

In past earnings calls, Adobe continued to mention acquisitions -- such as Workfront, Frame.io, and Figma -- after the acquisition, so that’s a lot of words to say chances of a Semrush mention are pretty high.

Fitting nicely with the potential Semrush mention…

LLM Optimizer is an interesting one. Adobe’s LLM Optimizer is a product designed to help brands optimize content for AI chatbots and LLM-powered search.

It launched in Q3 FY2025 and was mentioned substantively in both Q3 (7 times) and Q4 (4 times), anf is likely to be mentioned this time around in the context of the agentic web, brand visibility, and customer adoption metrics.

One interesting bit of note is that in a Semrush release, published earlier today, title “Semrush Unveils New Brand Identity to Command the AI Search Era” discussing “its strategic evolution from a search toolset to a unified intelligence engine that drives brand visibility for businesses of all sizes.” Writing that “Semrush is uniquely positioned to lead the convergence of SEO and AI Search and help ambitious markets understand where discovery happens and how to win visibility within it.” And most pointedly:

What’s interesting here is that the term LLM Optimizer wasn’t used. This may represent a shift in verbiage. Though, in a recent press release, discussing a Major League Baseball partnership, they mention Adobe LLM Optimizer, so the term might not be dead. 

We’re also leaning NO on Google Cloud. The Adobe and Google Cloud partnership was announced last October. In that press release the phrase “Google Cloud” was used many times. However, in the call following the partnership announcement, the phrase “Google Cloud” wasn’t used. Instead, we see the use of phrases like “Google” and “Google Gemini.” So, even if this partnership is discussed -- directly or indirectly -- a different phrase may be used.

In advance of tomorrow’s Adobe earnings call at Mar 12, 5:00pm EDT, we wanted to share some of the resources we gathered during our research. The last two years of Adobe’s earnings call transcripts are available on their website. You can check out the links here:

But we also understand that you might want to go further back, so below we’re providing the transcripts for Adobe back through 2020.

Adobe Earnings Calls Fiscal Year 2023

Adobe Earnings Calls Fiscal Year 2022

Adobe Earnings Calls Fiscal Year 2021

Adobe Earnings Calls Fiscal Year 2020

Subscribe to OddChain to get prediction market insights and potential edge sent straight to your inbox.

This afternoon (March 5, 2026) at 4:15 p.m. EST, Costco will hold its Q2 2026 earnings call. We dug into the possible mentions in the Kalshi market and used our earnings call tool to help us evaluate how we think this market might turn out.

Here’s what we’re thinking, along with some context to help you come to conclusions of your own.

The words in play:

Costco’s mention market for the earnings call today is an interesting one. While we think there are still some mispricings, as you can see below, the odds for a majority of the words sit at above 90%.

We attribute this, in part, to the remarkable consistency of mentions in past Costco earnings calls.

Inflation, Membership, Expansion, Kirkland, and Inventory are all frequently mentioned, and you can probably pick up some pennies on these if you’d like.

Tariff also sits at a high likelihood, with the recent Supreme Court decision striking down Trump’s emergency tariffs on February 20, 2026 and Costco’s active lawsuit providing catalysts for the mention.

Take a look at the counts from previous earnings calls for the mentions at play:

Now, we think that NOs on Automation and Valuation seem pretty good.

Although Costco has invested in technology such as AI-powered pharmacy inventory, self-checkout prescan technology, and AI tools in the gas business, we think there is a solid chance the word automation is avoided even if it is discussed indirectly.

Costco President and CEO Ron Vachris used the following language on their Q1 FY2026 earnings call when discussing their implementation of AI:

We think their emphasis on people and employees contributes to the case that they will avoid the word “automation” even if they are discussing the process of automation. Instead, they may use words like “technology,” “AI,” “prescan,” “digital,” or “efficiency,” rather than the word "automation." 

Regarding valuation, Costco management historically avoids commenting on their own stock valuation. This is standard practice for public company executives, as valuation discussion is typically the domain of analysts and investors, not company representatives. Plus, discussion of it by a company representative could be seen as forward-looking guidance on stock price.

Regarding dividend, even though Costco just declared a regular quarterly cash dividend in January, this is usually relegated to the press release text rather than the spoken transcript.

And though there is some analyst speculation about an impending special dividend announcement, we don’t see a strong signal for that. Costco historically announces special dividends through press releases rather than on earnings calls. This will likely depend on an analyst question, so we consider it a toss-up.

Hot dog is an interesting one that we think leans toward NO. Traditionally, hot dog has been mentioned in the context of media rumors of price increases or in response to analyst questions. The general familiarity and “meme-ability” of the Costco hot dog lends itself to a favorite-longshot bias that may contribute to the YES position being overpriced. We think unless an analyst questions them, a NO here is more likely than a YES.

This afternoon (March 3, 2026) at 2:00 p.m. PST / 5:00 p.m. EST, CrowdStrike will hold their Q4 FY 2026 earnings call. There’s a mention market for it on Kalshi, so we’re going to talk about it.

There are some positions that appear to be priced accurately.

Consolidation, Falcon Flex, and Acquisition are at 98%, 97%, and 92%, respectively.

Consolidation was mentioned on almost every earnings call since 2019. Since platform consolidation is one of CrowdStrike’s core strategic narratives, it’s likely they’ll continue to mention the word.

Falcon Flex, another key business driver, has been mentioned consistently since its release in 2023. As of Q3 FY2026, Flex had over $1.35 billion in ending ARR, growing 200%+ YoY. Q4 is a seasonally strong quarter, so new Flex deal metrics will likely be reported.

Acquisition was mentioned on nearly every call we reviewed. This might be slightly underpriced at 90% as CrowdStrike announced three acquisitions in the Q4 FY2026 timeframe:

• Pangea ($260M, September 2025) for AI security/shadow AI detection;

• SGNL ($740M, January 8, 2026) for identity security in the AI era; and

• Seraphic ($420M, January 13, 2026) for browser runtime security.

Likewise SGNL/Signal, at the time of writing, is at 95%. Although “Signal” has appeared a few times in past earnings calls, it was used as a generic term rather than in reference to the company SGNL that CrowdStrike just acquired.

We think the odds of this mention are pretty solid because, as mentioned above, the SGNL acquisition was announced on January 8, 2026, and management will almost certainly discuss the deal’s strategic rationale, the identity security market opportunity, and integration plans.

Now, before we look at a few more of the possible mentions, let’s take a look at a breakdown of the frequency of mentions on past CrowdStrike earnings calls that we put together using our earnings call tool.

AWS is currently at 85%. There is a solid chance of a mention. CrowdStrike named AWS 2025 Global Security Partner of the Year last December, and the CrowdStrike-AWS-NVIDIA 2026 Cybersecurity Startup Accelerator concluded March 3, 2026.

Historically, NVIDIA is mentioned with less frequency than AWS, but should CrowdStrike discuss their recently Cybersecurity Startup Accelerator, it’s likely NVIDIA will get the mention as well.

Hyperscaler seems quite likely as well. CrowdStrike’s cloud security strategy increasingly references hyperscaler partnerships (AWS, Azure, GCP). And the SGNL acquisition press release explicitly mentions extending identity security to “hyperscaler cloud access layers” including AWS IAM.

Odds for YES on China, Hack/Hacker, and Outage sit below 50%.

And this is in line with our analysis.

Regarding China, in the Q4 FY25 earning call, CrowdStrike said:

This was the only mention of China in a FY25 earnings call. There were no mentions until Q3 FY26, when a representative of CrowdStrike said:

In CrowdStrike’s recently published 2026 Global Threat Report China was mentioned multiple times in the context of China-Nexus adversaries. The YES here may be a bit underpriced, but due to the inconsistency of mentions on past earnings calls, we don’t have a strong position on it either way.

With regard to outage, many of these mentions have been by analysts and it is less likely to be said by management. So leaning toward NO on this one seems safer than a YES.

Hack/Hacker also seems to be a likely NO, as word is infrequently mentioned on past calls, and there are other words they may use, such as adversary, threat actor, intrusion, or breach.

Right now, approximately $54 million in prediction market contracts may hinge on what "for each Contract" means. Kalshi, the CFTC-regulated prediction market exchange, explicitly does not run war or death contracts. But what if someone dies?

It’s more than reasonable not to want to potentially incentivize assassinations. We’d go so far as to say that’s a positive thing, in most cases. However, in most markets involving people, death is always a potential outcome. So, should an untimely death occur, how does the market account for that potential outcome?

In the case of the “Ali Khamenei out as supreme leader?” market, Kalshi’s rules attempt to account for the outcome of his death, while at the same time being careful that the market is not about the question of whether or not he'll die.

The language Kalshi used is as follows:

The first two sentences are reasonable. If Khamenei dies, they'll settle the market based on the last traded price. If that price doesn't make sense, a committee figures out a fair split.

The last sentence is where the potential ambiguity arrives:

The ambiguity is: what exactly do they mean by “$1 for each Contract.” Is each position considered a “Contract?” Is each YES-NO pair that sums to $1 considered a “Contract?” Also what do they consider to be the trigger for "the avoidance of doubt?"

In Kalshi’s rulebook, they define “contract” as meaning “any contract, agreement, or transaction approved for trading on Kalshi pursuant to these Rules.”

And in their explanation of how to work with event contracts they say that “YES and NO contracts always come in pairs” while also describing the ability to buy individual YES and NO contracts.

So, which one is it? And who gets the $1. Everyone?

So, the potential issue here is that under one reading of this provision, if Khamenei dies, Kalshi, according to their own rules, may be required to distribute $1 to each contract.

A YES contract gets $1, and a NO contract gets $1. So Kalshi collected $1 (presumably the sum of the Yes and No positions) and now owes $2.

Now, Kalshi almost certainly didn't mean this. The most reasonable interpretation is that "for each Contract" is clarifying that the full dollar gets distributed to traders, split between sides according to the last traded price. That’s the likely intended reading, and the position they'll almost certainly take. But there is, at minimum, a colorable argument that the plain text says what it says. This is because the legal principle of contra proferentem holds that ambiguous contract language is interpreted against the party that drafted it. Kalshi wrote the rules and chose the words, so if a trader bought both YES and NO contracts and then argued in arbitration that they're owed $1 on each, they (theoretically) would have a sentence to point to.

All of this means that one of the most actively traded geopolitical prediction markets in the world, a market tracking the possible end of a 35-year theocratic regime, with implications for nuclear proliferation, regional stability, and oil prices, may contain a drafting error that entitles everyone to free money.

Below is the Q3 2025 earnings call transcript for eToro.

This transcript is provided for informational purposes only and may contain errors, omissions, or inaccuracies. While we strive to ensure the content is as accurate as possible, it has not been verified by the company or its representatives. This transcript should not be relied upon as a verbatim record of the earnings call or as the basis for any investment decisions. Please refer to the company's official filings and recordings for authoritative information.

Presentation

Daniel Amir: My name is Daniel Amir, Head of Investor Relations.

This webcast is being recorded and will be available for replay in the Investors section of eToro's website.

Our earnings press release, investor presentation and October monthly spreadsheet is now available on our website at investors.etoro.com.

Today I'm joined by Yoni Assia, our CEO; and by Meron Shani, our CFO. Following the prepared remarks, we will conduct a Q&A session and answer questions from both institutional research analysts and a selection of the most upvoted question previously submitted by eToro's retail shareholders.

But before we begin, I want to note that today's discussion contains forward-looking statements including statements about goals, business outlook, industry trends, market opportunities, expectations for future financial performance and similar items, all of which are subject to risks, uncertainties and assumptions.

And you can find more information about these risks and uncertainties in the press release that we issued today and in the risk factors section of our filings at sec.gov.Actual results may differ.

And we take no obligation to revise or update any forward-looking statements.

Finally, during today's meeting, we will discuss non-GAAP financial measures.

These non-GAAP financial measures are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP.

Definitions and reconciliation of GAAP to non-GAAP measures is available in our press release, investor presentation and on the sec.gov website as applicable.

With that, I will pass the call to Yoni.

Yoni Assia: Thank you, Daniel. And thank you, everyone, for joining us today.

Welcome to our third quarter's 2025 earnings call. After Meron and I conclude our prepared remarks, we'll open it up for questions.

We're proud to share another strong quarter that reflects eToro's continued momentum and a growing strength as a global leader in trading and investing. This performance with net contribution up 28% year-over-year to $215 million and adjusted EBITDA rising 43% to $78 million, delivering solid operating margins of 36% is a testament to the vision, resilience and scalability of our diversified business.

What excites us most is the remarkable pace of innovation unfolding across eToro, with AI accelerating our product development.

We're not only enhancing the platform that our users know today.

We're reshaping the future of investing.

We're expanding into new frontiers in crypto, tokenization and AI, while also broadening our global reach to serve an ever-growing community of smart investors. This wave of innovation is delivering tangible results.

Record assets under administration, robust growth across all segments and unprecedented engagement in copy trading. Together, these achievements underscore how our investment in technology is fueling growth, deepening client engagement and bringing us ever closer to our mission, which is to empower everyone to trade and invest in a simple and transparent way.

Our key performance indicators accelerated meaningfully this quarter. Funded accounts grew 16% year-over-year to $3.73 billion with organic funded accounts expanding at a double-digit rate. This growth demonstrates the growing appeal of eToro's differentiated platform, the trust we've built with our global community of investors and the success of our disciplined and data-driven marketing approach.

Assets under administration reached an all-time high of $20.8 billion, up 76% year-over-year, fueled by growth in new deposits and strong user investment returns in both crypto and equities this year. The positive momentum in our KPIs has continued into Q4 as shown in the October results released today. These results show our strategy in action.

We're democratizing and investing by making it simple, accessible and social while combining multi-asset trading with smarter products like copy trading and smart portfolios to empower investors everywhere.

As we execute on this vision, we're confident in our ability to deliver lasting value for our users and shareholders.

To achieve the strong performance we delivered in Q3, we've remained focused on executing across our four strategic pillars of trading, investing, wealth management and neobanking while accelerating product innovation across the platform.

Let me share a few highlights across each of these areas.

In trading, we're continuing to expand market [act] globally.

We continue to expect our 24/5 trading.

We began offering retail investors access to stocks listed on NASDAQ Nordic changes throughout our partnership with NASDAQ. This expansion comes amid rising retail investor interest in the global markets.

Today we offer access to [22] different exchanges worldwide and plan to increase to over 30 in 2026.

We've expanded futures across Europe and launched spot-quoted futures and partnership with the CME Group. Retail participation continues to grow in these markets, and we are pleased to give local traders the ability to execute in a more advanced and diversified strategies.

Our expanded to futures also establishes the infrastructure to support prediction markets in the future.

In investing, we're bringing together the power of community and AI to drive deeper user engagement on eToro's platform and help investors make smarter decisions with AI.

Our Pro investor community has grown to over 4,000 investors globally, supported by our new Pro investor program. This program is designed to support users on their journey to certify themselves as professional investors.Our Pro investors gain exposure and grow their profile by sharing their expertise with eToro's global user piece.

In addition, over 130 Pro investors now have north of $1 million assets under management with our top row investor growing from $50 million to over $250 million in 2025 alone, a milestone that reflects the growth and influence of our Pro investor program.

Last month, we launched CopyTrading in the U.S.

We're excited to bring our flagship product to the world's largest capital market. Copy trading today is experienced across approximately a third of our users.

With the launch of Copy Trading in the U.S., we anticipate increased user engagement and platform traction.

In AI, we're entering the next great leap in trading and investing.

One that puts the power of professional grade technology direct in the hands of every investor with our newly launched eToro apps, users can build share and scale their own investment tools across our global community.

What was once the privilege of sophisticated institutions is now being democratized.

Our AI-driven insights from sentiment to decision analysis, enabling investors to think and act as institutional investors by creating their own strategies, dashboards and innovations that shape the future of investing.

In Wealth Management, we launched a new subscription model for the eToro Club, our loyalty and rewards program, offering members premium benefits and exclusive features.

The subscription offers members smarter tools, monetary benefits and more personalized support while driving incremental engagement. Furthermore, we've expanded the eToro long-term savings offering for our U.K. users with the new eToro Cash ISA and Cashback rewards and management and do-it-yourself ISAs. And in Australia, we advanced integration of spaceship, giving users direct access to superannuation products through the eToro platform.

These developments demonstrate our commitment to continuously enhancing wealth management offering, which is a multitrillion-dollar opportunity.

With the Australian superannuation market valued at over $2.5 trillion and the U.K. individual savings account market exceeding $1.3 trillion.

In neo banking, we continue to enhance our global platform by delivering localized experiences that strengthen user trust, drive adoption and support sustainable growthacross our key regions. This quarter, we expanded our localization offering in the UAE, Singapore and Australia. Furthermore, across the U.K. and Europe on top of the 4% cash back in stocks, we now offer also a 1% cash back in stocks for new crypto deposits.

We plan to expand this offering to more countries in the future.

We're also very pleased with the strong traction we've seen with the eToro Money Card available today in Europe and the U.K., which saw a 2.4x increase in cards issued quarter-over-quarter.

Looking ahead, we believe there are five key factors that will drive eToro's continued growth and advance our mission to open the global markets, connect users to leading investors and give them tools they need to grow their knowledge and wealth. Number one, advancing at the forefront of innovation. Number two, continuing to expand globally. Number three, expanding our U.S. presence.

Number four, broadening our product offering with AI. And number five is leading to inevitable elastic macro trends.

We are focused on positioning our business to capture the significant growth opportunities presented by these five factors.

First, advancing the platform at the forefront of innovation. From day one, product innovation has been at the core of who we are.

We've built a strong track record of identifying early major trends like social trading and crypto and turning these trends into products that bring real financial utility to our users.

Our social investing products continue to define who we are today as a company. From the outset 15 years ago, eToro set out to transform retail investing experience by pioneering the concept of social investing, empowering individuals to learn from one another and invest together.

We believe there is a tremendous value and shared knowledge and collective insight.

Over the years, we have led the industry with innovations such as our social network, copy trading, crypto investing, fractional shares, machine learn driven analytics and our smart portfolios. Together, these capabilities make investing more inclusive, informed and collaborative for our growing global community.

Artificial intelligence is the next frontier in investing. And we're already seeing the impact across our business.

Our AI analyst story launched last quarter has already been used by over a third of our club members, a strong sign of engagement and potential. As we expand our AI capabilities, we're giving an investor powerful tools to make smarter decisions, improve performance and deepen their connection to our platform.

Our vision is also centered on building open, tokenizing borderless market to help millions to build wealth.

We're bringing U.S.-listed equities to the blockchain enabling 24/5 trading of over 500 prominent stocks with plans to move forward to 24/7 as the market's involved.

We're currently developing a crypto wallet that will open the opportunity for our users through decentralized finance to participate in innovations like on chain prediction markets, lending, loans and swaps of their assets of millions of different crypto assets and connect our tokenized assets ecosystem and benefit from tokenization of real-world assets, which will happen over time.

We expect these more advanced crypto products to continue to increase our client engagement within the crypto industry.

Second, continue to expand globally.

We're proud to serve users in more than 75 different countries across the world. The diversification of our revenues across geographies is the core strength of eToro and a differentiator of our business.

In Q2, we expanded our footprint in Asia by launching our capital market services license in Singapore, a meaningful step towards scaling our business in that region.

Looking ahead, we plan to further localize our offering in key markets and continue to expand globally in region where we're underpenetrated.

As part of our broader strategy to solidify eToro's position as a truly global platform serving investors everywhere.

Third, expanding our U.S. presence.

We're seeing solid growth in the U.S. Year-to-date, new funded accounts have already surpassed those of 2024 and we've achieved that growth while remaining disciplined and efficient in our marketing spend, consistent with our strategy. To continue this growth, we're focused on bringing the full product offering for which we are known globally into the U.S. market.

For example, in the third quarter, we brought in an crypto offering in the U.S. to include more than 100 different crypto assets and introduced staking in the U.S.

As a result, we saw a 3x increase in our crypto volumes quarter-over-quarter in the U.S. market. Additionally, last month, we announced the launch of Copy Trading in the U.S., a major milestone that lays the foundation for our next phase of expansion in this market. The copy trading product is a significant driver of user engagement with our platform is reflected in the increase of the number of trades we've seen this year.

We expect Copy Trading to be an important driver of growth in the U.S. going forward.

Fourth, broadening our product offering.

