Game Of The Year 2026 Odds: GTA 6, Resident Evil Requiem, And The AI Model Verdict

The 2026 Game of the Year race looks, at first glance, like the least suspenseful award on the calendar: Grand Theft Auto VI is coming, and it is Grand Theft Auto VI. Kalshi has turned the whole field into separate yes/no contracts, and traders make Rockstar's game a runaway favorite. I asked six AI models to score that field without letting any of them see a single price. They agreed GTA 6 is the favorite, and then, unlike the crowd, they backed off it, because the entire bet rests on one calendar date that has already moved twice.

The Quick Answer

Kalshi traders price Grand Theft Auto VI near 72 cents, about a 62% chance once you strip out the field's overround. Our price-blind, six-model panel is more cautious: it blends GTA 6 to 56.9%, a real fade below the market. The reason is not a rival out-playing it; it is the eligibility gate. GTA 6 is set for November 19, 2026, roughly one day inside the Game Awards' historical cutoff, so a third short delay would knock it out of the 2026 race entirely. If that happens, the panel's clear pick is Resident Evil Requiem, already shipped, already reviewed, and the board's most underpriced name.

The panel's read in one line: the market is pricing GTA 6 as if the only question is whether it wins. The models think the harder question is whether it arrives, and they are charging the crowd for ignoring it.

• Market Favorite: Grand Theft Auto VI, ~72¢ (about 62% once the field's overround is stripped out)
• Model Favorite: Grand Theft Auto VI too, but only 56.9%, roughly 5 points below the de-vigged market and 15 below the raw price
• The Underpriced Backup: Resident Evil Requiem, a 15.5% blend versus a de-vigged ~8.6%, the one shipped, high-scoring game that inherits the award if GTA 6 falls
• The Top Unreleased Challenger: Marvel's Wolverine, 9.1%, the Insomniac prestige title with a full review runway before December
• The Dark Horse The Panel Won't Dismiss: Onimusha: Way of the Sword at 3.0% versus a 1-cent price, a live longshot the market has written off
• The Honest Caveat: a shipped, well-reviewed GTA 6 is still the most dominant favorite on any board we have scored; the fade is about risk the market is under-charging, not a prediction it loses.

The full field, the case for and against the three names that carry the story, the dark horses underneath them, and a plain-English guide to reading these prices are all below.

New to event-contract markets? Start with our explainer on how to bet sports on Kalshi, this piece assumes you already know how a yes/no contract settles.

The Market At A Glance

Venue	Kalshi, a CFTC-regulated event-contract exchange (18+; availability varies by state, as of August 2026)
The Event	"2026 Game of the Year?", a separate yes/no contract per eligible game (Kalshi event KXGAMEAWARDS-2026)
The Contract	Each market resolves YES only if that game wins Game of the Year at the 2026 Game Awards, per Kalshi's settlement rules; the market closes following the show
Show Date	The Game Awards 2026 airs December 10, 2026; nominees are announced mid-November
Prices Below As Of	August 3, 2026 (prices on this page refresh during the news cycle; the "as of" stamp carries the evidence)

These are model estimates, not predictions of fact and not financial advice. Kalshi event contracts trade on a CFTC-regulated exchange; you must be 18+ and in an eligible state to participate.

What Actually Decides This Award

Before any game's name, two gates decide almost every Game of the Year outcome, and they explain why the panel built the board the way it did:

• Gate One: Release And Eligibility. A game cannot win an award it is not eligible for. The Game Awards requires a title to ship inside the eligibility window, historically around the third Friday of November, roughly November 20, 2026 for this cycle. That single rule is the entire GTA 6 story, and it is why the models refuse to price the game as a near-certainty.
• Gate Two: The Jury's Taste. Game of the Year is voted by a jury of critics and outlets that has, for years, rewarded big, critically acclaimed, single-player narrative games. Multiplayer-first and niche-genre titles almost never win, no matter their sales. That prior is why a shipped Resident Evil Requiem sits well above a bigger-selling multiplayer game on this board.

There is one more truth the panel kept circling, and it is the crux of the fade: being the favorite is not the same as being a lock. Red Dead Redemption 2 scored 97 on Metacritic, tied among the highest-rated games ever made, and still lost Game of the Year to God of War at the 2018 Game Awards. Rockstar at the absolute peak of its craft has been passed over before. A shipped GTA 6 is the heaviest favorite gaming can offer; it is not a coronation.

