How the contract works
A contract on this question settles at $1 if Google holds the #1 position on the specified LMArena leaderboard at the check time, and at nothing if any other company does. The check happens once, at 12:00 PM ET on 30 November 2026, using the Arena.ai Text Arena Overall leaderboard with style control off, filtered for Models and excluding AutoEval-tagged entries; ties are broken first by the underlying unrounded Arena score, then alphabetically by company name. The market price at any moment is simply what buyers and sellers currently think the chance of a Google #1 finish is โ a price of 0.30, for example, would imply the market sees roughly a three-in-ten chance, not this market's actual level. A position taken now can typically be sold before the 30 November check, at whatever price the market has moved to by then.
Yes49%
No51%
Probability
How the price has moved
The only figure available is the current consensus of 49%, from a single tracked venue, Polymarket, with $80,204 in total volume. There is no broader price history given here to describe a trend, but the level itself is informative: a reading this close to 50% says the market is not leaning toward either a Google or a non-Google outcome, and with volume this limited the price likely remains sensitive to new information, particularly any model launch announcement from a major lab.
Analysis
Context
LMArena (formerly Chatbot Arena) ranks AI language models by having users compare two anonymous model outputs side by side and vote for the better one. Those votes feed an Elo-style score, and the resulting leaderboard is widely cited in tech coverage as a rough proxy for which lab currently has the strongest general-purpose model. Unlike fixed benchmark tests, the ranking can shift within days of a new model launch, because votes accumulate fast once a model is added to the arena.
The contest for the top spot has been genuinely competitive. Google's Gemini line, OpenAI's GPT models, Anthropic's Claude, xAI's Grok and, at times, Chinese labs such as DeepSeek have each held or challenged for #1 at different points as new versions released. Each lab treats the leaderboard as a marketing and credibility signal, which means releases are often timed to coincide with periods when a competitor's model might otherwise be on top.
This contract asks specifically about the state of that leaderboard at a fixed moment: 12:00 PM ET on 30 November 2026, using the no-style-control, Models-filtered, AutoEval-excluded view of the Arena.ai Text Arena Overall board.
The consensus price sits at 49%, pulled from a single tracked venue, Polymarket, with total volume of $80,204. A reading this close to the midpoint says the market genuinely does not have a strong lean either way โ it is not pricing Google as the favorite or as the underdog, but treating a Google #1 finish and a non-Google #1 finish as roughly equally likely outcomes as of now.
That 49% figure also has to be read against how thin this market is. With only one venue tracked and $80,204 in total volume, there isn't a large pool of participants setting the price, and it can move more on a handful of trades than a deeper market would. That matters for interpretation: a 49% price here reflects the judgment of a relatively small set of traders rather than a broad consensus, so it should be read as a rough gauge of sentiment rather than a precise forecast.
The underlying mechanics of the leaderboard help explain why the market sits near even money rather than leaning hard toward Google or away from it. LMArena rankings are sensitive to recency โ a lab that releases a strong new model shortly before the check date has an advantage, because votes accumulate quickly once a model is live in the arena. Google, OpenAI and Anthropic have each demonstrated the ability to ship a flagship model on short notice, and any of them landing a release in the weeks before 30 November could flip the top spot. Because none of the facts available here point to a confirmed, scheduled release from any one lab timed specifically around that date, the market's near-50% price looks like an honest reflection of genuine uncertainty rather than a mispricing.
The resolution rules themselves add a layer of precision worth noting: the check excludes AutoEval-tagged models and uses the no-style-control view, and ties go to the underlying unrounded score before alphabetical order. That detail matters because it removes some of the noise that style-control adjustments or experimental-tagged models might otherwise introduce into who is counted as #1.
What moves the probability
Release timing before the checkpoint
Whichever major lab ships a materially stronger flagship model closest to 30 November has an advantage, because LMArena scores update quickly once a model accumulates votes. This is the single largest swing factor and can move the probability sharply within days of an announcement.
Google's release cadence
Google has shipped Gemini updates at a steady pace through 2025 and 2026 and has held the LMArena top spot at various points in the past, which supports the case that another strong release is plausible before the check date. This pushes the probability up when Google is actively iterating, but only if the timing lines up.
Competing labs' own roadmaps
OpenAI, Anthropic, xAI and other labs are capable of releasing a model that displaces any current leader, and have done so repeatedly across 2025 and 2026. Every credible rival release before the checkpoint pushes the probability of a Google #1 finish down.
Methodology specifics of the check
The resolution uses the no-style-control view, excludes AutoEval-tagged models, and filters for Models only, which changes which entries even qualify for the #1 slot compared with other leaderboard views. This narrows the field in ways that can favor models tuned differently than they would be under style-adjusted rankings.
The case for
- Google would need to hold, or regain, the #1 slot on the specified no-style-control Arena.ai leaderboard at the 12:00 PM ET check on 30 November 2026.
- A new Gemini release landing in the weeks before the check, strong enough to out-score current leaders on the unrounded Arena score, would secure the top spot under the stated tie-break rules.
- Google has previously held the top LMArena position during 2025 and 2026, showing the lab has the technical capacity to compete for #1 when it ships a new flagship model.
The case against
- Any other major lab โ OpenAI, Anthropic, xAI, DeepSeek or another company โ could release a stronger model before 30 November and claim the top spot instead.
- The leaderboard has changed hands repeatedly over the past two years, which argues against assuming any one lab's current position, or lack of it, will hold through the checkpoint.
- The near-50% consensus price itself signals that market participants see no clear favorite, which is itself evidence against confidently expecting a Google finish.
What to watch
The fixed event is the check at 12:00 PM ET on 30 November 2026, when the Arena.ai Text Arena Overall leaderboard (style control off, Models filter, AutoEval excluded) is read to determine the #1 entry. Between now and then, watch for any flagship model announcements from Google, OpenAI, Anthropic, xAI or DeepSeek, since a release landing close to the checkpoint is the most direct way the ranking could shift. Also watch the leaderboard itself in the days leading up to 30 November, since LMArena rankings can change as voting accumulates on newly added models.
Trade this contract
Open on Polymarket
- gas covered