In our core markets, we're focused on increasing engagement and share of wallet by broadening what we offer across our four strategic pillars. Trading, investing, wealth management and neo banking.

We continue to expand each of these areas quarter after quarter.

As mentioned, recent highlights include the launch of the savings products in the U.K. and France, expansion of the eToro Money card across Europe, the (inaudible) futures and options across key European markets and introducing localized trading and cash management capabilities in new regions as well. This approach has proven successful over time by driving both higher user retention and growth in assets per account.

And fifth, leaning into elastic macro trends.

Over the next two years, we will see the largest generational transformation of wealth in history with over $150 trillion moving to younger generations globally. These younger generations are digital first, and they're far more engaged with equities encrypted than their parents.

On their back end, we also expect to see $100 trillion moving on change as capital markets transform into digital assets. Moreover, in many non-U.S. markets, retail participation remains below U.S. levels.

We believe this cap will continue to narrow as younger generations, more inclined to invest in crypto and equities than their parents enter the market, offering us a long runway for growth across our global franchise. These secular shifts are powerful tailwind that will continue to support both assets under [reinstation] and new account growth over the long term.

Altogether, these five drivers gives us strong confidence in our ability to deliver sustainable, profitable growth and position eToro as the global broker of choice.

Lastly, we announced today a $150 million share repurchase program, underscoring our confidence in eToro's long-term growth prospects and our continued commitment to delivering value for shareholders.We believe that our stock is undervalued. And given our significant cash generation, we have the flexibility to buy back shares.

Our strong cash position also gives us the ability to consider M&A opportunities to drive inorganic growth.

We're evaluating a range of opportunities, though we will remain disciplined in any transaction we pursue.

In some, this quarter's results and KPIs for October clearly demonstrated that our strategy is working.

We're achieving double-digit growth in organic funded accounts, significantly increasing our assets under administration, expanding net contribution and adjusted EBITDA margins and doing so while entering new markets and launching new products that strengthen our leadership for the future.

With strong execution, financial discipline and a relentless focus on innovation, we believe we're very well positioned to drive long-term value, and we're just getting started.

With that, I will pass now along the call to Meron, our CFO, to discuss our financial results.

Meron Shani: Thank you, Yoni.

As Yoni mentioned, we are very pleased with our third quarter results. Third quarter net contribution grew 28% year-over-year to $250 million and adjusted EBITDA grew 43% year-over-year to $78 million.

In line with our focus on diversified, profitable revenue growth, our adjusted EBITDA margin was 36%, expanding 370 basis points from a year ago. These results are consistent with our view that our long-term adjusted EBITDA margins will increase from these levels.

Our momentum has accelerated in the third quarter.

Assets added administration for the quarter increased 76% year-over-year to a record of $20.8 billion, while our funded accounts grew 16% year-over-year to $3.73 million. The growth was driven by strong user acquisition and retention efforts and ongoing disciplined marketing.

Let's take a closer look at our third quarter financials by business plans compared to a year ago.Our net trading contribution from crypto grew 229% year-over-year to $56 million, which was largely driven by higher invested amount per trade and increased crypto activity, especially in the month of July and August.

Our net trading contribution from capital markets, equities, commodities and currencies declined 21% year-over-year to $73 million as investors shifted activity between crypto and capital markets. The 15% rise in the number of trades was driven by strong momentum across Copy Trading with record inflows into Copy, a clear reflection of growing customer engagement and platform strength.

Net interest income contributed $62 million, up 44% year-over-year, largely driven by a 52% increase in higher interest-earning assets due to an increase in customers' cash deposits, customers' margin book, staking and corporate cash. eToro money contribution grew 50% year-over-year to $21 million, largely driven by an increase in total money transfers.

In the third quarter, adjusted OpEx was $137 million, flat quarter-on-quarter.

Our adjusted selling and marketing expense was $47 million or 22% of net contribution.

Our business model provides us with flexibility in our selling and marketing expense, where approximately 70% of our expense is dynamic.

Adjusted R&D and G&A and operating expenses were $36 million and $54 million, respectively.

Our adjusted diluted EPS for the quarter was $0.60 compared to $0.51 in the third quarter of 2024.

Moving to our balance sheet.

We ended the quarter with $1.2 billion in cash, cash equivalents and short-term investments and generated $57 million in free cash flow from operations.

Now let me share a few comments on fourth quarter trends.

As part as our quarterly results today, we also released our October monthly KPIs.

Consistent with our commitment to greater transparency and enhanced disclosure, we will now begin publishing KPIs on a monthly basis.

We've also made a detailed spreadsheet available on our website, which includes historical monthly data to help investors better track our performance over time.

Our goal is to provide the investment community with the information and tools needed to more clearly understand, model and evaluate our business.The month of October continues the strength that we saw in Q3.

Our capital markets and crypto businesses saw significant year-over-year growth in both total number of trades and invested demand per trade.

Assets under administration was $20.5 billion, up 72% year-over-year and funded accounts were [3.76%], up 17% year-over-year.

To summarize, we are very pleased with our strong Q3 performance and positive momentum, and we believe we are well positioned to capture new opportunities, drive sustainable growth, and further strengthen eToro's leadership in global investing.

With that, Daniel, let's move to Q&A.

Daniel Amir: Thank you, Meron. The first question here comes from a list of questions that have been pre-submitted by our retail investors. This question is for Yoni.

So eToro announced its new subscription offering.

Can you comment on why eToro decided to launch the program?

And what is the opportunity?

Yoni Assia: Sure.

We've built the eToro Club program to provide great premium benefits including unique features, better service as well as discounts to various parts in eToro, as well as the card, the cash back and now the crypto cash back. The club originally was based only on the tiers of how much assets customers have, and we wanted to provide the same great benefits to people who want to subscribe to the new eToro subscription. This also consolidates to subscription models that we've had from acquisitions that we've done in the past, and we're very excited to see how the pickup is already to the new subscription model.

Daniel Amir: Thank you, Yoni.

Operator?

QUESTIONS AND ANSWERS

Operator: 

(Operator Instructions)

Our first question will be coming from Dan Fannon of Jefferies.

Daniel Fannon: Wanted to talk about account growth. And so obviously it's been good or quite strong.I was hoping you could talk about if there are regions or new markets that are contributing more and maybe the breakdown of that account growth here, not just for the quarter, but also October, if there's any more granularity, that would be helpful.

Yoni Assia: Sure. Meron?

Meron Shani: Thank you, Dan.

So with regards to our current growth, like organically, as we discussed in the past, we have achieved a double-digit growth.

We've seen great results coming out of the cohort of clients that we are bringing, increasing the level of deposits on to the platform and increasing their level also of engagement with the platform.

So overall, very good results.

With regards to geographic, we are not breaking down geographically, but I could definitely tell you that we are seeing early, but good signs coming out of the new regions that we launched in the -- recently and in the last few years as well as well as a significant strength coming from our core markets.

Yoni Assia: And we have seen also a very good increase in the size of accounts coming into eToro.

So it's both new funded accounts growth and their first time deposit, which is a very good indicator to the future size of the accounts.

Operator: And our next question will be coming from Devin Ryan of Citizens.

Devin Ryan: A question on artificial intelligence implications here.

So great to see 30% of club members engage with eToro.

It'd be great to hear about anything you can share on kind of conversion uplift from doing research to actually trading and trade frequency.

I know we're early days here, but I'm just trying to think about that. And then more broadly, taking a step back, just some of the other types of offerings you're planning to launch with artificial intelligence and just how you feel like that can maybe differentiate the eToro offering in the market?

Yoni Assia: Sure.

So during our Pro Investor Summit about two weeks ago, we've actually announced our new tools, an AI studio for Pro investors to actually vibe code tools. Those tools includeboth new types of analysis that can actually look at their portfolio using AIs, look at other people's portfolio, using AI read our entire feed and create automated strategy.

I think the biggest driver eventually of AI, and I started, I think I did my first automated trading strategy when I was about 16, 17 is the transition from just click to trade through the (inaudible) to eventually running automated strategies, quantitative strategies of our users.

And what we've built is the ability for our Pro investors to actually build those apps, those strategies.

It includes dashboards and analysis of how they generate their trades and then actually provide an app store, so all of our customers can access that. And we've seen some really amazing apps developed already by our pro investors. This is a quick example.

We've had one of our pro investors actually creating the persona of Ward Buffet Benjamin Gram, and I think it was Kathy Wood basically ranking his stocks in his portfolio and suggesting how to rebalance the portfolio.

So I do believe AI One is an amazing tool for people who are not necessarily technical to actually bring in sort of their ideas on how to automate their strategies, how to make their strategies more professional on top of the eToro platform. And I think over time that drives both high returns, so making our customers more successful and actually higher velocity of trades in eToro.

Operator: And our next question will be coming from Craig Siegenthaler of Bank of America.

Craig Siegenthaler: Our question is on your strategic focus by geography.

So please correct me if I'm wrong, but I see a company that is very focused on growth in the U.S., in Asia and maybe widely geographically as new markets.

But eToro has leading share in Europe with some scale. And this market also looks a lot less competitive than the U.S. and Asia.

So my question is, why not focus on your first-mover advantage in Europe, and continue to deepen our investor base here, especially given that there are less regulatory restrictions on your coffee charter model in Europe than in the U.S.?

Yoni Assia: Sure.

So as we think of our growth strategy, first of all, we think of long-term growth, and we do want to build significant presence in the U.S. and in Asia. Nonetheless, on the day-to-day when we drive our marketing and most of our marketing is a data-driven performance marketing approach, we always look at basically the ratio between CAC to LTV.

So when we have a strong region, and obviously Europe -- some of our strongest regions in Europe, UAE, Australia as well we're actually doubling down on growth there as well.

So we are focused on growing the business through data-driven approach and make sure that where we see the highest return on investment, that's where we continuously grow and invest more.

So we are definitely focused on maintaining our leadership in Europe in the retail brokerage industry.

Operator: And our next question will be coming from James Yaro of Goldman Sachs.

James Yaro: Could you help us think about the account growth algorithm here.

What do you see as achievable for account growth going forward? Is it high single digits or double-digit account growth on a yearly basis? And then in the press release, you talked about net new accounts in the first three quarters being stronger than all of 2024, but maybe any ability to comment more narrowly on the 3Q 25 trends?

Yoni Assia: Sure.

So first of all, strategically, we are aiming towards double-digit growth of funded accounts.

We believe, again, PAUSE the total available market of new generations coming into the market, the market itself will grow in double digits, and we believe we can outpace the market growth in all of our -- both existing strong regions in Europe and beyond as well as obviously new regions such as the U.S.

The comment in the press release, I do believe, was regarding the U.S., which post IPO, we actually started basically revamping both our product strategy in the U.S., bringing all of eToro's global products into the U.S. including now more than 100 crypto assets staking that was launched in the U.S. as well as very excited about launching CopyTrader now in the U.S.

So we do expect double-digit growth in funded accounts moving forward strategically across all regions.

Operator: And our next question will be Brett Knoblauch of Cantor Fitzgerald.

Brett Knoblauch: Maybe to one on the crypto side, it was a really strong quarter in terms of crypto volumes have been -- you guys bought outpaced global spot volumes by two 1, if not more. Can you point to comment to that was maybe from what you did in the U.S. and had new tradable assets in the U.S. post outside of the U.S.

And then as a follow-up, I think ECC can be a bit below where we were expecting.

Can you just talk about some dynamics at play within the DCC doing shorter as well?

Yoni Assia: Sure.

So I'll cover a high level and then I'll let Meron talk a bit about the numbers in the quarter.

But first of all, when you look at our numbers in the last eight quarters, and you see that in the investor presentation, what we've seen over time is very clear. When one asset class is very strong, we see a shift towards the asset class from other asset classes.

So every time crypto has a very strong momentum, we actually see the noncrypto revenues or capital markets revenue actually going down a bit.

And then as crypto goes down, we see capital markets significantly shift higher as well.

And that is a dynamic that we did see in Q3.

In addition to that, within capital markets, there are two segments. There's currencies, commodities and indices basically future based and their stocks. And we did see more stock trading volumes increase in Q3, which led to the lower -- basically take rate this take rate on stock trading is lower than on commodities and they (inaudible) FX.

Regarding the crypto market, of course, we've seen a couple of all times high during Q3 that always supports our crypto revenues and trading activities.

In general, we do expect to continue and see a strong crypto market with a very, very positive U.S. administration towards the crypto industry and a lot of new product rollouts for eToro within the crypto industry with that a bit about the actual numbers in Q3.

Meron Shani: Sure.

So we can see definitely a sequential growth from Q2 in a very high pace, both on the number of trades as well as the invested amount.

We saw elevated activity coming in July and August as we released our KPIs also early September about the summer.

So very good traction coming out of the crypto activities of customers in the summer.We did see a slight reduction in that in September, but remains at the same level also in October.

We're very happy about what we see in the industry.

And in general, as it goes to become more and more mainstream, we'll see the number of trades and the number of customers tapping into crypto growing. And together with that, our revenues as well.

Operator: Our next question will be coming from Chris Allen of Citi.

Christopher Allen: Maybe you could talk about capital allocation priorities here. You announced the buyback this morning, make some sense with a lot of exploration now but you've also talked in the past about looking at inorganic growth opportunities, particularly in the U.S. to build it out.

So help us think about the different levers here from a capital allocation perspective.

Yoni Assia: Sure.

So first of all, we've added a significant amount of cash flow already this year.

So our balance sheet grew from $1 billion to $1.8 billion in total with $1.2 billion in cash and short-term investments.

So we have a very strong balance sheet to look both at buyback. And again, at these levels, we do believe the price is undervalued, which is why we're buying as well.

And of course, leaves a significant amount of dry powder also to look at acquisitions.

We have been talking to various potential targets.

We're always excited to meet great teams of great founders with products that we believe add value to our customers and that our products will add value to their customers, and we're actively exploring these opportunities across different regions. And of course, as we announce them, you'll know what are the actual targets.

Operator: And our next question will be coming from Bill Katz of TD Cowen.

William Katz: This question coming back to your commentary about the longer-term opportunity for the prediction and the event market. Just wondering if you could maybe lay out your pathway of like how you get there.

Is this an organic opportunity? Is it inorganic? And then within that, I was just wondering if you could share your thoughts on just philosophically how you think about the sports or gambling opportunity within that?

Yoni Assia: Sure.

So prediction markets, we're looking at basically both alternatives.

So first of all, we've launched futures in Europe and the future rails that we've launched in Europe are the same rails that eventually enable basically to trade prediction markets in the U.S. as well.

In addition, we are launching our crypto wallet, which will enable our customers to engage with on chain prediction markets such as Poly Chain and others.

So we're actually going towards both of those directions.

We do believe that prediction markets on financial events, on geopolitics, on economic events have to create significant value and people thinking about their trading strategies or hedging what they want to do and that initially will be our focus to help our customers, basically find those financial opportunities that are related to their portfolios.

Operator: Our next question will be coming from Alex Kramm of UBS.

Alex Kramm: Just on Copytrade in the U.S., I know it's super early, but maybe you can talk a little bit about (technical difficulty) how you're looking to scale it in terms of marketing, you're leaning in.

I think you have 300,000 roughly accounts in the U.S.

I think you said earlier, CopyTrader globally has a third or so of your clients have uptake there.

So just how should we be thinking about the opportunity and how quickly gain scale and if the third is kind of like a good target for the U.S. as well?

Yoni Assia: Sure.

So our view holistically is always, as we enter new markets to roll out all of the products that we offer globally in that market. And of course, we look at the U.S. as a huge opportunity with now crypto back, taking back and Copy Trader now launch alongside stocks and options trading as well. And the way we look at each product that we release is an attachment of a customer.

So a customer engaging with CopyTrader.

What we see over time is the increase in both size of the wallet or share of wallet as well as over time LTV increases significantly. So the way we view it is we continue basically to do what we do across the globe now here in the U.S., which is promoting the CopyTrader feature within the app, increase the attachment rate of Copy Trading, building gradually more and more great investors that in the U.S. people can copy.

And over time we'll see that increasing LTV and enabling us to basically spend more or increase our tax in the marketing budget to bring in more users to the eToro platform.

So it's all about bringing all of the products to our U.S. product strategy, then make sure that throughout the product, we create those attachment rates that increase the lifetime value and share of wallet of our customers here in the U.S., and that obviously leads to us scaling up our marketing strategy here in the U.S. in a profitable way.

Operator: And our next question will be coming from Jamie Friedman of Susquehanna International Group.

James Friedman: (Inaudible) ask first about funded accounts as well. (Inaudible).

Yoni Assia: Let's move to the next question, and then I'll come back if we can get a better connection with him.

Operator: Our next question will be coming from John Todaro of Needham.

John Todaro: Congrats here.

I just wanted to follow up, and I'll make sure I heard it correctly. Would you guys look for kind of economic partnerships on production markets with like a telco poly market, and then if I could get kind of a second question.

It sounds like there is some -- I don't know if you want to call it cannibalization intra- quarter between some of these trading products, as you mentioned, where crypto maybe goes up and some of the other markets go down a little bit. How do you eventually get around that where they kind of are all kind of up and up on -- in the same quarter?

Yoni Assia: Sure.

So first of all, we are talking to (inaudible) poly market.

Obviously the market leader is in prediction markets. And we are excited about exploring the path of our users to be able to trade prediction markets on financial events.

On the -- sorry, do you want to take it?

Meron Shani: Yes. The switching between the...Yoni Assia^ I'm not sure whether cannibalization is the right way to describe where customers inter-quarter trade different products.

So over time when you look at the growth of the net contribution of trading, both in crypto and capital markets, you see them growing over time and we think it's a great opportunity for people to actually reshuffle or rebalance their portfolios.

We've seen by this -- by the way, the same also between U.S. stocks and European stocks earlier this year as European stocks were very strong and a lot of our customers started buying basically defense stocks.

We've seen, obviously in the past, the same amping with (inaudible).

Later, we've seen it with MIM coins.

So again, as we bring in more products to our customers, we expect constant rebalancing of customers between different assets. And actually, that's the strength of eToro versus pure crypto companies, which don't have stocks or significant capital markets, and it's a significant strength of eToro, the fact that we cover today 22 different global exchanges, so our customers can trade all of those different products, alongside, of course, both on-exchange and off-exchange derivatives like futures and options.

So we feel that the fact that customers are trading in between the different assets that is a strength and something that we actually want to see with all of our customers.

Operator: And our next question comes from Brian Bedell of Deutsche Bank.

Brian Bedell: Maybe just to go back to copy trading, the mechanics of that in the U.S., in particular, I know there was some different regulatory regime in terms of being able to offer it with, I guess, the advisory component being a -- critical component of being able to be paid.

Can you just go through the mechanics of how this is operating in the U.S. versus outside the U.S.? And if it's the same now or when you expect to have that aligned with how it works outside the U.S.? And any commentary on initial take-up.

I know it's very early, of course.

Yoni Assia: Sure.

So first of all, it's very early to look at take rates and engagement, but we have seen already people being copied and people are copying top investors here on the U.S. platform, which is great to see.

Second on the mechanics. The mechanics are quite similar from how it operates. So you can actually -- when you copy an investor, it copies their entire portfolio into your portfolio in the amount you chose.

So you copy somebody with $1,000, it fractionalize his entire account across stocks and crypto and it basically opens those trades in your account. And every time they trade and trade in your account at the same time the same price and the same proportion.

So the mechanics of copying works the same here in the U.S. and outside the U.S., we are looking right now outside the U.S., we've actually just revamped our Pro investor program, which is around how do the Pro investors actually get paid here.

We're still early stages.

We'll be building that as well as we move towards next year.

The second copy product, which is our smart portfolios product, where our Global Chief Investment Officer actually curates specific portfolios, about 120 different smart portfolios across different strategies. That part will fall under an RIA license, which we are in the process and expect that to launch in H1 next year.

Operator: And our next question will be coming from Dan Dolev of MIZ.

Dan Dolev: Congrats, guys. Great quarter, great October.

I have a question on banking. A lot of fintech companies view thinking as the Holy Grail kind of, that's what everyone wants to do.

So Yoni, Meron, like how do you guys view sort of like banking services globally in the U.S., et cetera, in terms of the long-term opportunity for eToro?

Yoni Assia: Sure.

So I generally say everybody is looking at the concept of the super app, right? So we believe our core is actually helping our customers find data in alpha into their portfolio to generate over time returns than to compound wealth over time.