Here is the field, sorted by the model blend, with each model's raw number and the gap against the market price.

Game	Studio	Market	Opus	Sonnet	Fable	GLM	Kimi	DeepSeek	Model blend
Grand Theft Auto VI	Rockstar	72¢	54	54	54	61	55	60	56.9%
Resident Evil Requiem	Capcom	10¢	16	16	15	14	16	15	15.5%
Marvel's Wolverine	Insomniac	7¢	9	9	9	8	10	9	9.1%
Control Resonant	Remedy	6¢	6	7	6	6	7	7	6.6%
Crimson Desert	Pearl Abyss	8¢	3	4	3	3	3	3	3.2%
Onimusha: Way Of The Sword	Capcom	1¢	3	3	2	3	3	4	3.0%
The Duskbloods	FromSoftware	1¢	2	2	2	2	2	2	2.0%
The Eleven Other Games (Pragmata, Fable, Saros, Marathon, Stranger Than Heaven, Slay The Spire 2, Cairn, Phantom Blade Zero, Esoteric Ebb, Half-Life 3, Tie)	various	1¢ Each							≤1% each

These are model estimates, not predictions of fact and not financial advice. Market figures are raw YES quotes from Kalshi and include the exchange's bid/ask spread, so the listed prices sum to about 116 cents, over 100%. The de-vigged comparison is worked through below.

A word on that last row. No model gave any of the eleven bottom games more than noise. Some are shipped titles without a Game of the Year-tier critical breakout (Pragmata, Saros, Marathon); some are 2026-window titles with no confirmed release or reviews yet (Fable, Stranger Than Heaven, Phantom Blade Zero); Half-Life 3 has never been officially announced by Valve at all, so it cannot be nominated; and "Tie" resolves only if the show declares co-winners, which has never happened. Rather than print eleven near-identical floor rows, we group them honestly. Nothing is dropped from the field, every eligible contract is priced, these are simply the games the models placed at the floor.

A note on the panel. Our standard Model Verdict lineup runs eight seats. On this run two were unavailable: the ChatGPT and Gemini seats were not installed on the machine that produced these verdicts. This board therefore runs on six models: Claude Opus, Claude Sonnet, Claude Fable, GLM, Kimi, and DeepSeek. All six scored the field on August 3, 2026, without seeing any prices, then read one another's anonymized reasoning in a second round and were free to revise. Every model's number here also lands in our back-tested Model Verdict scoreboard, where past price-blind calls are graded against settled results so the method is measured over time, not just asserted.

The one number that matters most: the panel's 56.9% on GTA 6 sits below the market's de-vigged 62%. Strip the exchange margin out and the models are still fading the favorite, the rare board where our panel is more skeptical of the crowd's top pick than the crowd is.

With the field in view, three names carry the story: the favorite the models back but fade, the shipped game they think the market has underpriced, and the unreleased challenger with the pedigree to matter. Here is the case for and against each.

Grand Theft Auto VI: The Favorite The Panel Fades

If the panel has a center of gravity, it is still Grand Theft Auto VI at 56.9%, every seat named it the clear favorite, landing between 54% and 61%. A GTA game reviewing anywhere near Rockstar's historical level, released on time, is about as close to a default Game of the Year winner as the medium produces. Grand Theft Auto V and Red Dead Redemption 2 both scored 97 on Metacritic; the jury's big-narrative taste points squarely at this kind of game.

So why fade it? Because the bet is really two bets stacked on top of each other. First, GTA 6 has to ship. It has already been delayed twice, from a 2025 window to May 2026, then to November 19, 2026, and that November 19 date sits roughly one day inside the eligibility cutoff. A third slip of even a few days does not cost GTA 6 points; it removes the game from the 2026 award completely. Take-Two has publicly ruled out further delays and pre-orders have been live since June 25, which the panel treats as real but not decisive, locked dates and live pre-orders are exactly what precede some delays. Second, if it ships, it still has to win, and the Red Dead Redemption 2 precedent shows a 97-Metacritic Rockstar epic is no lock for the trophy. The seats put the on-time-ship probability near 78-80% and the win-if-eligible probability near 70-75%; multiply those and you land in the mid-to-high 50s, not the low 70s the market implies.