So our view comes more from our core expertise on trading in investing.

What we've seen over time is basically expanding the business also into savings, the various pension schemes around the world where we have already launched U.K., Australia and France recently including the now new launch of cash sizes in the U.K.

And lastly, of course, neobanking.

And what we wanted to see and what we're seeing in our strategy is basically when we provide our customers the virtual bank account on eToro, interest rate that they can get on their local currency and the ability to convert easily FX from euro to dollars or pounds to dollars to efficiently trade also U.S. capital markets and then attach into that also the Visa debit card, we remove the need of a customer to actually take their money back to the bank.

So I think us as a financial super app, we're looking for customers to increase their share of wallet over time on eToro.

We believe most of customer assets should be compounded towards capital markets in crypto markets and they should easily spend from that account without the need to actually take their money back into the bank. And we've seen that significantly improve basically attachment rates of customers, lifetime value of customers, which is why we're also providing great incentives at user Visa debit card with a 4% stock back.

So when you spend and we see a lot of our customers now posting on X when they're spending on a meal and then they get a 4% stock back into their portfolio. Again, why the stock back? Because we do believe that all of our customers would actually have most of their funds or the vast majority of their funds in capital markets and crypto markets, and that's the efficient way to compound wealth over time.

Dan Dolev: Great. Makes a lot of sense. Congrats again on a great quarter.

Operator: And our next question will come from Jamie Friedman of Susquehanna International Group.

James Friedman: Yoni, the press release indicates that the 16% growth in funded accounts was driven both by the user acquisition, but also the retention.

I was hoping at a high level, you could unpack that those dimensions. And then if I could just sneak in another one, Meron.

Can you remind us about the typical seasonality of the business as we build out our models?

Yoni Assia: Sure.

So we currently don't break down win backs churn and retention, but we've been engaging significantly in going back into basically our database of funded accounts across time and making sure that they are kept up to date on all of the new product releases, whether it's, again, the debit card in Europe, whether it's futures trading, so or whether it's the new crypto wallet and the ability to use our crypto to buy stocks.

So as we do that, we bring in also old customers into eToro sort of customers that churned in the past that are now joining back eToro for our new products. The second question, you can answer, Meron.

Meron Shani: Yes.

With regards to -- you touched about the breakdown, we do not provide it, but we can definitely see strong signs coming out of both the retention side as well as the acquisition as we aim -- as we discussed in the past, for a double-digit growth on the funded accounts on an annual basis.

Yoni Assia: Regarding seasonality, I would say, historically, we see strong Q1s and Q4s.

It's mostly though related, I believe, to the market.

So I would say we're enjoying usually seasons of the markets. And it seems, again, if I look at the last three years, Q1s and Q4s are usually good markets in both crypto and capital markets, but of course, we don't control the rain or the seasons.

Operator: Next question will be coming from Alex Kramm of UBS.

Alex Kramm: Just wanted a couple of follow-up questions. Hopefully, I didn't miss this earlier, but on the expense side, I think previously it said flattish for the remainder of the year.

Is that still right for the fourth quarter? And then just another housekeeping one on the interest income.

I know there's a lot of moving pieces, but I think the implied rates -- interest rate actually went up quarter-over-quarter if you just use the interest earning assets.

So again, I know there's a lot of things in there, but maybe just unpack what you're seeing on the interest income side?

Yoni Assia: Sure.

I'll start on the first one, which was related to the expenses.

So yes, definitely, we -- as we mentioned last quarter, we're aiming to keep our cost base the same quarter-on-quarter. And indeed, we came with a flat view.

So we are roughly looking at staying within those lines also in Q4.

With regards to the -- what was the second question?

Second question, Daniel?

Alex Kramm: The interest income...

Yoni Assia: So we have seen like definitely, there is a decrease in the interest rate coming from the two rate cuts.

But what we have witnessed in those periods is that actually customers balances in all different asset classes have actually grown.

Now -- so even though there is a direct reduction expected in interest income.

Over time we do see those balances increase. And therefore, we should be able to grow also on the revenue, that also taking into consideration that customers are inclining when the interest rates are lower, to bring their assets more into a more riskier portfolios that will generate higher return for them than the lower interest rate in the market.

Operator: And we have no further questions. And for closing remarks, we'll bring it back to Daniel.

Daniel Amir: Thank you for attending our earnings call today.

We're looking forward to seeing you at our upcoming investor conference during the quarter. These conferences are on our website that you can go see on the Investor

Relations section. And thank you.

And have a great day.

Yoni Assia: Thank you very much.

Meron Shani: Thank you.

Below is the Q2 2025 earnings call transcript for eToro.

This transcript is provided for informational purposes only and may contain errors, omissions, or inaccuracies. While we strive to ensure the content is as accurate as possible, it has not been verified by the company or its representatives. This transcript should not be relied upon as a verbatim record of the earnings call or as the basis for any investment decisions. Please refer to the company's official filings and recordings for authoritative information.

Presentation

Daniel Amir:
Thank you, everyone, for joining eToro's Second Quarter 2025 Earnings Call.

My name is Daniel Amir, Head of Global Investor Relations. Here with me today are Yoni Assia, Founder and Chief Executive Officer; and Meron Shani, Chief Financial Officer.

This morning at 7:00 a.m. Eastern Time, eToro released its second quarter 2025 results.

In the Investors Section of our website, you will find the earnings press release and presentation to accompany today's discussion.

This webcast is being recorded and will be available for replay in the Investors section of our website.

Yoni and Meron will offer opening remarks and then open the webcast to Q&A.

During the Q&A portion of the webcast, we will answer live questions from publishing research analysts. And we'll address the selection of the most up-voted questions previously submitted by eToro's shareholders.

As a reminder, today's webcast will contain forward-looking statements, which relate to our financial outlook and market positioning among other items.

Such forward-looking statements are not guarantees of future performance.

Actual results may differ materially from these expressed or implied in the forward-looking statements due to a variety of factors.

These factors are discussed in more detail in the press release that we issued today and related earnings presentation and our filings with the SEC including our prospectus dated May 13, 2025, as such factors may be updated from time to time in our filings with the SEC, which are or will be accessible on the SEC's website at www.sec.gov.

We do not undertake any obligation to update forward-looking statements.

Throughout this webcast, we will also present and discuss non-GAAP financial measures.

Reconciliation of our non-GAAP financial measures to their most directly comparable GAAP financial measures appear in today's earnings press release and presentation, which are available in the Investors section of our website.

And now I will turn the call over to Yoni, Founder and CEO.

Yoni Assia:
Thank you, Daniel. And thank you to everyone joining us today for your interest in eToro.

And welcome to our second quarter 2025 earnings call our first earnings call as a public company. After Meron and I conclude our prepared remarks, we'll open it up to analyst questions.

Before sharing some highlights from our second quarter results, I would like to briefly introduce eToro for those of you who are new to our story.

We founded eToro in 2007 with a vision of opening the global markets for everyone to trade and invest in a simple and transparent way.

At that time investing was largely the domain of traditional financial institutions catering to seasoned investors with significant capital. Retail investors face daunting barriers, complex platforms, high fees and a lack of educational resources.

Our goal was to change that by using technology to level the playing field and create a platform that could empower millions to participate in the financial markets.

Today eToro provides a gateway to the global financial markets, offering a social and investment platform that empowers our users from around the world to trade, invest, save and spend in a way that suits their unique needs.

We've built a social investment network designed to provide users with educational resources and tools they need to grow their knowledge and wealth.

We provide what retail investors around the world care about the most, simple access to the assets they want to invest in an intuitive and user-friendly mobile app and a trusted and transparent source for financial education and tools including the ability to draw the knowledge and insights of the top investors from all around the world.

On eToro's platform, users can invest in stocks for more than 20 different stock exchanges and trade over 130 different crypto assets alongside indices, commodities and currencies.

We serve customers in 75 different countries and 20 different languages, users can trade directly, invest in a portfolio or use our patented CopyTrader technology to replicate the investment strategies of the top investors on our platform. eToro Money, our money management offering enables our customers to have a virtual bank account and a debit card and manage multiple balances as they trade both global and local stocks.

We also provide mid valuable investment tools and services including sophisticated charting and AI-powered tools.

We've built a track record for identifying and adopting key investment and technology trends.

We've been a pioneer in the crypto space since 2011 due to my own passion around blockchain technology. Crypto's underlying principles of transparency, security and inclusivity resonate deeply with eToro's mission.

As one of the first global investment platforms to embrace crypto, we've helped millions of investors access this new market in a safe way.

As technology continues to evolve so does our ability to create more inclusive financial systems, artificial intelligence holds immense potential to revolutionize investing.

We're already using AI to provide users with personalized insights, identify trends and optimize their strategies, but this is just the beginning.

We believe that the next wave of financial innovation will be driven by collaboration between people and technology. Editorial, we're committed to staying at the forefront of this transformation using technology to enhance the investing experience while keeping the human element of its core.

Our goal is to create a platform that not only meets the needs of today's investors but also anticipates the opportunities and challenges of tomorrow.

Now let's move to the quarter results.

We are very pleased with our strong second quarter results, which reflect the continued success of our goal to be the leading trading investment platform.

Our net contribution for the quarter increased 26% year-over-year to $210 million, and our adjusted EBITDA increased 31% year-over-year to $72 million, while improving operating margins.

In terms of geographic diversification, we recently expanded our global footprint, activating our license in Singapore. This milestone positions eToro to serve Asia's growing investor base from a key financial center as we continue to grow our global presence.

Over the past quarter, we also made real progress in advancing our product development across our four strategic pillars: trading, investing, wealth management and neo banking, delivering innovative features that strengthen user engagement and expand our addressable market.

In trading, we continue to enhance the trading experience by expanding and flexibility for our users.

We announced the expansion of 24/5 trading and on exchange futures, and we intend to expand this offering to provide our users around the world with the flexibility to trade on their own time. The ongoing efforts to remove boundaries in training and financial markets underpins our interest and focus in tokenization. Back in 2012, we wrote the colored coins white paper, a method for representing and managing real-world assets on top of the Bitcoin blockchain.

We first offered tokenized assets back in 2019, following our acquisition of Smart contract start-up Firmo. Building on this heritage, we're launching tokenized stocks.

We believe that blockchain technology will facilitate the greatest ever transfer of wealth as traditional assets move on to the blockchain.

As a platform that bridges between traditional trading and a blockchain-powered future, we believe we're very well positioned to participate in the transformation of financial markets and grow our business.

In the U.S. recent regulatory development around crypto represents an important milestone and a great tailwind for our U.S. expansion.

As a result, in recent months, we've significantly expanded our crypto offering for our U.S. users and now offer the ability to trade over 100 different crypto assets.

We'll continue to add more crypto assets and more features for our U.S. consumers and will be soon launching staking as well. Turning to investing.

We continue to introduce innovative solutions that make investing simpler, smarter and more personalized. This quarter, we rolled out a suite of our Alpha portfolios as part of our smart portfolio offering. These AI-powered long/short strategies are built using advanced analysis of multiple sources of data including eToro's proprietary retail trading data.

This gives our users assets to exclusive quantitative trading strategies typically used by hedge funds and institutional investors only.

In addition, we also announced a partnership with Franklin Templeton to launch new portfolios focused on long-term target date investing. These efforts broaden our investment offering and personalized user experience to help our customers achieve their financial goals at each stage of their financial journey.

As part of our efforts to provide a global platform with a localized experience, we've expanded our wealth management offering to provide access to long-term tax advances, savings and investment solutions.

In the second quarter, we expanded our savings from Australia and the U.K. to also France, launching retirement and life insurance products to give our users more ways to invest in the future.

Additionally, we rolled out recurring investments and saving plans in crypto and stocks in the U.K., Europe and the UAE, enabling our users to automate their investments in making disciplined investing easier and more accessible. Moving to our neo-banking solutions.

We recently launched the eToro Money Card across Europe. This card turns everyday spending into investing by offering a 4% stock back reward on every purchase alongside premium benefits.

Before I conclude, I want to also address the topic of harnessing AI technology. eToro has been leveraging artificial intelligence for many years, but we believe that we are now at an inflection point.

We are harnessing AI to redefine social investing.

We're rolling out a suite of powerful AI-driven tools that empower popular investors with sophisticated professional capabilities.

With these tools, we're not democratizing only access to the markets and data.

We're now democratizing quantitative trading and investing. Community is at the heart of everything we do, and these AI-powered tools will enable our popular investors to build apps and trading strategies that combine social trading with advanced execution capabilities. equipping them to innovate like the top hedge funds around the world and scale their impact with the eToro ecosystem.

In line with this vision, we recently launched Tori, our powerful AI system, transforming how users interact with eToro, answering questions, surfacing personalized insights, guiding them into the platform and helping them make better decisions, all through natural conversation. Tori helps users make smarter decisions and navigate our platform based on the advanced models out there including recently GPT 5.

We truly believe that AI tools are a game changer for investors.

I suggest you all try that. Together, these product innovations and enhancements reflect our focus on empowering our users from around the world to trade, invest, save and send in a way that suits their needs.

We feel great about our product development velocity that's been accelerated by AI and the business result it is driving.

Looking ahead, I'm excited about the opportunities in front of us, and we are well positioned to execute on our growth initiatives to drive sustainable long-term growth.

With that, I will pass the call to Meron to discuss our financial results for the quarter.

Meron Shani:
Thank you, Yoni.

Before we begin with an overview of this quarter's financial results, I would like to share some insights on eToro's financial philosophy.

Our philosophy is focused on sustainable, profitable revenue growth, which is supported by three pillars.

First, diversification of revenue streams to complement transaction-based revenue, with a focus on growth in asset-based revenues, expansion of our eToro Money offering and the introduction of new revenue streams.

Second, strategic investment to support growth in funded accounts and increase our share of the user's wallet as we focus on product innovation, organic and inorganic growth and a dynamic approach to sales and marketing operations.

And third, disciplined cost management.

We focus on streamlining operations, optimizing costs and improving efficiency across the organization, using technology including leveraging AI by automating routine tasks, optimizing resource allocation and enabling faster data-driven decision-making.

We believe this financial philosophy will deliver best-in-class financial results and an increase in shareholder value.

Now turning to the quarter results.

As Yoni mentioned, we are very pleased with our second quarter results, our first full quarter as a public company.

Second quarter net contribution grew 26% year-over-year to $210 million and adjusted EBITDA grew 31% year-over-year to $72 million.

In line with our focus on diversified profitable revenue growth, our adjusted EBITDA margin was 34%, expanding 140 basis points from a year ago.

As our first quarter momentum continued into the second quarter, AUA for the quarter increased 54% year-over-year to $7.5 billion, while our funded account grew 14% year-over-year to $3.63 million. The growth was driven by customer acquisition efforts, supported by strong customer retention.

Let's take a closer look at our second quarter financials by segment comp to a year ago.

Our net rent contribution from equities, commodities and currencies grew 37% year-over-year to $114 million as invested amount per trade grew 54% year-over-year due to elevated levels of trading activity in the capital markets during the quarter, especially in the month of April.

Our net contribution from crypto grew 34% year-over-year to $27 million and was largely driven by higher invested amount per trade, partially offset by lower crypto activity, in line with the overall market.

Interest-earning assets contributed $46 million or up 1% year-over-year, largely driven by higher interest-earning assets that was partially offset by lower interest rates.

Our eToro Money contribution grew 17% year-over-year to $18 million, largely driven by an increase in total money transfers.

In the second quarter, adjusted OpEx was $138 million, our adjusted selling and marketing expense was $51 million or 24% of net contribution.

Our business model provides us flexibility in our selling and marketing expenses, where approximately 70% of our expense is dynamic.

Our historical performance has proven that our marketing expense has consistently returned positive ROI within the first year of the investment and sustained commission growth over time. Adjusted R&D and G&A expenses were $36 million and $51 million, respectively.

Our adjusted EPS for the quarter was $0.56 compared to $0.51 in the second quarter of 2024. Moving to our balance sheet.

We ended the quarter with $1.2 billion in cash, cash equivalents and short-term investment.

And for the quarter, we generated $61 million in free cash flow from operations.

During the quarter, we also announced a new credit facility of $250 million to increase liquidity optionality.

Now let me make some comments on the third quarter trends. Through the first month of the quarter, crypto activity has been higher than second quarter levels, while capital markets (ECC) activity has returned to a more normalized level following the elevated levels of the second quarter.

We are very pleased with the first quarterly financial results as a public company. This sets the foundation for future growth as we focus on increasing shareholder value.

With that, Daniel, let's move to Q&A.

QUESTIONS AND ANSWERS

Daniel Amir:
Thank you, Meron.

So okay. The first question in our Q&A session today comes from a list of questions that have been pre-submitted by our retail investor community.

Yoni, this question is for you. When does eToro plan to provide calls and puts options to European investors.

Yoni Assia:
Thank you, Daniel.

Our product rollout strategy is to continuously launch the products where we see the highest potential product market fit and revenue opportunities for eToro.

With options, for example, we sit in the U.S., we expanded into the U.K., and we plan to expand over the next 12 months also to Europe, UAE, Singapore and Australia. The same with futures that we recently launched in Europe and continue to expand it across the globe and with all of our products across the globe eventually offer to our customers in all of our regulated regions.

Daniel Amir:
Thank you, Yoni.

So operator, we'll open it up to questions from our institutional analysts.

Can you provide instructions?

Operator:
(Operator Instructions)
Our first question will come from the line of Dan Fannon with Jefferies.

Daniel Fannon:
I was hoping you guys could expand a bit upon just the cadence throughout the quarter since we don't get monthly.

I was hoping just to get a little more color.

Obviously April was quite active from a macro perspective.

So if you could talk to the change in environment as the quarter progressed? And then maybe expand a bit more upon July in terms of how that country did versus what we saw in the first quarter.

Yoni Assia:
I'll let Meron answer that.

Meron Shani:
So obviously as we saw with the market following the tariff announcement in April, we saw elevated trading activity of our customers on the platform.

We saw invested amount growing per trade of 54%.

We saw invested demand growing in general. And the Q2 revenue per trade also grew 54%.

So pretty much aligned with the invested amount.

What we saw is that those numbers normalized throughout July, as we moved from the spike in April and normalized towards the end of the second quarter as well. When you talk about crypto, we saw lower levels in the second quarter normalized also with the market and also taken into consideration the fact that our customers were actively trading on the capital markets, they diverted from trading on crypto. When we come to July, what we see on the crypto side is that customers following the outgoing altcoin season as well moved to a higher pace of trading than quarter two.

Operator:
Our next question is going to come from the line of Devin Ryan with Citizens.

Devin Ryan:
Question to start here just on kind of digging into the opportunity on the back of the launch of the tokenized stocks recently. And as we think about all the other types of assets that are going to come on chain over time it's great to just double-click into how you would quantify the revenue opportunity for eToro. How fast do you think under asset classes are going to move on, assets that are currently not available to eToro's customers? And then the overall kind of bigger picture opportunity from this long-term existing like right now stocks like a lot of other assets that are currently not available to eToro customers will if you could be available (inaudible).

Yoni Assia:
So we've been very early on for the concept of tokenizing world assets starting with the white paper of colored coins that we wrote together with (inaudible) between 2012. I believe that the digitization and tokenization of assets is a very significant process that will take time but we do expect $100 trillion of assets over the next years to move on chain, now quoting the SEC Chairman talking about moving U.S. capital markets on chain.

We already support stocks from 22 different capital markets as tokenization happens across stock markets, bond markets and potentially real estate. People are talking about private shares. And I do believe we'll see that significantly scaling up over the next sort of midterm of the next three years. Everybody is waiting, and we're waiting for regulatory clarity.

So it was very hard to think of tokenizing real-world assets and securities without clarity on regulation with micro regulation now in Europe, that's a possibility, and we're talking to many participants in the market about various opportunities of tokenizing assets, as you probably seen also the largest asset managers in the world are all involved in how they can tokenize their potential funds and products and then obviously listed them on platforms like ours to introduce them to retail investors.

So we do see a very large opportunity, both in scaling up the investment universe in eToro over time as well as opening all those assets to 24/7 trading.

So we just recently launched both trading and futures, which trade 24/5. And we do believe that by tokenizing assets and expanding liquidity over the weekend will eventually enable customers to transfer assets 24/7, like crypto assets are transferable 24/7 and to actually trade on assets 24/7 continuously.

Operator:

Our next question will be from the line of Bill Katz with TD Cowen.

William Katz:
Just maybe a couple of questions embedded.

So the first question would be, could you give us an update on any progress you're making for a CopyTrader opportunity into the U.S. And then secondly, just given all of the growth vectors in front of you included that we laid out on the global webinar, how should we be thinking about marketing spend either in an absolute sense or as a percentage of contribution.

Yoni Assia:
So I'll take the first question, and or maybe you can start actually with the second question, and I'll jump to the first question.

Meron Shani:
So when we think about marketing spend, we think about it as a percentage of net contribution, and we are normally around the 20% mark of net contribution investing into the growth coming out of new acquired customers.

Yoni Assia:
Can you repeat the first question just for clarity.

William Katz:
Yes. If you could just give us an update on any progress you're making on introducing in the U.S.

Yoni Assia:
So as we announced, we've been able to significantly scale the number of assets on the platform from three crypto assets for U.S. customers to now 100 crypto assets in the U.S.

We are also planning to introduce staking soon with clarity from U.S. regulators across staking, and we plan to launch a CopyTrader for U.S. customers on crypto and stocks later this year.

Operator:
Our next question will come from the line of Chris Allen with Citi.

Christopher Allen:
I wanted to ask about account growth dynamics from about 8% year-to-date. Maybe you can give us a color on regional account growth, where you're seeing the fastest growth and best opportunities here.

Meron Shani:
So we manage that growth on a global scale.

We see some growing regions for us, like we launched a couple of years back in OEE, and we see growth coming from there.

But altogether, we see the growth coming from a very, call it, a balance across the different regions that we are managing. There's no significant player in the regions that is significantly higher or lower than any of the others.

Operator:
Our next question will come from the line of Craig Siegenthaler with Bank of America.

Craig Siegenthaler:
So my question is a simple one on growth. Funded account growth was 1.4% quarter-over-quarter.

I'm wondering, is this a good go-forward organic run rate Were there any unusual factors in the quarter, especially given that sharp correction we had in April. And I know M&A has aided your account growth rate historically, but I wanted your perspective on the organic trend.

Meron Shani:
So we have proven in the last few years that we are able to grow our funded accounts on a double digits on a yearly basis.

We did see a slower quarter in terms of the growth.

However I could definitely say that we see a better quality of customers. They bring more share of their wallet onto the platform.

And we could also note about the average account size that grew 34% year-on-year, thanks to customers loading more of their share of the wallet as well as market gains. And also to say that in terms of marketing, as I mentioned before, we manage it as a percentage of net contribution, but we also manage it to bring the best quality.

So we have the ability to be very flexible in how we invest. 70% of the investment is dynamic, so we could upscale and low scale as we see right coming out of the results, we are very comfortable about our ability to continue and deliver growth and continue to deliver great ROI returning the investment within the first year and growing over time.

When we come to growth of the funded accounts also in the future, we are seeing good numbers coming out of rolling out more products that serve as a good product market fit, as Yoni mentioned, that will drive not only better engagement of existing customers, but also drive more growth of funded accounts.

Operator:
Our next question will come from the line of James Yaro with Goldman Sachs.

James Yaro:
I just wanted to touch quickly on the net contribution per trade across both ECC as well as crypto.

I think on the ECC side, it was up about 25% sequentially and crypto was up about 30%.

Could you maybe just talk a little bit about the drivers of that and perhaps the outlook for take rates going forward?

Meron Shani:
So the main correlation that we see and we quoted also on the percentage is the growth of the invested amount per trade that grew 54% year-over-year and 20-plus percent also sequentially.

So this is more or less aligned on the ECC side of how we see that the invested amount has also elevated the revenue per trade.

We did introduce ticket fees outside of the U.S. and U.K., and that increased the invested amount and reduced smaller transactions.

So there's some balancing of a slight decrease in the number of trades, but comes with a higher invested amount per trade. When we look into crypto, this is touched immediately to the invested amount per trade as we charge on a percentage basis.

We saw slightly better ability to analyze our volume also in the second quarter, which resulted with slight uptick in terms of the revenue per trade, but altogether, the growth of the invested amount is what drives the majority of that trend.

Operator:
Our next question is going to be from the line of Brian Vieten with Needham.

Brian Vieten:
With the new licensing in hand, can you talk about the glide path for new funded accounts growth in the APAC region more near term and then aspirationally. And then just on that, can you talk about the geographic composition of the 40 million registered users.

Is that in line with the funded accounts proportionally?

Yoni Assia:
So historically, there is a proportion between the total registered users or funded accounts.

It's worth noting that the total registered users is in a cumulative number since 2007 and includes basically everybody who registered since then, not only the monthly active users of the virtual accounts and in our more mature growth markets, Europe, U.K., Australia, we are seeing obviously higher conversion rates than basically cross-border markets where we might see a lot of registrations that don't necessarily convert into funded accounts.

When we think of our growth opportunities, we definitely think that both Asia now through Singapore and the U.S. represent significant opportunities to grow as a percentage of the business over time. Right now U.K. and Europe represent roughly 65% with Australia being a part of of course, our strategy in Asia.

But as we doubled down on our high potential growth markets, Asia, Latin America and of course, the U.S., we do believe represents a higher potential growth over the next couple of years.

Operator:
Our next question comes from the line of Joseph Vafi with Canaccord.

Joseph Vafi:
Congrats to you on your first public earnings quarter. Just thought maybe we drill down a little bit more on the U.S. strategy now with more cryptos available, and it sounds like the platform is a little more robust. And with CopyTrader potentially coming, just how should we think about marketing spend in the U.S. moving forward versus where it's been and maybe an update perhaps M&A still makes sense in the U.S. to kind of grow the platform materially here over the short term?

Yoni Assia:
So as we've said during our road show, during the sort of pre-IPO where around last administration, we took a step back from our growth strategy in the U.S. since the IPO, and obviously very excited to see the new administration and share talking about innovation in public markets in tokenization and getting capital markets on chain, which we believe we're positioned in a very good place to take part of that opportunity in the U.S.

So we'll continue to expand all of our products right now.

I'd say about only 50% of our products that are operating globally or offered in the U.S.

We do plan to aggressively, basically pursue and drive all of our product expansion into the U.S. in the next 12 to 18 months including CopyTrader and looking at all of the additional global products that we have. And as we expand and add new products, we expect to see higher lifetime value, obviously product market fit, which means more organic growth. And on top of that, we'll scale the marketing budget as well.

So we are looking to invest more in the U.S. market, and we're looking to do this in a disciplined approach where we're looking at the investments part of the revenue potential and existing revenues in the U.S.

We are looking and are opportunistic about M&A opportunities in the U.S.

We believe there's a huge opportunity right now of many great fintech companies that have a great product and team in the U.S. that are able to be independent. And we are also thinking of doing this, obviously in a disciplined approach looking at basically how this adds to the total KPIs and the growth opportunities in the U.S.

Operator:
Our next question comes from the line of Brett Knoblauch with Cantor Fitzgerald.

Brett Knoblauch:
Maybe just more broadly on kind of crypto trends during the quarter. Volume kind of declined, I would say, much more than global spot volumes.

So anything to read from that? And then as a follow-up on crypto in the U.S. meaningfully expanded the number of assets available to trade for U.S. users.

Can you talk about the initial kind of uptake you're seeing as you broaden the U.S. breadth there?

Yoni Assia:
So as Meron mentioned before, as a platform, which is actually quite uniquely positioned where if you look at the assets of our customers, roughly 50%, 50% stocks and crypto since we've been early into both. When we see exciting markets and capital markets as we've seen in April, we do see a transition of our active traders who are actively trading both crypto and stock .

We've actually seen our customers excited about European markets and the European defense stocks or that was the version of meme stocks in April, we saw a lot of our retail investors actually buying the deep, actually participate in capital markets, buying stocks that sometimes are related to crypto markets like Nvidia and Tesla and others.

So we did see that cycle of increase, basically capital markets versus crypto, obviously coming to July, where we saw all-time high, not only in Bitcoin, but also a significant scale up in a lot of altcoin activity generally.

Since the IPO, just under roughly three months, we've managed to add a significant amount of crypto assets actually reaching over 100 in the U.S. this week, and we have seen significant growth to our U.S. business in the U.S. after those additions, obviously also driven by very good crypto markets environment in July.

Operator:
Our next question will come from the line of Matthew O'Neill with FT Partners.

Matthew O'Neill:
Share the congrats to the public markets that others have mentioned.

I was just curious, how you guys are seeing your users utilizing stable claims.

I feel like there's probably some interesting insights given the geographic diversity of your user base and would just love to kind of hear some anecdotes around that. And then separately, I know it was asked, but maybe you could just dig a bit deeper into the increase in the take rate on crypto. Was that driven by the kind of outsized activity we saw in April, mix shift, pricing dynamics would just be helpful to kind of understand the drivers there.

Yoni Assia:
So first of all, regarding the crypto take rate, then we charge 100 basis points.

We actually recently launched also crypto fees that are separated from the spread, improving significantly at both transparency of fees and understanding of users. Anything to add to that, Meron?

Meron Shani:
We had a better ability to internalize some of the volumes.

So that increases the take rate a bit in the second quarter, saying that this is still within the range that we offered in the history as well about our ability to enjoy this take rate.

Yoni Assia:
Regarding stable coins, we just recently launched we're receiving the (inaudible) license in Europe, basically the ability to deposit crypto into eToro and then use crypto assets to buy stocks and invest in capital markets we're actually quite conservative in the past as there wasn't from our point of view, sufficient regulatory clarity.

So we did not enable customers to actually deposit stable coins or crypto into the eToro platform.

So with the new launch of enabling our customers across Europe, U.K. and expanding that to all of our global operations to now deposit crypto assets including stable coins, we do believe this opens up a new user audience, which are the crypto-native user audience.

So our platform originally is more capital markets rather than crypto.

So we started before crypto and most of the activity on eToro historically was people funding their accounts, of course, with fiat money with either local currency in dollars and then trading crypto assets or stocks with it.

Now as we double down on crypto as well we're now enabling that crypto to fiat. And so we are talking to some of the stable coin issuers to understand the various opportunities of partnerships there as well.

Operator:
Our next question will come from the line of Dan Dolev with Mizuho.

Dan Dolev:
Congratulations on a great first quarter with a public call. Really quick question on Tori.

I've noticed that you're using that. Just wanted to ask a broader question about sort of AI use in investing for your funded accounts, and how they perceive it, how they're using it?

And what are you seeing in the future in terms of the use of Tori and AI for investing?

Yoni Assia:
So first of all, I have to say I'm super excited about seeing the developments in AI, both from accelerating our ability to deliver products to making our organization more efficient.

I myself now have an actual coding platform and write things with AI on top of our APIs.

We just recently launched Tori, the AI analyst. Just 30 minutes ago, I checked Tori's view on eToro's quarterly financials.

So you can immediately see what's happening in the markets, analyze them.

It's, by the way, a multi-model so we can actually very easily switch in the back end, the different models.

We already started testing with users GPT5 as well as Gemini 2.5 and Grok, so we are very excited about the opportunities in investing. And we've launched our API that's powered by AI and the marketplace to enable our popular investors to actually build apps on top of the eToro platform.

I think this brings us to a future where retail investors can actually become significantly more sophisticated with their tools. They can basically build quantitative strategies related also, of course, to the rise of digital assets and crypto assets and 24/7 trading, even myself.

I've been passionate about capital markets, trading and investing my entire life, but I never had a 24/7 AI agent trading in my account.

So I have no doubt that these AI tools will help our customers both increase their velocity and engagement on the platform and become significantly more successful investors and more sophisticated investors with these tools at their hands.

So this is, I think, still early days, but with a huge opportunity ahead of empowering retail investors with tools that just one or two years ago were really reserved just to the top quantitative hedge funds in the world.

Operator:
Our next question is from the line of Craig Siegenthaler with Bank of America.

Craig Siegenthaler:
So my question is of the $114 million of net trading income from equities, commodities and currencies, I'm curious what was the percentage mix from CFD? And how has that mix been trending over the last year?

Meron Shani:
So as we mentioned before, we don't break it down really by the nature of the transaction, the type of the transaction, but we adopt a broader way of assessing which asset class that is coming from.

We offer both CFDs and futures on equities and commodities.

So we don't really segregate the results by then.

I think we mentioned in the past as well that we are aiming to get to a 20% mark within the 12, 18 months ahead.

So that's still where we are aiming to get.

Operator:
Our last question is going to come from the line of Charles Bennett with Rothschild and CEO Redburn.

Charles Bendit:
Just a follow-up, if I can, on the CopyTrader product in the U.S.

Is there any update regarding licensing of U.S.-based popular investors? Are there any regulatory teething issues there? Are you expecting this to roll out pretty seamlessly?

Yoni Assia:
We do expect, as I mentioned before, to roll it out this year as we also spoke during the road show that we believe that the CopyTrader technology could work and be translated into the U.S. markets relatively easily, and we've basically made more of a risk decision and not to launch it prior to the IPO regarding smart portfolios, we are working towards an RIA license to be able to launch our RIA -- our smart portfolios, which are basically our own strategies that we developed 120 smart portfolios from our customers.

Outside the U.S., we are seeing record high assets under CopyTrader, so very excited about launching it later this year in the U.S.

So that was the last question, Operator.

Operator:
Yes. Sir.

I'm showing no further questions at this time.

Yoni Assia:
Thank you very much.

So quickly closing remarks, as we close our first quarter earnings call as a public company, it's clear we are entering a new era for both toro and in the investing world at large. The future of investing will be defined by AI-powered intelligence that can process data at unprecedented speed, uncover insights, (inaudible) sight and personalize the investing journey for every user.

We're harnessing these technologies to empower our community, enabling smarter smarter decisions, more adaptive quantitative strategies and building their own apps now on top of eToro. Alongside this transformation, the global financial landscape is embracing digital assets as a core pillar of diversified portfolios from bitcoin and crypto assets to tokenized real-world assets.

The boundaries between traditional and digital markets are dissolving. eToro has sat at this intersection since 2013, giving investors seamless access to both worlds with trust, simplicity and scale.

We believe the combination of AI-driven investing and the continued rise of digital assets will reshape wealth creation for the next generation and we are committed to leading this change not just as a platform but as a global community, where anyone can trade and invest in a simple and transparent way. This is just the beginning. The age of intelligent investing is here and eToro is ready to lead it. Thank you, Grok, for just writing this. And thank you all of you for joining us.

Operator:
This concludes today's conference call. Thank you for participating. And you may now disconnect.

Everyone have a great day.

Hello friends, we’re glad to have you with us! Today we’re going to look at:

• the chances of another government shutdown,

• some potential presidential hopefuls,

• how Gaeten Dugas finds edge, and

• some markets to watch in the weeks ahead.

To kick things off…

It seems we’re headed toward another potential shutdown.

A bunch of House and Senate members left DC Thursday afternoon, jetting off to the Munich Security Conference, so, it’s going to be particularly difficult -- if not impossible -- for there to be a last-minute vote to save the day.

That said, lawmakers in the House and Senate are on notice to return if a deal seems to be on the horizon, but that probably won’t happen before midnight.

Plus, there’s a chance that the shutdown might drag on. With the exception of Senator John Fetterman, the Ds are staying united. And on Wednesday (the 11th), The White House sent the text of their offer for DHS funding, which the Ds rejected. And with folks out of town, it’ll likely be a few days before any progress is made.

So, why then, you might be wondering, are prices on Polymarket and Kalshi for “Government shutdown tomorrow?” and “Government shutdown on Saturday?” so divergent?

Well, as we’ve discussed before, it comes down to the rules and a statement by Scott Kupor, the director of the Office of Personnel Management:

And because of the difference in rules of resolution for Polymarket and Kalshi, each market responded differently.

On Polymarket, the rules state:

With the added additional context:

Kalshi’s rules state that:

In other words, Kalshi doesn’t simply come down to the OPM website, but Polymarket does. This is notable, because there is a real chance that a part of the government will shutdown this evening, and this occurs whether or not the OPM website is updated. If one of the intentions of prediction markets is to accurately reflect what will occur, then the rules of resolution need to be correctly written. Because, right now, on Kalshi people are taking positions on whether or not there will be a shutdown, and on Polymarket people are taking positions on whether a website will be updated. Those are not the same thing.

On the topic of the Munich Security Conference

AOC spoke today at the Munich Security Conference, offering a “working class” perspective on U.S. foreign policy.

The news that AOC would speak at the conference was first widely reported around February 6, 2026, when the New York Times published the profile "Alexandria Ocasio-Cortez Steps Onto a Wider Stage."

Although she still trails Newsom on Polymarket -- who’s at 27% -- there was a small bump in the odds of AOC being the Democratic Presidential nominee in 2028 after the article broke. And professional and amateur Twitter pundits are hinting at this being a test for her on a more global stage.

Beyond AOC, Newsom, Whitmer, and Gallego -- other potential nominees -- are all also speaking at the conference.

And, even though it’s still early, they all maintain a comfortable lead over Mr. Beast, The Rock, and Hunter Biden:

I wouldn’t expect too much movement in these markets until the midterms are done, but keep your eyes out for signals in the coming months to help build your model. And if you’re not doing that, we’ll be doing some of that for you.

How Gaetan Dugas Finds Edge in Spotify and Billboard Markets

Below is an excerpt from our recent conversation with top Kalshi Trader Gaetan Dugas…

You talk a lot about how to build models for markets, and about really data-backed theses for these things and how to find edge with those models. Can you go into a little bit more detail about what that actually means from a practical perspective?

Sure.

Like, if you have an edge, how do you build to get that?

Sure. So, I'll walk through Spotify as an example since it's pretty easy compared to a lot of the other ones and the data is so easily accessible.

So with Spotify, they've got their whole daily history of streams of the Top 200 for global and in every country, going back to like 2017. So, you've got access to this entire database of streams. And you can pull the data and start analyzing it, and you'll start to recognize patterns.

So, there are patterns that happen when a song is released. Friday is typically the biggest day. And then Saturday, it will drop by a certain percentage. Sunday it will drop another percentage. On Monday it usually bounces back. On Tuesday it bounces back a little more. And then from there, it does a slow decay. So you can spot that type of data. And you can look for comparable songs in the history to new ones that are coming out and kind of map out what and how you think it will do.

So, when it comes to modeling, I've got different models for different types of songs. A song like "Please, Please, Please" by Sabrina Carpenter or "we can't be friends" by Ariana Grande. Those are songs that did very well; they didn't drop very much.

And then there are songs that kind of flop, like "Aperture" by Harry Styles was a recent one where it dropped a ton.

And then there's kind of the average one.

So, basically what I do is I try and come up with a range of how songs will perform, and then once I track it a couple days, then I can get a pretty good idea of which of the categories it falls into. And based on that data, I can start projecting how it's gonna do the rest of the week.

Does that make sense as an example?

Yes, I can kind of see the spreadsheet for that too, right? You sort of slot it in. When you talk about building a model, sometimes people build more of a mental model, but it seems like for you it's very numbers-based. I imagine you sort of run these songs through the Excel sheet?

Yeah, so I'll run the numbers in Excel and get a high, medium, and low range for what the streams will be throughout the week, and then I just narrow that down as more data comes in.

So, that's a fairly simple one, but for something like Rotten Tomatoes, that's actually about six models that run at different points throughout the week because the amount of data that you get is different. So, before the social media embargo lift, all you can really find is random people rating it on Letterboxd, IMDb, or sometimes Google, or pulling reactions from Twitter. But as you get closer to release, you get more and more data. And then once reviews come in, you need to account for all the reviews that you can find on websites that haven't been published yet and how likely they are to be posted to Rotten Tomatoes in time. So, I take all that information and each step of the way has a different model that applies to it.

That's very cool. I've seen you talk too, though, that sometimes you do bets for fun, kind of based on gut feel. I'm wondering if the two betting styles ever overlap. Like, you have the model, but every now and again you're like, "No, this song's a banger. I bet it actually does better than the model says."