The case against the fade: if GTA 6 makes its date and reviews as expected, none of this matters and the market's 72 is closer to right than the panel's 57. The fade is a bet on tail risk the crowd is under-pricing, not a prediction that Rockstar loses a head-to-head.

What would flip this board: a single Rockstar or Take-Two statement, official or leaked, of another delay. Even a two-week slip past the cutoff converts GTA 6's 57 to roughly zero and roughly triples the entire shipped-contender tier overnight. It is the one piece of news that most changes every number here.

Resident Evil Requiem: The Underpriced Backup

Every model named the same inheritor, and it is the one game on the board that has already done the hard part. Resident Evil Requiem blends to 15.5%, well above its de-vigged market price of roughly 8.6%, the largest below-the-radar gap on the board.

The case is that it is finished, and it is excellent. Requiem released February 27, 2026, scored 88 on Metacritic as the highest-rated game of the year so far, the best-reviewed mainline Resident Evil since RE4 in 2005, and sold more than 7 million copies as the fastest-selling entry in series history. It fits the jury's big, single-player, critically acclaimed profile precisely. Crucially, it carries no release risk at all, the trait that makes GTA 6 a gamble is exactly the trait Requiem has already retired. If the favorite slips out of eligibility, the jury's prior needs a shipped, acclaimed, high-profile target, and Requiem is the only field member that clearly fits today. That is why the panel prices it as the clean beneficiary of every scenario where GTA 6 falls, and why it sits above the two unreleased September prestige titles beneath it.

The case against: an 88 is a very good score, not a historic one, and Requiem's own reviews flagged pacing and narrative issues. In a year where GTA 6 ships and reviews in the mid-90s, an 88-rated horror game is a clear runner-up, not a winner, which is exactly why the panel keeps it in the mid-teens rather than pushing it toward the favorite.

Marvel's Wolverine: The Unreleased Challenger With Pedigree

The strongest not-yet-released name is Marvel's Wolverine, which blends to 9.1%. It is the kind of game the jury tends to reward: a single-player, narrative-driven action-adventure from Insomniac Games, whose Marvel's Spider-Man 2 was a 2023 Game of the Year nominee.

Its advantage over GTA 6 is timing in reverse: releasing September 15, 2026, it clears the eligibility window comfortably and gives critics and jurors a full review cycle before December. Its disadvantage is that it is entirely unreviewed, the panel is pricing pedigree and release runway, not a known quantity. A 90-plus Metacritic once its embargo lifts would meaningfully compress GTA 6's cushion; a merely-good reception leaves it a nominee rather than a winner. The models keep it in high single digits precisely because both outcomes are live and the evidence to separate them does not exist yet.

Dark Horses The Panel Won't Dismiss

The story is not only the top three. A field board rewards looking underneath the favorites, and a few games drew real numbers from at least one seat despite trading at the floor:

• Control Resonant (6.6% Blend, 6¢). The sequel to Remedy's multi-award-winning, Game of the Year-nominated Control, out September 24, 2026. Every seat gave it a real number; Remedy's prestige pedigree and a clean review runway make it the fourth name the panel keeps truly alive, a breakout review away from the Wolverine tier.
• Onimusha: Way Of The Sword (3.0% Blend Versus A 1¢ Price). Capcom's first new mainline Onimusha in more than 20 years, moved up to September 4, 2026. DeepSeek and the Claude seats all gave it triple the market's implied chance, a live, jury-friendly revival the crowd has written off entirely.
• The Duskbloods (2.0% Blend, 1¢). A FromSoftware title in the 2026 window; the studio's name alone earns it more than noise, though an unconfirmed date and multiplayer framing cap it. If FromSoftware ships and reviews at its usual level, this is the longshot most likely to embarrass the floor.

None of these is a prediction of a win. They are the places the panel thinks a 1-cent price is too cheap for the studio behind the game, the honest reason a field is worth reading past its top line.

The Convergence Round: One Model Changed Its Mind

These numbers came in two passes. Each model scored the field blind, then read the other seats' anonymized reasoning and was free to revise. Most held. One moved sharply, and the reason is instructive.