Yeah. So, today's Thursday and at midnight Eastern they're going to drop some new songs, like the J. Cole album's gonna drop. And if something big is coming up I'll usually stay up until midnight and just listen to them. And at that point, you don't really have any data on that specific song, so it really is a matter of listening to it and seeing if it's going to be any good or perform well. But for something like J. Cole, where it's an album people have been waiting for, I'll go back and compare how he did on his last studio album, which I think was in 2021. He did a mixtape release in 2024, as well. So I'll see how he did in the past and then consider if his popularity has dropped or if it's risen. I think it's probably dropped a little, but not enough that he won't get number one for the day tomorrow. So, it's a little bit of gut and a little bit of research on how that particular artist has done.

I like that a lot. A mixture of hard data and cultural analysis.

Yeah, for sure.

But yeah, I think that's a good overview of what it means to build a model. Let's say somebody's just starting out today. They've probably lost some money gambling on sports or whatever and they're trying to get more sophisticated. Where do you think the highest upside is these days? Where are the most inefficient markets? Where could somebody actually go out and find edge, in your mind?

Read the full interview here.

Some markets to watch for the weeks ahead:

Polymarket has a market on “Measles cases in U.S. by February 28?” We’ve written on measles markets in the past. This is one to watch for a late-month resolution. Currently the odds that there will be at least 1000 this month are at 97%, which seems pretty reasonable given we’re currently at 910 cases this year today.

Oscars markets -- best picture, best actor, best actress, etc. -- will resolve on March 15, 2026. These markets often have lower liquidity than some of more headline-grabbing markets in politics and the like, but many might be worth your attention. One Battle After Another leads “best picture” with odd of 75%. And Jessie Buckley (89%) and Timothée Chalamet (79%) are favorites for actress and actor, respectively.

Coming off the government shutdown “When will DHS receive full-year funding?” is another one to watch. We find this one particularly interesting because if you’ve been following these markets for the past few weeks, you can take a lot of your research and insights and apply them here. Expect more here.

Welcome to the transcript of OddChain’s conversation with top Kalshi Trader Gaëtan Dugas. Here we discuss:

• Finding Edge in Spotify and Billboard Markets

• Where are the most inefficient markets?

• Are the markets getting more sharp and sophisticat …

• How to tell if there are sharks in the water

• Trading Tools, LLMs, and AI

• What markets Gaëten is looking forward to in the m …

• Trade before settlement or hold until the end?

• Where the edge often lives in politics markets

• Other markets

• The Origin of Gaëtan Dugas

Let’s dive on in!

This transcript has been edited for clarity and length.

Welcome to the inaugural episode of what will eventually become the OddChain podcast. For now, we're going to be interviewing some of the top minds and traders in the prediction market space. To that end, we're very excited to welcome as our first guest, none other than Gaëtan Dugas. I really appreciate having you here, Gaëtan.

Hey, I didn't realize I was the first guest ever. What a tremendous honor.

Introduce yourself to the listeners a bit. Want to tell everybody who you are? How long you've been around in prediction markets?

My avatar, my nickname I go by online is Gaëtan Dugas. I guess, I've been doing it for a little over ten years now. I started on PredictIt back in 2015. I mostly did politics up until like 2021 or so when Polymarket started to offer more options.

I'm probably about a Top 100 trader on Kalshi right now. I've made $581,000 on Kalshi, and I've only been doing it for a little under two years on Kalshi.

And then, regarding my background, I studied finance, and I got an MBA as well. And then I keep my day job as a business analyst. It's an IT role. I'm basically telling developers what to do.

I love that. Do any of your coworkers know what the night shift gig is or do you keep that to yourself?

Yeah. Like, when I bring it up, or if we're in a meeting and I bring it up, it becomes the topic of the rest of the meeting. They're pretty fascinated by it. And then I'll send them The Washington Post articles and The New York Times and stuff like that as well.

That's so cool. It certainly makes for good water cooler stuff, I'm sure. But one of the things that I appreciate about your presence is that I think you're one of the social media figures who's really good at giving good advice to newcomers. 

You talk a lot about how to build models for markets, and about really data-backed theses for these things and how to find edge with those models. Can you go into a little bit more detail about what that actually means from a practical perspective?

Sure.

Finding Edge in Spotify and Billboard Markets

Like, if you have an edge, how do you build to get that?

Sure. So, I'll walk through Spotify as an example since it's pretty easy compared to a lot of the other ones and the data is so easily accessible.

So with Spotify, they've got their whole daily history of streams of the Top 200 for global and in every country, going back to like 2017. So, you've got access to this entire database of streams. And you can pull the data and start analyzing it, and you'll start to recognize patterns.

So, there are patterns that happen when a song is released. Friday is typically the biggest day. And then Saturday, it will drop by a certain percentage. Sunday it will drop another percentage. On Monday it usually bounces back. On Tuesday it bounces back a little more. And then from there, it does a slow decay. So you can spot that type of data. And you can look for comparable songs in the history to new ones that are coming out and kind of map out what and how you think it will do.

So, when it comes to modeling, I've got different models for different types of songs. A song like "Please, Please, Please" by Sabrina Carpenter or "we can't be friends" by Ariana Grande. Those are songs that did very well; they didn't drop very much.

And then there are songs that kind of flop, like "Aperture" by Harry Styles was a recent one where it dropped a ton.

And then there's kind of the average one.

So, basically what I do is I try and come up with a range of how songs will perform, and then once I track it a couple days, then I can get a pretty good idea of which of the categories it falls into. And based on that data, I can start projecting how it's gonna do the rest of the week.

Does that make sense as an example?

Yes, I can kind of see the spreadsheet for that too, right? You sort of slot it in. When you talk about building a model, sometimes people build more of a mental model, but it seems like for you it's very numbers-based. I imagine you sort of run these songs through the Excel sheet?

Yeah, so I'll run the numbers in Excel and get a high, medium, and low range for what the streams will be throughout the week, and then I just narrow that down as more data comes in.

So, that's a fairly simple one, but for something like Rotten Tomatoes, that's actually about six models that run at different points throughout the week because the amount of data that you get is different. So, before the social media embargo lift, all you can really find is random people rating it on Letterboxd, IMDb, or sometimes Google, or pulling reactions from Twitter. But as you get closer to release, you get more and more data. And then once reviews come in, you need to account for all the reviews that you can find on websites that haven't been published yet and how likely they are to be posted to Rotten Tomatoes in time. So, I take all that information and each step of the way has a different model that applies to it.

That's very cool. I've seen you talk too, though, that sometimes you do bets for fun, kind of based on gut feel. I'm wondering if the two betting styles ever overlap. Like, you have the model, but every now and again you're like, "No, this song's a banger. I bet it actually does better than the model says."

Yeah. So, today's Thursday and at midnight Eastern they're going to drop some new songs, like the J. Cole album's gonna drop. And if something big is coming up I'll usually stay up until midnight and just listen to them. And at that point, you don't really have any data on that specific song, so it really is a matter of listening to it and seeing if it's going to be any good or perform well. But for something like J. Cole, where it's an album people have been waiting for, I'll go back and compare how he did on his last studio album, which I think was in 2021. He did a mixtape release in 2024, as well. So I'll see how he did in the past and then consider if his popularity has dropped or if it's risen. I think it's probably dropped a little, but not enough that he won't get number one for the day tomorrow. So, it's a little bit of gut and a little bit of research on how that particular artist has done.

I like that a lot. A mixture of hard data and cultural analysis.

Yeah, for sure.

Where are the most inefficient markets?

But yeah, I think that's a good overview of what it means to build a model. Let's say somebody's just starting out today. They've probably lost some money gambling on sports or whatever and they're trying to get more sophisticated. Where do you think the highest upside is these days? Where are the most inefficient markets? Where could somebody actually go out and find edge, in your mind?

So, I think some of the most inefficient markets are in culture as well as in politics. I think a lot of people come into those markets and just bet based on vibes or guts without really understanding what's going on. And then they'll do so in just ridiculous volumes, especially on the politics side.

For example, on Tuesday, they had the House vote for reopening the government. And when the Speaker of the House opens or schedules a vote, he doesn't do it without having a pretty good idea that he's going to win the vote, right? Because it's pretty embarrassing if you put something on the floor and you lose, especially something as high profile as a government reopening vote. So, he put it on the floor, and the vote came in, and it stalled with four Republicans not voting, and two Republicans voting no.

And if you have any experience following this stuff at all, you would know that that happens every single time. And only one of those no votes is gonna stick, and that's Thomas Massie. It's the same guy every single time.

So, the market just went nuts. Like the price of it not reopening went down from like 91¢ where it was when it peaked in the early morning to the 60s, which was completely stupid because if you follow this stuff, you know that not only does this happen every time, it's the same people every time and they cave every time.

So, I was able to put pretty much my whole portfolio, like $160,000 I had open, onto this bet that the government would reopen. Then, of course, they ended up voting yes. The one no flipped, exactly like anybody that's following it knew would happen, and I ended up making like $40,000.

But it's one of those markets where there was almost infinite volume coming in on the wrong side, because I think they must have been pushing it hard on Robinhood or Coinbase or some of these affiliates, where people see it as a chance to hit a big bet when people that follow it in detail and know what's going on know that it's a big loser.

But it's one of those things where there's an edge available there, but you've got to do the research to know which side is going to be correct.

Yeah, the sheer amount of money coming in, I think that's a really interesting point. I remember when the shutdown first happened, there was this rumor going around that there was a lot of insider money betting that the shutdown would not happen. And insider money, you know, historically just means it's a fresh account and it's a big bet. But there were a lot of people coming in and voting no who were just new accounts and clearly not insiders because they were wrong.

I think it's an interesting time for prediction markets to the extent that new volume records almost every month sort of changes how people have to look at these things from an analytics perspective.

And some of this stuff isn't that susceptible to insider trading because in this case it's 435 people voting, or in the Senate it's 100 people voting. Who's going to have insider information on how all of that's gonna go? Often they're not sure until it gets to the floor and the votes start coming in.

Are the markets getting more sharp and sophisticated over time?

Yeah, impossible to know anyway. That's funny. I think that leads to a more interesting point. You know, you've been here a couple of years now. It really seems like the landscape's changing due to this sheer amount of volume coming in. I've sort of heard the scuttlebutt about more professional market makers -- the Jumps or Citadels of the world or what have you -- coming in to market make. And more professional sportsbooks and oddsmakers and things like that. Are you seeing the markets become more sharp and sophisticated over time? Or is it mostly retail money coming in? Do you see that on a market-to-market basis -- depending on the subsection -- or is it more that the whole vibe is shifting?

The world that I play in, I think it's actually become dumber as the volumes have grown. Because me and my friends who are sharp at this stuff, we're constantly running out of cash because there's just so much dumb money coming in for us to take.

So I think I've got almost $300,000 in my account right now, and I rarely have much cash available because there's just so much free money that I can throw it at.

That's an incredible answer. I love to hear that.

And I wouldn't have predicted this. I thought it would get sharper over time. But it seems to be getting -- or at least staying -- pretty dumb so far.

How to tell if there are sharks in the water

It does seem like spreads are closing a little bit, just with the amount of liquidity. I heard you mention on a different podcast that tight spreads to you signaled sharp money. Is that still the case? What are the signals you look for that indicate there might be some sharks in these waters?

Well, it's how tight the spreads are, but it's also what types of markets they are, like in the sports. I still think sports are extremely sharp. I think the financials are extremely sharp. And I think politics can be sharp. It's just, I happen to be on the right side of that.

So they'll be tight spreads, but I'm the one with a 100,000 share bid that's hoping that the dumb money will buy it.

Trading Tools, LLMs, and AI

That makes sense. Let's talk about how you can get sharper. I think that with Claude Code and some of these tools… you were already familiar with how to work with an API and how to call some more advanced data from some of these websites and get edge. Have you started incorporating any more advanced tools into your trading at all or do you just already have your setup and you're good to go?

So I'm actually kind of a moron when it comes to the tech stuff. I've got a friend that sets all that stuff up for me. Like I know Visual Basic 6 from 2000 or something like that. I just haven't bothered to learn anything new. So mostly, I come up with the ideas and my friend codes it for me.

We have been pulling in some AI tools, but we don't really use them to generate models or anything like that. What we do is have them interpret data for us. When we find a review or a tweet or feedback on a movie for Rotten Tomatoes, we don't want to manually interpret all of those. So we pass them into an AI tool to get a rating and to be able to incorporate that into our existing models.

I also use AI tools as kind of a sanity check on a bet. I don't think just asking it for odds works; it's almost always way off. But if I've got an idea for a play I want to make, I'll run it by the tools to get some feedback on it. Oftentimes the feedback is stupid and I have to say, "Oh, you didn't think of this, did you?" and it'll be like, "Oh yeah, you're right." So, I don't think it can replace human logic yet, but it is useful as a second check.

I really like that idea. A lot of people use it to just ask, "Give me the odds that this will happen," right? But all it can really spit out is what the wisdom of the crowd already is. Like it's just doing an internet search, right? Just a really glorified internet search. But in the converse, if you bring it to an idea, it is almost asking the crowd if the idea is good or not. That's very smart.

But some of the answers are comically bad. On the music stuff, it was saying Taylor Swift was a 90% favorite last year when she was actually like a 99% underdog at the end. It doesn't have access to the same level of granular data that I can access from Spotify and Kworb and stuff like that. So it's very unreliable.

But that still makes sense to me, though. Because if all the internet is just wine moms who like Taylor Swift, then that's what the AI will tell you, right?

You'll see it in the comments all the time, just people repeating verbatim what the AI tools are saying and then taking positions on it. It's just a good way to lose money fast.

What markets Gaëten is looking forward to in the month ahead

I love that. Let's get into some actual markets. You had a tweet going over some of the stuff you're excited about this month. Would love to hear you walk through what's going to keep you up until midnight, what you're excited about, what you're going to be trading?

Probably the biggest one is continued congressional action on the Department of Homeland Security. They passed a continuing resolution that goes through February 13th. So that means that all the other departments are fully funded through October 1st, but DHS is only funded through February 13th, and it was funded at the current level, so they didn't implement any changes. So, the standoff here is that the Democrats want to put a bunch of guardrails around what ICE agents can do, and the Republicans don't want that. So, they're kind of at a standoff and they've got very little time to resolve it. So there's a market on when DHS will be fully funded for the year.

I've got a ton of shares at 50¢ for “No” that it won't be done by March 1st. And that one's gone up quite a bit. It's at 70¢ now. So, that's one to watch. Once they start talking about doing another continuing resolution that goes through, like, March 5th or whatever, I might hammer on that a little bit more. I'm kind of just waiting for it to get to the point where the path is clearer. Because they could still come to some sort of resolution and pass it. But, at some point, they're gonna probably choose a different path, and once that path is clear, then I'll be more comfortable sizing up on this bet. So, that's one thing I'm watching.

Another one that's been great is they introduced a new market, actually two new markets. One of them is: Who will have a #1 song on Spotify USA in February? And the other one is: Which songs will be #1 on the Billboard Hot 100 in February?

So, for the Spotify one, that's 28 days of data that can resolve the market. So far, I think it's been Noah Kahan and Bad Bunny have been the two. But there's still 24 days or whatever of data that can come in that can have somebody else get number one. So, that's kind of fun to watch and kind of scope out. And Bruno Mars is gonna drop an album on the 27th. Will his album be popular enough to get number one?

And then J. Cole drops his tonight. Will that be popular enough to get number one? Because Bad Bunny is doing some huge numbers right now. But I think that with the album drop, J. Cole will be able to pass him for at least one day. But it's that type of thing, kind of looking forward to what new releases are coming out. And then Taylor Swift is going to drop a music video for "Opalite" on Saturday, and she's dropping it on Spotify and Apple Music, because YouTube doesn't count towards Billboard anymore.

So, this is kind of a new thing. We haven't really seen this before. And we can see how that is going to impact the Spotify numbers. Will that video be enough to get her in first place on Saturday? I don't know. So those types of things are pretty interesting to me.

It's kind of interesting how there's some overlap in there as well, right?

Yeah, definitely.

Trade before settlement or hold until the end?

These markets might interact with one another a little bit. I'd like to talk a little bit more about some of those, but before I do, you were talking about your shutdown market position. Do you ever trade around the positions before settlement or are you the type of person who tends to hold until the end?

Most of the time I buy and hold until settlement, but for some markets I don't. Like the last shutdown market, I probably took six different positions throughout the week, just playing different news plays. And my win percentage wasn't great. I think I probably got half right and half wrong. But the ones that I got right were such big wins that it offset everything else. And the ones I got right, I bought the morning of the shooting. I made like $15,000 just by buying “Yes” on shutdown for so cheap, but then I ended up selling out of that and taking profits and then flipping to the other side to “No” because “Yes” was at like 80¢, so I picked up some “No” at 20¢. And that play wasn't profitable.

And then I did a couple more plays like that where I just thought it would go one way or the other and it didn't move much, so I ended up cashing out. And then the last play that I made that was profitable was when a tweet came out where Thune basically said, "Well, if we're close, Trump just won't shut down the government." So, as soon as I saw that, I was one of the first to act and I just hammered. I probably bought 30,000 shares, and then I sold it a couple of minutes later for an $8,000 gain.

So I do that sometimes, but the ultimate goal is to be on the right side at the end and be on the right side with sufficient size that I can make a good profit.

Yeah, I often find it's much easier to sort of hold the swing if you think you're going to be right in the end anyway. And so I try to avoid taking the opposite end.

With the congressional markets, it usually gets to a point where the path is clear. And once it gets to that point, like when Mike Johnson started saying, "Okay, we're going to do the vote tomorrow." Or, even before that when they started saying the House wasn't gonna come back in time, so they weren't gonna vote until probably Tuesday. So, at that point, it was clear that the shutdown was going to happen and that's when I was able to shift a large position onto the winning side. But I still ended up making a lot of money before that just playing the swings.

Where the edge often lives in politics markets

Yeah, nice. I think a lot of the politics markets, you know, I don't think it'll ever be my specialty. It's just not the style that I think we're developing over at OddChain, but like headline-based stuff and the huge swings that can come out of it, it seems like politics is full of that.

Yeah, for sure. And what's nice about the politics markets and where the edge comes in is that it's not a market where there's a data point announced and somebody can create a bot to find that data point and immediately trade it. You need to do some level of analysis usually.

So, that lets somebody like me, where I'm just looking at Twitter -- I don't have any bots scraping or looking for information here -- I'm just looking at Twitter. I can see a headline and I can quickly interpret what it means and then go and trade based on it.

Whereas if it was something like the TSA numbers get released, as soon as those numbers are released, the market's over because it's black and white. You either win or you didn't.

With these things, you need to put some level of political analysis into the news.

Other markets

Yeah, I think the one I sort of have my eye on with all that in mind is the Iran strike markets. I think they're like crazy deep liquidity, but seem just laughably headline driven. You know, somebody shoots down a drone, the president says something absurdly bellicose, the Ayatollah says something absurdly bellicose. Like these things can go up and down and zig-zaggy so quickly. I'm wondering if you're watching that one at all or if it's just one of those things that's too high variance? Like it's really hard to find out how it'll actually end. How are you thinking about that one?

So, that's not one that I have on my radar. I don't do a ton of geopolitical stuff. Especially stuff that you have to monitor almost 24/7. Because the Iran stuff can happen… Like the Venezuela stuff happened at like three in the morning. And Iran stuff can happen pretty much any hour.

With the stuff that I do, I can kind of time-box it. Like usually not a ton of stuff happens off hours in politics. And the music stuff I do, there's a data drop each morning for Spotify. There are data updates for Apple Music throughout the day, but I just have those pushed to Telegram for me. And then the Billboard stuff drops on Monday afternoon. So, I don't need to be glued to my phone all the time.

But I think for something like Iran, where it's an ongoing story, there's no clear end date, you really have to make a commitment to follow this stuff almost 24/7 for an indefinite amount of time, and I'm just not willing to make that commitment to a market like that.

Yeah, I mean, somebody who is, though, you can imagine the guy who has the alert set up, the pager duty. Something happens, they can sell out of a position, they can sell into a position… It seems like where the edge is just your ability to either be awake or wake up. I don't think I'm there.

Yeah, and every trader is different. There are lots of people that are willing to put in 12, 13, or more hours a day on this stuff, and that's just not where I am at this phase of my life. I don't have any interest in doing that. I want to be profitable, but I also want to have a good balance. I've got three young kids. I like to travel a lot, so I don't want to make my life prediction markets. I want it to be something that I do to make money, but not something that consumes my life.

Healthy as hell. Yeah, you were talking on a different podcast, like you took two months off last year, which I think is really cool. Like, you don't hear about serious prediction markets people often touching grass to that extent.

Yeah, I took the entire summer off. Just because it gets stressful to monitor markets for months and months on end. So, you've gotta find a balance so that you can keep your sanity.

Makes total sense. Well, we've been chopping it up for about half an hour. I want to let you go soon. But before we do, are you doing any Olympic stuff?

No, I kind of avoid sports because I'm awful at them. And one of my New Year's resolutions was to not do sports this year. I think I lost like $20,000 on sports last year. So I thought, you know what? That's probably not for me.

Again, incredibly healthy. I guess, like, the reason why I think I like sports is also the reasons why they're really tough to trade. Like big upsets can happen. You can have these kinds of miracles in the last minute. That's like why you watch this stuff, but also it just makes it gut-wrenching to move around. I don't know though. And we were talking earlier in the pod about how they're also just filled with sharps at this point. But something tells me that, like, curling isn't going to be as…

Right, right. There's edges to be had, but it's just not my thing. I'm pretty comfortable in my space of politics and culture and I'm able to put almost all my capital into play almost all the time, and I'm making a ton of money doing it, so I don't really see a need to expand my field of expertise right now.

The Origin of Gaëtan Dugas

Really smart. Really appreciate it. So, one last question before you go. Gaëtan Dugas is an anagram?

No. If you look it up, he was rumored to be patient zero for AIDS. So, he was like the first guy. He was a Canadian flight attendant and he was kind of a slut, so he'd just fly around and have sex with a bunch of men. And they thought he was like the guy that really spread AIDS around. But it turns out that he probably wasn't. But he's just an interesting historical character, and I thought it'd be kind of funny to pick a guy like that where people might… like a lot of people probably just think that's my real name, but if you get curious and look it up, it'll be like, oh, that's kind of interesting.

I love it. Interesting energy to bring to your prediction market alter ego.

Yeah, well, it was back in the days of PredictIt, and the PredictIt comment section was kind of a cesspool with people insulting each other and kind of like a toxic masculinity thing, so I just thought it'd be funny to pick a historical figure that was obscure and primarily known for having sex with men, even though I'm a 46-year-old heterosexual guy.

Incredibly based.

I do stupid shit all the time. If you follow my Twitter, you're probably well aware of that.

Look, we're all in prediction markets, man. This is good stuff. But yeah, really appreciate you coming on. Anything you want to shout out, any projects you're working on or let the people know where to find you?

Yep, so I'm @GaetanD on Twitter. That's pretty much the only place I do anything. So, follow along. I post a lot of prediction market stuff, a lot of politics, a lot of culture, a lot of bad jokes.

Down to earth and not just posting the same one tweet about some profitable trader someplace.

No, no. It's basically a place for me to capture my thoughts and if people are interested in reading that, you can.

Awesome. But yeah, once again, really appreciate you jumping on. Really love that you took the time and, um, yeah, looking forward to seeing how you perform in February.

Yeah, thanks. Thanks for having me on, and I look forward to watching your platform grow.

Thanks, man.

Cheers.

If you haven’t already, we hope you’ll subscribe, so you don’t miss a thing. And if you want to run your own mention analysis, check out our Mention Markets x Earnings Call Dashboard.

Also, if you haven’t listened to it yet, we just had a great conversation this morning with one of Kalshi’s top traders, Gaeten Dugas. Give it a listen, and be sure to follow us on X, so you don’t miss the next one:

Traffic. Reddit Pro. Machine Translation. Developer. Game. Emplifi. Holiday. Algorithm. Smartly. OpenAI / Open AI. TikTok.

The call is Thursday, February 5th, 2026 at 5 p.m. ET.

This market is a tricky one, and we’ll get to why…

Currently, some potential mentions are aligned with our expectations and analysis.

Traffic and Reddit Pro are at 99% and 97%, respectively, and they’ve been mentioned on most calls. Reddit Pro, however, was skipped in Q2 of 2024.

Traffic is central to Reddit's earnings narrative. On the Q3'25 call, CEO Steve Huffman noted that “Reddit is the #3 most visited site in the U.S., per Semrush October 2025.” And then there’s the recent Max Campaigns beta that is explicitly for “select advertisers running traffic and conversion campaigns.”

So we -- and apparently a lot of other people -- think “Traffic” is more than likely a safe position, even if it’s only pennies at this point.

Now, Reddit Pro is a key part of Reddit's pitch to both advertisers and publishers, and we figure any update on publisher adoption or new Pro features for Q4 would naturally reference this.

One reason why machine translation may be slightly underpriced is that it’s been mentioned on every single earnings call to date. 

Although, it’s worth mentioning that the number of mentions dwindled to only one mention on the last call: 

But a “problem,” or rather complication, with this market is that because Reddit recently went public, there aren’t many earnings calls to reference. Because of that, we’re uncertain about many of these and the market seems to reflect that…

We think two likely to be safe positions are “No” on both OpenAI/Open AI and TikTok. 

OpenAI was mentioned in 2024 in the context of their deal with Reddit, but there has been no mention of them in 2025 earnings calls. Although it was reported last September that Reddit wanted better licensing arrangements with Google and OpenAI, there was no mention of OpenAI in the earnings call shortly following that report, and we don’t currently see a strong signal that they’ll be mentioned. 

And even though Reddit overtook TikTok in the UK, we just don’t see that as strong enough evidence for Reddit to mention a competitor, unless it’s brought up in the context of a question, but even then a Reddit representative would have to say the word.

Then that brings us to a whole lot of maybes…

Developer has been mentioned on 5 of 7 calls, but the trend is trailing, with nine mentions in Q1'24 then four, then zero, zero, one, one, and two. On the last earnings call, Huffman specifically said:

There’s still a chance it’s mentioned, but based on Huffman’s own framing, it doesn’t seem like it’s a centerpiece of their earnings narrative.

Holiday has only appeared on two of seven calls, but that number is a bit misleading, because mentions of “Holiday” are most relevant for Q4 calls, and there has only been one so far. In the last Q4 call, the word did appear:

Last October, Bloomberg reported that Reddit's Q4 guidance was literally framed as a "holiday-quarter forecast." And on the Q3'25 call, an analyst specifically asked about the "4Q holiday shopping season," putting the word in play. However, we don't have strong conviction here. The sample size is painfully small, and management might prefer phrases like "Q4 seasonality" or "fourth quarter strength" over "holiday" specifically.

That said, after tonight, we'll have more data to determine whether this is a pattern.

Smartly probably won’t come up unless it’s asked about. Smartly has been mentioned on the last two calls (Q2'25 and Q3'25), but the trend is declining.

In Q2'25, it was in prepared remarks:

In Q3'25, it was mentioned in response to an analyst question about lower-funnel products:

So unless it's asked about, we don't think there is a high likelihood that it is mentioned. The integration is now established, so it's no longer the "new thing" that necessarily merits a prepared-remarks callout.

Emplifi we don’t have a strong take on, but there’s some background we wanted to provide that you might find helpful. Emplifi has never been mentioned on any Reddit earnings call because it's a brand-new partnership.

On January 15, 2026, Emplifi announced a strategic integration with Reddit's Enterprise API. Reddit's VP of Business Development Jonathan Flesher was quoted in the announcement.

Reddit does have a history of name-dropping new partners in prepared remarks. In Q3'24, CFO Drew Vollero announced the Meltwater partnership by name:

In Q4'24, they announced a partnership with Intercontinental Exchange. And in Q2'24, they named mParticle and Tealium. 

Naming new partners is part of the earnings call playbook, but as of late, their cadence shifted. In Q1'25, partner mentions dropped sharply to two generic references. And by Q2 and Q3'25, the focus moved from naming individual partners to discussing ecosystem-level progress. The precedent for name-dropping new partners exists, but the recent trend is away from it. Since Emplifi is a martech middleware company, not a well-known consumer brand, we genuinely don't have a strong take here.

Now, one last thing…

Will there be another government shutdown?

If you ask Polymarket, there’s a 67% chance there will be. We have more to come on this in the coming week, but to get caught up on what’s going on, and to get ready, check out our past write-ups on the government shutdown:

• Roughly $24.5 Million Hangs in the Balance

• A looming government shutdown?

If you want to run your own mention analysis, check out our brand-new Mention Markets x Earnings Call Dashboard.

Chipotle Lane / Chipotlane. Inflation. Supply Chain. Transaction Growth. Al Pastor. Promotion. Tariff. Carne Asada. Share Purchase. Honey Chicken. Autocado. Automation

This afternoon Chipotle has its 2025 Q4 earnings call at 4:30 pm ET. We read through their last 12 earnings call transcripts (sourced from Morningstar) and counted each potential mention using our Mention Markets x Earnings Call Dashboard Tool.

In many cases, the markets are currently aligned with our expectations.

Chipotle/Chipotlane, Inflation, and Supply Chain all seem highly likely to us.

Right now, we’re looking at “Supply Chain” potentially being slightly underpriced, given that Supply Chain has been said in the past 10 of 12 calls; it’s usually mentioned within the context of the CEO’s discussion of strategic priorities.

Another market that we think might be underpriced is…

Share Purchase, which is at 69% at the time of writing.

The exact phrase "share purchase" appeared exactly one time on each of the three 2025 calls, always in the same CFO boilerplate line:

You’ll see something interesting about this when you look at our mentions breakdown:

Prior to Q4 of 2024 there were no mentions of “Share Purchase.” Chipotle’s old CFO, Jack Hartung, instead used the phrase “repurchased” during calls, saying things like "During the third quarter, we repurchased $226 million of our stock…”

But starting Q4 2024, the language flipped. The new CFO Adam Rymer used the phrase “Share Purchase,” saying “the board authorized an additional $XXX million to our share purchase authorization.”

If Rymer stays on script today -- which he has for three straight calls -- we can probably expect exactly one mention of this.

We think Al Pastor and Carne Asada are solid maybes, and the pricing reflects that.

Chipotle recently announced it’s bringing back chicken al pastor starting February 10th as a limited-time protein option. Executives plan to launch three to four proteins in 2026, up from two historically; they may very well frame this as part of an accelerated innovation strategy to drive traffic.

Carne Asada was a past limited-time protein, so they may mention it in the context of how it performed and contributed to Q4 results.

That said, for ourselves, if we had to pick between “Carne Asada” and “Al Pastor,” we’re currently leaning “Al Pastor.”

We know this is a quick one, but a few final points…

We’re not touching Autocado. It appeared on 10 of 12 calls historically, but the trend is clearly fading, going from 2 mentions to 1 to 0 over the last three quarters. And since management now talks about the same kitchen tech using phrases like “equipment package” and “produce slicers” without naming Autocado specifically, we don’t see enough signal to take a position on this one.

We think Automation is probably a “No.” The lack of any mention of “Automation” on the last three calls might indicate that it was retired from management’s vocabulary. And we’re not seeing strong signals that would indicate a “Yes.”

Best of luck! 

- OddChain 

With ~$11,275,212 in volume on Kalshi and ~$13,242,144 in volume on Polymarket for "How long will the Government Shutdown last?" a lot of money is on the line.

We recently wrote on the markets leading up to the shutdown. So, at this point, if you're reading this, you probably know that the shutdown occurred and we're now all trying to figure out how long it will last.

Speaker Mike Johnson said on Fox News Sunday that he plans on getting it done by Tuesday, saying: "We'll get this done by Tuesday, I'm convinced." He added that “[t]he president is leading this."

The positions that are the most in play right now are "More than 4 days" and "More than 5 days"

“More than 3 days” seems like a lock at this point, and -- barring anything unexpected happening -- you can probably still pick up some pennies, but moves like that often aren't worth it.

The Pipeline of the Funding Bill

Because Democrats refused to fast-track it under suspension of the rules, the Senate-passed funding bill has to move through regular order. This involves three sequential gates, each of which the bill must clear:

1. Rules Committee — Monday 4 PM ET

2. Floor rule vote — likely Tuesday

3. Final passage — Tuesday, possibly late (could be delayed depending on how the floor rule vote goes)

Now, the rules committee vote this afternoon is important but probably not decisive.

The committee is 9R-4D.

Assuming that all Democrats vote no, Republicans need 7 of 9 to get it out of the committee.

Two potential R dissenters are Chip Roy (TX) and Ralph Norman (SC). Roy voted against seven procedural rules in the 118th Congress, and both voted against the One Big Beautiful Bill in this committee last July, which still passed 7-6.

However, Roy might not break from party lines. Roy told Fox News Saturday that Democrats were "playing politics" and expressed optimism the bill could pass.

Even if Roy and Norman defect again, this bill passes the Rules Committee 7-6. It’ll only die with a third defection, and none of the other members (Foxx, Fischbach, Houchin, Langworthy, Austin Scott, Griffith, or Brian Jack) have signaled that they intend to do so.

One thing to watch with this…

If the Rules Committee issues an open or modified rule, matters would likely be complicated as it would allow for floor amendments, and Anna Paulina Luna (R-FL) or allies could force a vote on the SAVE Act. And if that amendment passes on the floor, the bill is no longer identical to what the Senate passed and it must go back to the Senate for another vote. On this note, Schumer has called the SAVE Act "dead on arrival."

We’d be surprised if there were changes to the bill during the Rules Committee today.

The Real Chokepoint is the Floor Vote

Speaker Johnson needs nearly every R to vote “Yes” for this bill to pass. If it fails the House vote, the rule goes back to the Rules Committee for revision. Johnson could bring a new rule, but that adds at minimum a day, pushing resolution to Thursday at the earliest.

One variable that impacts the numbers here is whether newly elected Democrat Christian Menefee gets sworn in before the vote.

• Before Menefee the House membership is 218 R and 213 D. In this case, Johnson can lose 2 Republicans.

• After Menefee the House membership is 218 R and 214 D. In this case, Johnson can lose 1 Republican.

There are some potential Republican holdouts.

• Anna Paulina Luna (R-FL) is demanding the SAVE Act be attached to the bill. She earlier claimed (on the 29th) to have enough votes to shut down the floor. But her language has since shifted. On Monday she told Punchbowl News: "I cannot control the outcome... But what I can do is control my vote."

• Tim Burchett (R-TN) is named by Politico as sympathetic to Luna's SAVE Act push. Whether he's a firm no on the rule or just grumbling is unclear.

• Unnamed ultra-conservatives want to scrap the two-week DHS stopgap entirely and push for six weeks or longer.

• Some people might simply be absent. On any given House vote, 2-5 members are typically absent. And in a world where Johnson can lose only 1-2 votes, a single absent Republican is functionally a defection.

While drafting this piece, news broke that Rep. Anna Pauilina Luna is currently phoning Rs today to whip against advancing the funding bill:

The Final Passage

On a Sunday Democratic caucus call, senior lawmakers Steny Hoyer and Jim Clyburn voiced support for the funding deal. A CNN source said that if Johnson can clear the rule hurdle, there should be enough Democratic votes for the bill itself to pass. That means once the rule is behind them, Johnson has breathing room.

The exception would be progressive Democrats. Congressional Progressive Caucus chair Greg Casar has pledged to vote no, saying "not another cent to ICE until we stop the chaos." But that bloc might not be enough to prevent the bill’s passing.

A Note on Phrasing

When Do We Think It Will End

Our expectations are currently aligned with the markets.

Here’s what some scenarios might look like…

• Tuesday, Feb. 3. Rules Committee passes the bill Monday. Luna votes no on the rule but is isolated. The rule passes Tuesday, final passage follows the same day with some Democratic support. Trump signs promptly. Shutdown lasts about 4 days total.

• Wednesday, Feb. 4. Minor procedural delays, the Rules Committee session runs long, floor scheduling slips, or Trump signs Wednesday morning after a late Tuesday vote. Shutdown lasts 5 days.

• Thursday–Friday. The rule vote fails on first attempt, Johnson regroups, cuts a side deal with Luna (perhaps a promise of a standalone SAVE Act vote later), and brings it back. Or a surprise second Republican holdout surfaces. Shutdown lasts 6-7 days.

• Beyond Friday. The bill gets amended. SAVE Act attached or DHS provisions changed — and bounces back to the Senate where it dies. We're in an extended standoff. Or Luna's coalition is larger than anyone has reported. Shutdown lasts 10+ days.

Where we stand: If things continue as is we’re eyeing a “No” on “more than 4 days.” But given last week’s late night deal between Trump and Schumer, we’re currently waiting for a bit more signal.

So, follow us on X, -- if you haven’t done so already -- to stay up to date on any, well, updates.

If you want to run your own mention analysis, check out our brand-new Mention Markets x Earnings Call Dashboard.

Competition. Subscriber. Cruise. Abu Dhabi. Avatar. Zootopia. Moana. NFL. Doomsday. Acquisition. YouTube TV. WWE. DraftKings. Fortnite. Buyback. Weather.

We used our Earnings Call x Mention Markets tool to analyze Disney’s past earnings calls. (Results below.)

We think there are some outcomes that have a high probability of hitting, and the markets agree.  

Competition, Subscriber, and Cruise all seem highly likely based on past earnings calls and current market expectations. If you’re up for it you can probably squeeze out a few more dollars, but when it gets this close to 100% it can often feel like picking pennies off a train track.

One interesting note is that Disney is transitioning away from publicly reporting quarterly subscriber numbers and Average Revenue Per User (ARPU) for Disney+, Hulu, and ESPN+, as of the first quarter of fiscal 2026. 

As reported by Variety. “The execs said that “while we will no longer disclose subscribers and ARPU, we will provide information on Entertainment Direct-to-Consumer profitability.” That said, there are other contexts that “Subscriber” will likely be mentioned.

Abu Dhabi sits comfortably at 93%. Bob Iger visited on January 25th, 2026, with the presumed next CEO of Disney Josh D’Amaro. (Did anyone else catch that move yesterday evening when the Bloomberg piece dropped?)

There are also some markets that we think are likely underpriced. 

Here is our overview of past Disney earnings call mentions.

We removed the forward looking statements page when analyzing the transcripts as that skewed some of the results. For example, with the forward looking statements page “weather” was mentioned on every single earnings call, BUT that is not the case once the page is removed. Weather was actually only mentioned once in Q1 of FY23 in the context of an action that would “better [position] [them] to weather future disruption.”

So, “weather” -- which currently sits at 12% -- is pretty unlikely.

Zootopia and Avatar may be underpriced. 

As of writing, they’re at 93% and 88% respectively. And they were the number 3 and 4 top grossing movies for the domestic box office of last year. 

If we were Disney, we’d mention these successes, so we think there is a good chance they’re mentioned. And to top things off, they were both also mentioned in the last three Disney earnings calls. Also, Disney’s CFO explicitly flagged Avatar on the Q4 FY2025 call as a key driver of Q1 guidance, and analysts are very likely to ask about its trajectory.

Acquisition is a bit of a toss up. The word has been used in the past in the context of a Fox acquisition, subscriber acquisition, and the acquisition of the NFL network. The “problem” is that if they say “acquire” or "acquired" it won’t count. So, while we think they’ll mention this, the grammar-based variable gives us pause.

Doomsday was mentioned twice in the last earnings call, and it is expected to finish out the 2026 calendar year. An Avengers movie was mentioned in most earnings calls.  

An Avengers movie wasn’t mentioned in Q3 of fiscal year 2025. We’d say Doomsday is probably underpriced currently at 70%, but it isn’t a guarantee. 

We think “NFL” is currently underpriced.  An ESPN-NFL mega-deal literally closed today (February 1, 2026), with ESPN getting  regulatory approval to acquire NFL Network and RedZone in exchange for a 10% equity stake in ESPN. 

On November 14th Disney announced that they came to an agreement with YouTube TV. The Q4 call referenced YouTube Negotiations and this deal’s impact falls within Q1 FY2026; however, we don’t feel that there’s strong enough signal for much conviction on this one at the moment.

And then some of these we find pretty unlikely.

Weather, which was mentioned above, is one of them.

Buyback is another that we think is unlikely to be mentioned. They regularly do buybacks, but it’s not exactly a hot topic for earnings calls. And they were only mentioned in two of the earnings calls we analyzed. And they’ve instead actually used the phrase “share repurchase” more than buybacks on past earnings calls.

“Share repurchase” mentions:

• fy25-q4-earnings-transcript.pdf | 2

• fy24-q1-earnings-transcript.pdf | 2

• fy25-q2-earnings-transcript.pdf | 1

• fy24-q2-earnings-transcript.pdf | 1

We’re thinking a no on this.

DraftKings is probably unlikely. Disney announced that ESPN and DraftKings entered a multi-year agreement on November 6, 2025, and there was no mention of it on the earnings call only a few days later on November 11, 2025. DraftKings stock is also currently down. 

Best of luck! 

- OddChain 

A potential shutdown looms, as the U.S. government faces its second funding crisis in four months. Six appropriations bills funding roughly 75% of discretionary spending --including the Departments of Defense, Homeland Security, Health and Human Services, and Treasury -- expire at midnight on Friday. 

To pass, and for the government to remain open, the package requires 60 votes in the Senate, where Republicans hold 53 seats. Only one Democrat -- Senator John Fetterman of Pennsylvania -- has committed to voting yes. 

Republicans need six more votes. Meanwhile, Senate Democrats (for the most part, see above) have unified against the package following the fatal shootings of two U.S. citizens by federal immigration agents in Minneapolis this month, which an OddChain writer recently discussed. 

Since 1976, there have been 21 funding gaps. Most recently, in October and November 2025, we experienced the longest shutdown in government history. Democrats demanded extension of enhanced ACA premium subsidies, and after 43 days, eight senators broke ranks and voted to reopen the government. Democrats ultimately lost that fight, and the subsidies expired at the end of 2025.

As of writing, Kalshi has the odds of a shutdown at 73.2% and Polymarket has is at 74%.

I expect there will be some price movement that can be captured as news ebbs and flows throughout the day. 

Yesterday, late night reports that President Trump and Senate Minority Leader Chuck Schumer were progressing towards a deal led to a nearly 20% swing on Kalshi and Polymarket. Odds moved to close to 50-50 before adjusting this morning.

Before this -- based on research and conversations with current and former government employees -- my thesis was moving in the direction of a “Yes” being overpriced by about 11% and that was the safest approach based on the unexpected nature and rapid positional swings was to take advantage of limit orders and trade on headline-based sentiment swings throughout the day. I managed to capture 16% of the swing before bed, and I anticipate further volatility.

The markets already corrected back up this morning, I reckon due to the fact that even if Schumer and the White House come to an agreement, there may still be a funding lapse this weekend as things get sorted, and a partial shutdown would be sufficient --according to Kalshi and Polymarket’s rules of resolution -- for the markets to resolve to “Yes.”

That said, since this also relies on OPM’s website updating in time, to minimize risk traders might want to consider using limit orders as the market continues to shift based on the news. 

Onchain Indicators

On Wednesday night, amateur Twitter sleuths flagged a number of suspicious wallets that appear concentrated on the “No” side of the market. 

A common “flag” to spot a potential insider is a fresh account that takes a large position on a single market (sometimes the only market they ever bet on), and tends to market buy. (Because, as one OddChain writer puts it, why would a corrupt shmuck -- and presumably a new user of prediction markets --  know the intricacies of resting orders on an AMM?) 

As of 10:30 EST Thursday, six of the top ten “No” holders have accounts that were created in the last two months, which alone certainly raises eyebrows. On closer inspection, though, the allegations of an insider cabal seem a little overblown. 

One account has been accumulating their shares steadily using resting orders over a 48-hour period. Another is an active trader across multiple political markets. One started his account rather unprofitably, so if he is an insider, he’s a bad one. 

The remaining three are legitimately suspicious. All bought between 22 and 23 cents. Two addresses bought in batches of three market buys around the same time, and one bought (after a test buy) first two weeks ago. 

But: who knows? 

Big, single-sided buys could be from traders catching the arb between Polymarket and Kalshi, which has been as large as 3%. Prediction markets are routinely hitting all time highs in trading volumes as well -- could simply be new bettors. 

We’re wishing WickRick in particular good fortune in his pain avoidance strategy:

So, what’s going on and what might happen?

Initially, there was indication of progress being made in negotiations. Some members of the GOP indicated a willingness to separate the DHS bill. Thune, however, is not a fan of this, having refuted this idea multiple times and saying that “I would prefer that there be a way that we keep the package together.” If this bill were to split, the House would need to come back; Johnson has not signaled that to be the case.

Trump also signaled that he did not want a shutdown to occur, with Bessent telling CNBC that Trump was “‘urging’ lawmakers to avoid a looming partial government shutdown.”

Then, in the last day, a lot happened.

• In the morning there was some optimism. Senator Josh Hawley said, "We're at a situation where this just isn't safe…" Senator Thom Tillis said, "We're losing trust with the American public on an issue that we should be winning on." And Democratic negotiators sounded hopeful, with Senator Brian Schatz saying "I think we should be able to land this." and Senator Tina Smith saying, "Nobody has said, 'That's ridiculous.'" 

• In the afternoon, after a caucus meeting, Schumer announced three conditions: 1) End "roving patrols" and tighten warrant requirements for arrests, 2) Establish a uniform code of conduct for use-of-force accountability, and 3) Require agents to remove masks, wear body cameras, and carry visible identification. (It’s worth noting that these demands involve legislation, which require House approval.)  

• Senators dismissed some of Schumer’s demands and The White House unequivocally rejected them, saying (as reported in Roll Call): “A demand for agreement on legislative reforms as a condition of funding the Department of Homeland Security with a government funding deadline just 48 hours away is a demand for a partial government shutdown.”

• Politico reported that the White House is now pinning its hopes on border czar Tom Homan who was dispatched to Minnesota earlier this week to "cool temperatures." A White House official described Homan's efforts to "appease" Democrats as "the only arrow left in the quiver." There is no Plan B, the official said. "It's all wait and see."

• Then the big move in the market occurred when Schumer and Trump met last night. Progress was made negotiating a framework to pass five of the six funding bills. However, complicating matters is the structural difficulty of reaching a last-minute deal. Any changes to the six-bill would force House lawmakers to return from recess to approve the revised measure, and they’re gone until February 2nd.

Why the market might resolve “Yes”:

• There is no deal. In this case, the Thursday test vote fails, and there is a shutdown on Saturday. There is solidarity among Democrats, and there’s a solid chance they hold strong. Plus, Democrats who defected to end the government shutdown last time -- such as Cortez Masto -- have said they’re no’s this time around. 

• The package splits and there is a temporary shutdown. Several Republican senators have publicly endorsed stripping DHS from the six-bill package, and instead passing the other five measures that have bipartisan support. Again, this may still lead to a temporary shutdown. Even then, not everyone supports this; Senator Mike Lee of Utah, for example, has indicated he will reject any effort to strip DHS from the package. 

Why the market might resolve “no”

• The discussions that Schumer and Trump had last night pan out in a way where there is no temporary shutdown.

• The Democrats cave to Republican pressure. There’s no split and no changes to the bill. This requires an additional 6+ Ds breaking ranks or Rs to agree to split. I find this unlikely as the senators who broke ranks in October -- such as Cortez Masto, King, Hassan, Shaheen, Rosen, Kaine -- have all said they're voting no this time. However, one reason they might cave is that the shutdown wouldn’t actually be a big hit to ICE. Last summer's reconciliation bill gave DHS roughly $176 billion in separate funding, including $75 billion for ICE. Senator John Fetterman has made this his central argument: "A vote to shut our government down will not defund ICE.” 

Expect markets to continue to move as the news changes. Right now, it does seem to be leaning toward yes -- especially given the chances of short term shutdown even if the bill is split. However, things are changing fast, so looking to trade positionally as the market swings might be a more strategic option. Take advantage of those limit orders.

Keep an eye out for market moving headlines or for Ds breaking rank or indicating that they’re holding firm. I expect Thursday’s test vote to be a strong signal, and that there’s a good chance markets keep moving down to the wire. 

A final note on the matters of resolution. We’ve written on the complexities of resolution in the past (here and here).

Both markets on Kalshi and Polymarket resolve based on the OPM website. We were not able to confirm when exactly the OPM website was updated last time. (If you know this please get in touch.) If you’re worried about the OPM website updating, you can consider setting limit orders to capture large swings in your favor while minimizing risk of resolution disputes or delays.

This article does not represent the consensus views of Oddchain writers.

Exhaustion and analysis. 

• Consequences Markets

• Boots on the ground?

• Libertarians Find A Spine

Dead Americans make it easy to be wrong about things. Classic hot blood/red mist stuff. Facts tend to lose their axiomatic stability under tectonic pressure from reality-warping discourse that can last decades -- or even centuries -- after the relevant death. Pat Tillman, MLK, Aaron Bushnell, Assata: there’s a rattlebag of names that each make a fine example of the phenomenon I’m describing where historical consensus becomes, at times, permanently obscured as sweaty-collared polemicists apply their personal faction’s topspin to the relevant corpse. It’s the type of stuff that makes Thanksgiving tables a touch uncivil all over America, and your family probably has a personal fave they beef over. 

I don’t know for sure the circumstances surrounding Alex Pretti’s passing, the latest Dead American for us to fight over. Despite my efforts to hone my intuition for moneymaking reasons, I continue to lack the capacity to peer into another’s soul and perfectly divine their intentions. Who knows what he or what the agent who emptied clips into his body was thinking. 

But -- at the risk of alienating some portion of Oddchain readership in order to make what I hope will be an insightful observation about prediction markets -- I will say this: I believe, beyond any reasonable doubt, that Pretti was murdered without cause by the state. I believe that the Hague would be too good for the people involved. I believe any civil society would immediately disband ICE and prosecute everyone involved with its formation, operation, and administration, going back multiple presidencies.

When it comes to prediction markets, however, my beliefs don’t quite matter. In fact, they can actively impede my rational participation in the relevant markets. Let me explain:

On Sunday, as thousands of Minnesotans protested ICE occupation throughout the state, some friends and I were idly debating early Oscar predictions. One friend, in possession of extremely good taste and generally sharp logical faculties, insisted that Sinners was a strong underdog choice on the market, currently sitting around 25% odds. She believed, and argued with great conviction for myriad reasons, that it was a superior film when compared to the frontrunner, One Battle After Another. 

It took a good deal of convincing that the respective quality of the movies didn’t matter. The Academy is a whole lot of old people who’ve chosen a fantasy or Sci-Fi flick for best picture a grand total of three times throughout its 100 year existence (The Shape Of Water, Lord of the Rings: Return of the King, and Everything Everywhere All At Once). With that in mind, an argument can be made that on that basis it’s highly unlikely that the Academy will choose Sinners, which is a fun, sexy, vampire romp. On other hand, the Academy has nearly doubled in membership in the last 14 years, demographics have changed, places like Variety currently have Sinners as a favorite, and the narrative momentum of their Oscar’s campaign may be picking up. (Our full analysis is forthcoming. Subscribe to get it.)  

This type of thinking is a mistake that even seasoned prediction market participants can make. What you think should come to pass due to your beliefs and what will happen can be very different things, especially when you have strong ethical or aesthetic views. This was a market inefficiency that plagued the early days of Polymarket in particular, back when liquidity was lower and the average participant skewed crypto-libertarian in flavor. 

Often, the brighter the flashpoint the more obscured the historical truth. Emotions are running high; I am personally exhausted with debating right and wrong on social media. In the context of the markets, however, moral right and wrong are somewhat irrelevant, and instead we look to history to try and glean lessons and venture a guess as to what actually happens next. 

Thankfully, while we have to pretend to be rational market actors for this exercise, politicians emerge from their slime-sacks in such a state. The reptiles largely lack beliefs entirely, which to a certain extent make them extremely predictable -- unlike arguing about Fred Hampton with your drunk uncle at a holiday, suddenly we’re all playing the same game. 

So, for the sake of rationality, let’s look at a few examples. 

Consequences Markets

There are two live markets essentially devoted to the question of whether or not the killers of Alex Pretti and Renee Good will be held accountable. 

The one devoted to Alex Pretti’s killer is difficult to parse. A Minnesota judge has ordered the federal government to preserve evidence related to the death, which seems like something you’d do if you’re planning to press charges. 

Conversely, however, nobody likes charging cops with murder. Estimates peg the frequency that LEOs face charges for on-duty killings to be around 2%. 

The counterargument, of course, is that pressing charges against a member of an increasingly unpopular institution would be a fine way for some DA to kick off a political career -- they would hardly be the first, and the Minnesota judiciary seems particularly incensed. This market feels like a coin flip. 

By contrast, there is a market for Good’s killer that is structurally different. Instead of facing charges (though there is also a charge market), this market revolves around whether he will be fired or resign. 

Renee’s killer seems unlikely to be going anywhere before the 31st of March as even if he is charged by the Minnesota Government, and I believe he’s likely to be placed on paid leave, if anything, not fired or forced to resign. Administrative leave while under investigation is considered standard practice, and it can last upwards of years. 

In both cases, there is video evidence supporting the innocence of the victims. In the Derek Chauvin case, for example, video evidence is partly what led to multiple officers being quickly fired, but at the time Joe Biden was the President of the United States. 

Even though there is similarly strong video evidence that Alex Pretti was merely carrying a phone at the time of his public execution, the Trump administration has doubled down on the narrative that Alex Petti was a terrorist with a gun. Additionally, in this killing, the officers were federal and not local as in the Chauvin case. This means that Minneapolis and the Minnesota government aren't calling the shots this time, the federal government is. 

Interestingly, unlike in the Chauvin case, the NRA and other gun activist groups have pushed back against the Trump administration for killing a licensed gun owner. This means that there is pressure from more than one side of the United States political spectrum this time, but less sympathy, and arguably much more animus, from the Federal Government for the victim. 

There is a third Minneapolis public execution case that we can also look to for potential guidance: the killing of Justine Damond (Ruszczyk). Like in the Pretti case, Damond was white and Donald Trump was also the President of the United States at the time. Unlike in the Pretti case, there was no video of Damond’s murder (both officers had bodycams, but had switched them off), though a 16-year-old on his bike did take videos of the scene using his cell-phone after the shooting ended. However, unlike in the Pretti case, Damond was murdered by a Minneapolis police officer, not a federal officer. 

Damond’s death triggered widespread street protests, however, they were mostly localized to Minneapolis. In Damond’s case, the officer was fired nine days after the shooting, while charges took over 8 months to file (much more slowly compared to the Chauvin case).  

And then of course, as discussed above there is also the killing of Renee Nicole Good, 17 days before Pretti’s death. Similar to Pretti, Good was shot and killed by a federal agent. Like Pretti, she was also white and there are bystander videos which appear to contradict the claims of the officers who were present at the scene of her death, and like with Pretti, Donald Trump was President of the United States at the time of her killing. In Good’s case, the FBI and the United States Attorney’s Office immediately claimed exclusive federal jurisdiction, and blocked the Minnesota Bureau of Criminal Apprehension from accessing evidence, which the Minnesota government (perhaps learning from Good’s case) has attempted to preempt by issuing the injunction to preserve evidence in Pretti’s case.

The biggest factor affecting these cases appears to be whether the law enforcement officer who conducted the shooting is a federal or state employee. Or to put things another way, whether the police officer is technically protected by the Supremacy Clause.

What this means for traders is that it’s unlikely that any federal agent is getting fired any time soon. And, if they are punished, it’ll likely take the form of paid administrative leave. If anyone is going to pursue charges it will be the state and local government, and there will likely be a severe, multi-year delay, engendered by interventions from the federal government.

Boots on the ground? 

Resolving in four days, there’s an open question as to whether President Trump will deploy troops in response to the rioting. 

Recent reporting from the WSJ indicates Trump is distancing himself from ethnonationalist Stephen Miller, and he has implied he’s thinking about withdrawing federal officials from Minnesota entirely. 

I believe this one may be a bond.

Bonus: Easy Arb

Democrats should let the government shutdown as a defacto way to withhold funding ICE. Will they? I have no idea, but might be able to catch upwards of a 3% arb settling in three-to-four days if they click buttons at the right time. In short, by purchasing YES on one venue and NO on another with an equal amount of money on both, a user can earn the spread if both markets settle in the same manner. 

Libertarians Find A Spine

Politicians watch the streets. 

If either the rioting reaches a certain threshold of violence or the latest round of public polling emerges that shows a significant sentiment shift on ICE ahead of midterms, I think you can expect one self-styled libertarian -- for instance, Thomas Massie -- to suddenly stand up for their “beliefs’” And, if so, it’lll be the first time they stood since emerging from the pond where they spawned. 

Oddchain will be waiting for a big molotov-induced fire or a particularly damning survey, but will likely bid yes at some point. 

This article does not represent the consensus views of Oddchain writers. Stay safe. 

If you ask Polymarket, the United States did not invade Venezuela on January 3rd, 2026.

They did, however, arrest Venezuela's President Maduro during a military raid in Caracas. They captured him and his wife, flew them to the USS Iwo Jima, and then to Brooklyn, where Maduro was detained and pleaded not guilty to narco-terrorism charges. At least 80 people were killed. 

Later that morning, at a press conference from Mar-a-Lago, President Trump declared: "We are going to run the country until such time as we can do a safe, proper and judicious transition."

So, when the market for “Will the U.S. invade Venezuela by…” failed to resolve to  “Yes,” Polymarket users weren’t shy about their frustration:

And:

They felt the market’s rules had been incorrectly interpreted. 

The original rules for the market “Will the U.S. invade Venezuela by…” read:

The commenters here are correct that there’s some ambiguity in what’s considered to be “a military offensive intended to establish control over any portion of Venezuela” as some amount of control over some portion of Venezuela was necessary for the operation to be completed. 

Polymarket acknowledged this ambiguity when they issued a clarification:

The clarification, however, came on January 4th; this was after the raid and after Trump's "run the country" statement.

Although the clarification tried to be more precise, it still created new ambiguities:

• What counts as "intended"?

• What counts as "control"?

• Does removing a head of state establish control even without territorial occupation?

Plus, it added information that would have influenced traders' positions if known beforehand.

Unsurprisingly, this clarification didn’t go over well with everyone. Responses ranged from: 

And: 

To: 

Fair question.

What is an invasion?

It’s not super clear. Neither international law nor U.S. law seems to actually define "invasion."

In UN General Assembly Resolution 3314 (1974), the term “invasion” appears as an example in the definition of “aggression” where it says in Article 3 of UNGA Res 3314:

“Invasion” here is never explicitly defined, rather it’s used if the meaning of the word is self-evident. 

The Rome Statute's 2010 Kampala Amendments, which created the crime of aggression for the International Criminal Court, adopted the same approach; invasion appears as an example of aggression, but remains undefined. 

A clear definition is also not found in U.S. law. The Constitution uses "invasion" four times:

1. Congress can call forth militia to "repel Invasions." 

2. Habeas corpus can be suspended during an invasion. 

3. States can engage in war if "actually invaded." 

4. The federal government must protect states "against Invasion." 

But these all refer to being invaded, not conducting invasions abroad, and none address what constitutes a U.S. invasion of another country.

Article 42 of the Hague "Regulations concerning the Laws and Customs of War on Land" identifies "occupation" as: 

Article 2 of the Geneva Conventions, relative to the Protection of Civilian Persons in Time of War, makes it clear that: 

This makes it clear that even a partial occupation -- that is neither the subject of a formal declaration of war nor involving armed resistance -- will still be considered to fall under the rules of war.

Black's Law Dictionary defines “invasion” as "the incursion of an army for conquest or plunder." This would arguably cover Venezuela, given Trump's explicit statements about taking Venezuelan oil. But no federal statute provides a definition.

Further, the meaning of "invasion" is actively being litigated right now; however, this is in the context of immigration, not military operations. Trump issued an executive order claiming it does, which would trigger powers under the Alien Enemies Act of 1798. Legal scholars sharply disagree. And the Fifth Circuit is hearing cases on whether migration across the southern border constitutes "invasion" under the Constitution. 

A full exploration of this is beyond this piece's scope, but the point is that the “correct” interpretation of “invasion” is up to debate. And Polymarket’s use the word “invasion” as though it had a clear legal meaning when it, in fact, does not, added to people’s frustration and confusion.

A Matter of Resolution

A tricky aspect of prediction markets is that you’re not necessarily taking a position based on the outcome of the question; you’re taking a position based on the rules of resolution for that market.

It’s sort of like how, when Stephen Colbert asked Ruth Bader Ginsburg if a hot dog was a sandwich, her response was, “You tell me what a sandwich is and I’ll tell you if a hot dog is a sandwich.”

You can watch the full exchange here: 

Once he provided a definition, her answer was that it was considered a sandwich according to his definition of a sandwich. 

The same logic applies to this market. 

If you tell me what an invasion is, I’ll tell you if what happened in Venezuela was one.

The Problem with Vague Terms

A common pitch is that prediction markets are truth-discovery mechanisms. The premise is that if you aggregate enough bets from people with skin in the game and you surface the probability of real-world outcomes you can arrive at the “truth.” However, when contracts use vague or contested terminology, this starts to fall apart, because instead of measuring what actually occurred you’re measuring whether or not you think the outcome will align with the stated rules of resolution.

This isn’t the first time Polymarket has faced a definitional controversy. In July 2025, a $237 million market asking whether Ukrainian President Volodymyr Zelensky would wear a "suit" before July resolved to "No" -- despite BBC, New York Post, Reuters, and The New York Times all describing his June 24 NATO summit outfit as a suit.

The contract's criteria stated: "The resolution source will be a consensus of credible reporting." More than 40 global media headlines called it a suit. Yet UMA's oracle ruled there was no "consensus of credible reporting" and finalized the outcome as No.

The problem was that the market never defined what a "suit" meant. As menswear expert Derek Guy (who recently started a column in Bloomberg) told Wired: "It meets the technical definition [of a suit]... I would also recognize that most people would not think of that as a suit."

Replace “suit” with “invasion” and you have the same problem with the Venezuela market.

Should this be considered an invasion?

There are arguments on both sides.

The Case for an Invasion

Before the January 3rd raid, the U.S. had already imposed a naval blockade on Venezuela. UNGA Res 3314 explicitly lists this as an act of aggression. Under this definition, even a brief military occupation qualifies as aggression. The resolution doesn't require permanent territorial control, any military occupation, regardless of duration, counts. By this standard, the Venezuela operation, which involved armed forces entering sovereign territory and temporarily occupying parts of Caracas, appears to meet the definition of an "invasion or attack."

Trump said the U.S. would "run the country." He added that "U.S. oil companies would take charge of oil-rich Venezuela's struggling nationalized operations." And when pressed on how the U.S. would oversee Venezuela, Trump said an unspecified group was being formed that included Secretary of State Marco Rubio and Secretary of Defense Pete Hegseth to manage the country. This language directly suggests "establishing control,” which is an operative phrase in the contract.

The Brookings Institution noted that “[t]he closest parallel may be the invasion of Panama in 1989, when the United States captured the Panamanian dictator Manuel Noriega and put him on trial on U.S. soil."  Although Panama involved 27,000 troops who stayed for weeks and Venezuela involved a smaller force that withdrew within hours, both operations captured a head of state on drug charges, were framed as "law enforcement" domestically, and violated the UN Charter according to international legal consensus. Today it's universally called an invasion, including by official U.S. government sources.

Harvard Law professor Jack Goldsmith, who served in the Bush administration's Office of Legal Counsel, called the Venezuela operation an invasion in his analysis -- while noting the administration frames it as "law enforcement." And because different labels trigger different legal frameworks, the administration has strong incentives to avoid the word “invasion” regardless of whether it applies.

The Case Against

U.S. Army doctrine explicitly distinguishes between raids and operations to seize and hold terrain, defining a raid as a surprise attack as “for a specific purpose other than seizing and holding the terrain... [it] always ends with a withdrawal.”

The Caracas operation fits that category: forces entered, captured Maduro, and withdrew. The 1976 Israeli raid on Entebbe isn't called the "Israeli Invasion of Uganda." The 2011 operation that killed bin Laden isn't called an "invasion of Pakistan." 

Joint doctrine for "forcible entry" operations, by contrast, explicitly covers "seizing and holding a lodgment" in hostile territory. Under this doctrinal framework, the Venezuela operation looks more like a raid than an invasion:

• Forces entered with a specific objective (capture Maduro)

• They withdrew after accomplishing that objective

• No U.S. troops remain in Venezuela

• No territory is being held

Secretary Rubio explicitly told NBC: "…this was not an invasion. This is not an extended military operation. This was a very precise operation that involved a couple of hours of action." UN Ambassador Mike Waltz told the Security Council: "We are not occupying a country." Secretary of Defense Hegseth called the operation "a joint military and law enforcement raid." 

That said, there is a counterargument to be made that military doctrine serves operational purposes, not legal or semantic ones. The fact that the Pentagon calls something a "raid" doesn't mean it isn't also an "invasion" under other definitions. A raid can be a type of invasion. 

But when the resolution criteria require a “consensus of credible sources,” administration framing matters. And in terms of contract resolution it doesn’t matter if what occurred is objectively an invasion if there is not a consensus of credible reporting that it was one.

Is there a consensus of credible reporting?

Multiple governments used the word "invasion" or "aggression": 

• Brazil's PSDB party president Aécio Neves said it "repudiates the North American invasion against Venezuela"; 

• South Africa's government viewed it "as a manifest violation of the Charter of the United Nations" and referenced "military invasions against sovereign states";

• Russia's Foreign Ministry called it "an act of armed aggression against Venezuela";

• China's Foreign Ministry called it "blatant use of force against a sovereign state";

• Colombia's President Petro called it "aggression against the sovereignty of Venezuela and of Latin America"; and

• French politician Jean-Luc Mélenchon said, "There is no such thing as a good invasion, only bad ones."

However, mainstream English-language outlets described the event in softer terms:

• Reuters —> "military abduction"

• Washington Post —> "military raid"

• UK House of Commons Library —> "U.S. military raid in Caracas"

Add to this statements by the White House and other government agencies, a consensus around the word “invasion” did not materialize, at least not in the sources Polymarket would likely consider authoritative for resolution purposes.

What Traders Should Learn

The most effective resolution risk management is simply to avoid markets with ambiguous resolution criteria and to stick to ones with objective, binary, easily verifiable outcomes.

But if you decide to do so anyways, here are some things to keep in mind:

1. Read the resolution criteria, not the title. "Will the U.S. invade Venezuela by…" invites one interpretation, and a "[m]ilitary offensive intended to establish control" specifies another.

2. Identify the hinge words. Here they were "intended" and "establish control." In the Zelensky market, it was "suit." Pay attention to these terms.

3. Understand "consensus of credible sources." Model the language, not just the event. It’s possible for something to objectively happen and for there not to be a consensus of credible sources that the platform deems worthy.

4. As mentioned, avoid markets with ambiguous criteria. Stick to markets with clear outcomes that you can verify based on authoritative data sources.

5. Model the resolver, not reality. The question isn't "What do I think happened?" but "What would people applying these specific rules to these specific sources conclude?"

The Uncomfortable Bottom Line

Was this an "invasion" by some reasonable definition? 

Armed forces entered sovereign territory, killed at least 80 people, removed the head of state, and the president said the U.S. would "run" the country. A naval blockade -- itself an act of aggression under international law -- preceded the raid. UN experts called the operation a "grave, manifest and deliberate violation" of the UN Charter. Multiple governments used the word "invasion." International law's definition of "aggression" includes military operations "however temporary." The DOD's own manual describes occupation by "invading forces."

Did it satisfy this specific contract? 

The contract required "military offensive intended to establish territorial control" plus "consensus of credible sources" supporting that characterization. 

That consensus didn't materialize in U.S. media, one could argue in part because the administration itself had legal incentives to describe it differently, and partly because major outlets opted for softer terms like "raid" and "intervention."

Polymarket's resolution may be defensible under its stated rules. It may also feel like the rules changed mid-game, with clarifying criteria added after positions were taken, after events unfolded, and after many traders concluded the market should resolve to "Yes.”

Both things can be true.

When resolution criteria are clear, prediction markets can surface the truth. But when the contract relies on vague or contested terminology, we’re no longer predicting reality, we’re predicting whether reality aligns with someone else’s definition.

“If the number of measles in 2025 is above X, then the market resolves to Yes. Outcome verified from CDC…. If this event occurs, the market will close the following 10am ET. Otherwise, it closes by Dec 31, 2025 at 11:59pm EST.”

On its face, the question is straightforward: How many cases of measles will there be in 2025?

As of December 23, 2025, the CDC has reported that 2,012 cases of measles have been confirmed this year. Interestingly, the Johns Hopkins measles tracker currently reports that there have been 2,023 confirmed cases of measles as of December 26, 2025. 

And as of writing this, the odds for over 2050 confirmed cases stands at only 63%. But why?

The Market in Perspective

The Kalshi measles market for 2025 started at 100 cases. Until this year, this was a pretty reasonable ceiling for a measles market. In 2024, the US had 285 confirmed cases, the year before that 59. And before 2010 cresting 100 was an anomaly, instead of the norm for almost a decade.

Since 2000, the measles virus has been considered “eliminated” in the United States. What fewer readers may know is, two years earlier in 1998, measles was considered “eliminated” in Canada. (Elimination is defined as an interruption of endemic measles transmission for greater than or equal to 12 months.) As of November 10, 2025, Canada’s measles elimination status was revoked after almost 27 years.

Obviously, this is not great.

So, what happened?

A Measles Explosion

South Carolina

Between December 5 and December 9, there were 27 new confirmed cases of measles in South Carolina, meaning that, on average, between those dates, approximately 6.75 people contracted confirmed cases of measles per day.

As of December 10, news sites began to cover an “accelerating” outbreak in South Carolina, with hundreds of people reported to be in quarantine at that time. The outbreak itself appears to have originated from a single church in Inman (Way of the Truth Church) and multiple schools in Spartanburg County.

As of December 10, there were 111 confirmed cases of measles in South Carolina since July 9, 2025. Of those 111:

• 105 people were unvaccinated, while three people were only partially vaccinated;

• 254 people had been placed in quarantine; and

• 16 were currently in isolation.

Regarding demographics, of the 111 cases:

• 20 were under the age of 5;

• 75 were between the ages of 5-17;

• 10 were over the age of 18; and

• 6 minors under the age of 18 were reported with confirmed cases of measles.

On December 12, reported cases had risen by 15 more people (since the previous Tuesday, 12/9/25) bringing the total up to 126, with 303 people quarantined and 13 in isolation.

Utah

Meanwhile, in Utah, as of December 23, 2025, there have been 142 cases. Of the 142 cases, 102 occurred in Southwest Utah, 134 were in people who had not received the measles vaccine, and 85 were under 18.

The consistent thread between almost all of the outbreaks has been that the main victims have primarily been unvaccinated children.

Arizona

The measles outbreak in Mohave County, Arizona (which as of 2024, was reported to have a 78.4% vaccination rate for the MMR vaccine), appeared momentarily at the start of December to potentially be puttering out. However, as of December 29, 2025, the website had updated, reporting 195 confirmed cases in Arizona, with new cases being actively confirmed on a weekly basis.

Texas

Earlier this year, in Texas, there were 762 confirmed cases of measles reported by Texas’s HHS. Thankfully, as of August 18, 2025, Texas’s DSHS announced that the West Texas measles outbreak had officially ended. What we also learned from this outbreak was that most of the cases were clustered in low-vaccination areas within Texas. We also know that it’s likely that the measles virus essentially crept up from Texas and across New Mexico, then through Arizona, and then finally made its way to Utah, since August.

Predicting the Spread

According to CDC guidelines, for there to be effective herd immunity from measles via vaccination, 95% of a population must be vaccinated against the measles virus. Currently, the estimated percent of South Carolina residents who are vaccinated is at 92.1%; the estimated percent of vaccinated Arizona residents is 89.3%; and the estimated percent of Utah residents who are vaccinated is 88.8%. 

However, it’s important to keep in mind that these percentages do not represent an even spread across the state. Instead, you’ll find large communities with over 95% vaccination rates and higher, next to extremely small communities with vaccination rates that in some cases may be almost 0%. 

When epidemiologists attempt to predict the proportion of an at-risk population that is likely to contract a pathogen within a specific time frame, they calculate something known as an “attack rate.”

The attack rate is defined as:

Attack Rate = Number of new cases /Total exposed at-risk population

(Note: An “at-risk” person is defined as a person with no immunity against the pathogen.)

The attack rate for measles is 90%. This means that there is a 9 in 10 chance that you’ll get sick with measles if you’re unvaccinated or no longer have the titers against the measles virus and are exposed to measles.

Next, measles has an R₀ of ~12-18. R₀ is another number used in epidemiology, which is used to express the number of people a single sick person will infect on average. 

• R₀ < 1 typically indicates that the pathogen is likely to die out.

• R₀ = 1 indicates that the disease is endemic to an area or “Stable.”

• R₀ > 1 implies that the disease is likely to spread in the population.

To put things into perspective, the flu has an R₀ ranging from 0.9 to 2.1. 

A person with measles is contagious approximately four days before they begin to show a rash. Additionally, measles has been reported to remain infectious for approximately 2 hours in enclosed spaces without high-efficiency airborne-contamination removal systems, i.e., most places you normally go in public. 

This depressingly growing number of cases of human tragedy, and the circumstances of their creation, ironically, also creates a dataset, which should make the math for trading on this market somewhat trivial. However… 

While we’ve been tracking this outbreak for weeks and our model has consistently predicted the reported growth in confirmed measles cases, we also noticed something important which led to us holding off on sharing our findings earlier. 

Even though the title of the Kalshi market would suggest that, by accurately predicting the number of confirmed measles cases in 2025, a trader could successfully anticipate the growth of these cases and therefore make informed trades, there are three large hiccups. 

1. The rules of the market state that: The Last Trading Date of the Contract will be the sooner of the date of the first 10:00 AM ET following the occurrence of an event that is encompassed in the Payout Criterion or the last day of <year>. The Last Trading Time will be 11:59 PM ET;

2. The CDC website does not update in real time or even on a regular basis. In fact, it will often also add cases that occurred in the past, several weeks later; and

3. The measles virus takes approximately 7-14 days to incubate before an infected person begins to show signs of infection.

This means that it actually does not matter what the true number of confirmed cases for 2025 will be. Instead, the outcome of the market is dependent on whether the CDC website updates with an accurate number for confirmed cases in 2025 in time for the market’s resolution.

This turns what should’ve been a guarantee that there will be over 2,050 confirmed cases of measles this year, into a near coin flip. Because the true number of confirmed cases for 2025 will not be able to be ascertained until after the Kalshi measles market has already resolved, and because it’s not clear that even if there was a huge surge of confirmed cases between December 23rd and December 31st that the CDC website will update its results in time.

With all that said, Happy New Year folks. Stay vaccinated and get your titers checked, because 2026 is going to be a great year for measles.

As you probably know if you’re reading this, Polymarket allows you to trade on the outcomes of future events. Prices between $0.00 and $1.00 reflect collective forecasts. Once an event concludes, the market must be reconciled or resolved to a binary “yes” or “no.” The outcome of this determines which traders profit and which lose.

Trust in resolution is essential. Obviously, you don't want bad actors manipulating outcomes in their favor. Because Polymarket is a decentralized protocol, they don’t rely on a central admin, like Commodus here, to decree the outcome.

Instead of someone like Commodus, Polymarket relies on rules, bonds, and third-party arbitration to align incentives.

How Polymarket Resolution Works

For a market to be resolved, an outcome must be “proposed.” To do this, the proposer submits an outcome and puts up a bond of (usually) 750 USDC, which acts as a bounty for anyone who can prove the assertion wrong.

After a market is proposed for resolution, it goes into a challenge period for two hours. If the proposal is valid and verified the proposer will receive a reward. If someone does not agree with a proposed resolution (say you don’t think Maximus should be sentenced to death) that proposal can be disputed. 

If someone disagrees with a proposed outcome, they can place a dispute (along with a 750 USDC challenge bond) through the UMA dapp. UMA, or Universal Market Access, describes itself as “a decentralized truth machine,” an Optimistic Oracle (OO) that records verifiable truth onto a blockchain. The oracle is considered optimistic because it assumes honesty until challenged, and the answer stands if nobody contends the proposal. 

When a dispute is placed, the case escalates to UMA’s Data Verification Mechanism, or DVM. You can think of UMA’s DVM as effectively a decentralized court.

This begins a 24-48 hour debate or evidence period. During this time, participants present sources and dispute evidence in the UMA Discord server. After the debate period, UMA token holders vote. And staked $UMA voters commit their votes in secret; this is intended to reduce collusion by preventing voters from simply copying the majority as the votes come in. 

The intention is to guide voters to disregard their personal opinions and to encourage them to predict what other rational, rule-observing voters will conclude. Voters are rewarded if they vote with the (eventual) majority, and those who are inactive or vote with the minority are "slashed,” losing a portion of their staked $UMA tokens.

Once the results are in and the voting period is completed, there are four possible outcomes:

1. The proposer wins and receives their bond back plus half the disputer’s bond as a bounty. The disputer loses their bond.

2. The disputer wins and receives their bond back plus half the proposer’s bond as a bounty. The proposer loses their bond.

3. It is deemed too early. This outcome is for when the underlying event has not happened yet. In this case, the disputer receives their bond back plus half the prospers bond as bounty, and proposer loses their bond.  

4. The outcome is unknown (50/50). This is rare, and is for events where none of the other options are appropriate. In this case the market price resolves to 50% yes and 50% no. The disputer receives their bond back plus half the proposer’s bond as a bounty, and the proposer loses their bond.

When the System Breaks

The trust in this process holds as long as the Cost of Corruption (CoC) exceeds the Profit from Corruption (PfC). When participants stand to lose money, they’re motivated to find fault. If they believe the rules were misapplied (and there is evidence to support them), they can dispute.

This system appears to work most of the time.

Although, in theory, someone could buy enough $UMA rig a vote, doing so would destroy the token’s value, making the attack self-defeating. The system, however, is not perfect. Issues can arise when a single market’s payoff is greater than the honest voting base. There are a few notable cases where this happened:

• March 2025 — “Ukraine mineral deal” market. A high‑stakes market on whether Ukraine would agree to a Trump‑brokered mineral deal before April was resolved “Yes” despite no verified agreement. Reporting traced the outcome to a concentrated $UMA holder (“whale”) whose voting power with “as much as five million governance tokens” swung a premature DVM result. Honest voters who chose the factual outcome were slashed, which is an inversion of the intended incentives. Polymarket said they would take steps and build systems to make sure this didn’t happen again.

• July 2025 — “Zelensky suit” controversy. A market on whether Zelensky wore a suit before July turned into a lightning‑rod dispute over definitions and voter concentration. “Suitgate” exposed the consequences of when subjective criteria smashes into whale‑dominated voting. You can read UMA’s debrief of the situation here.

Until recently, Polymarket only relied on UMA for resolutions. However, that is starting to change.

In September 2025, Polymarket quietly began to migrate a portion of its resolution stack to Chainlink’s Data Streams and Automation, particularly for price-based markets such as the 15 minute “BTC Up or Down” and similar markets. Polymarket continues to use UMA for subjective or rules‑heavy questions where human interpretation matters.

With the rumored $POLY token set to introduce in-house oracle governance, Polymarket appears to be moving toward its own “truth layer.” While this would bring market resolution under native, token-weighted governance, it also raises new questions around decentralization, incentives, and regulatory risk.

Did you know you can still take positions on disputed contracts?

An interesting bit about UMA is that you can still buy contracts while a market is in dispute, and the public conversations on the UMA discord can provide additional insight into how the dispute might go. (Please note, you do not have to be voting to be in the conversation. And based on our own research the conversations do not seem to be very lively or involved.)

How to read a dispute before it resolves

Here’s an approach you can use to try to gain insight into which way a dispute might swing:

• Start with the rules. Read the ancillary data line by line. Definitions, time windows, and exclusions usually decide the case. Disregard what traders "assumed" if it isn't written in the rules.

• Write the test. If possible, in one sentence, state the criterion as a boolean: Resolve Yes iff X occurs by T as reported by S.

• List admissible sources. Use only what the rules authorize (official sites, specific APIs, named outlets). Do some sources have priority over others?

• Track the arguments. Follow proposer/disputer summaries and community analysis in public channels, such as the UMA Discord or X. This does not mean any of this is necessarily an admissible source, but it might give you insight into how people are thinking. 

• Watch the bond & timing. Rising bond sizes, repeat challenges, or “too‑early” flags might signal uncertainty.

• Anticipate edge cases. Words like exclusively, consensus, recognition, suit often carry the dispute.

A mental model you might want to try: Ask, What would a rules‑maximalist majority choose using only the allowed sources?

By applying these guidelines, you can approach disputed markets more deliberately, using a clearer grasp of Polymarket’s resolution mechanics to look for, rather than assume, potential edge.