In the first round, DeepSeek was the board's outlier, pricing GTA 6 at 75%, essentially treating a shipped game as a near-lock. After reading the other seats' reasoning, it revised down to 60%, writing that a 97-Metacritic game is "slightly discounted from a lock because RDR2 lost to God of War in 2018." A 97 Metacritic score, it reasoned, does not settle the award. That is the convergence process working as intended: not conformity, but a specific, checkable precedent moving a number. The rest of the panel had weighted that same 2018 result from the start, which is why the group landed in the mid-50s rather than the low 70s the market implies.

How To Read A Kalshi Field Price Like This

The models can tell you where they disagree with the crowd, but the number in front of you still has to be worth understanding. Here is how to read a field price like the ones above, plus the honest friction that comes with a market that will not settle for months.

Work an example. GTA 6 shows a 72-cent YES price. That is not a clean "72% chance." On Kalshi the YES and NO quotes carry a spread, and across a mutually exclusive field the individual YES prices collectively add up to more than 100 cents, an overround baked into the quotes. Here the contracts sum to about 116 cents of implied probability. To get closer to a fair read, compare each name's price to that total and re-scale: GTA 6 at 72 cents is really about 62% once you strip the overround out (72 divided by 116). That is why the panel's 56.9% blend reads as the models leaning below the market rather than merely disagreeing at the margin, and why Resident Evil Requiem's 15.5% against a de-vigged 8.6% is the board's real value flag.

Game	Raw YES price	De-vigged (÷1.16)	Model blend
Grand Theft Auto VI	72¢	62.1%	56.9%
Resident Evil Requiem	10¢	8.6%	15.5%
Marvel's Wolverine	7¢	6.0%	9.1%
Crimson Desert	8¢	6.9%	3.2%

De-vigged column re-scales each raw price by the field's ~116¢ total to strip the overround. These are model estimates, not predictions of fact and not financial advice.

That habit of reading a fair, vig-free version of every number is line shopping applied to a field board: checking what a spread-free version of a price looks like before you trust it, the same way you would convert Kalshi's cents into real odds for comparison. OddsShopper's live odds screen and free expert picks today do that price-comparison work for weekly game slates, while a season-long award board like this one is priced on Kalshi itself. If you want the same price-blind treatment applied to a different awards field, our 2026 Ballon d'Or model verdict scores that race the same way. A few honest points before anyone acts on a number above:

• Eligibility. Kalshi is a CFTC-regulated event-contract exchange, not a sportsbook. You must be 18+ and located in an eligible state, and availability varies by state and can change.
• Liquidity And Spread. The tail games on this board are thin. Many of these contracts show a wide bid/ask gap, so the "price" on a 1-cent game is softer than a single number suggests.
• Time And Capital. This market does not settle until after the December 2026 show, so any position ties up money for months with no interim payout.
• Not Advice. Everything here is analysis for information, not a recommendation to trade. If you want the fuller comparison of how these exchanges stack up against traditional books, see our Kalshi vs. sportsbooks breakdown. These are model estimates, not predictions of fact and not financial advice. Every number in this piece gets graded in public once the market settles, you can check our past calls on the full Model Verdict scoreboard.

FAQ

What are the odds GTA 6 wins Game of the Year 2026? Kalshi traders price Grand Theft Auto VI near 72 cents, roughly a 62% chance once the field's overround is stripped out. Our six-model AI panel is more cautious, blending GTA 6 to 56.9% because a third release delay of even a few days would push it past the Game Awards eligibility cutoff, and because a shipped GTA 6 is a heavy favorite but not a lock, Red Dead Redemption 2 scored 97 on Metacritic and still lost Game of the Year to God of War in 2018.

When is The Game Awards 2026 and why does GTA 6's release date matter so much? The Game Awards 2026 airs December 10, 2026. A game must release inside the eligibility window, historically around the third Friday of November (about November 20). GTA 6 is currently set for November 19, 2026, roughly one day inside that window, so even a short additional delay would make it ineligible for the 2026 award entirely.

Which game does the AI panel back if GTA 6 slips or loses? Resident Evil Requiem. It released February 27, 2026, scored 88 on Metacritic as the highest-rated game of the year so far, and sold more than 7 million copies as the fastest-selling entry in the series. The panel blends it to 15.5% against a de-vigged market price near 8.6%, making it the board's most underpriced name and the clear inheritor if the favorite falls.

How does the Kalshi Game of the Year market settle? Kalshi lists a separate yes/no contract for each eligible game (event KXGAMEAWARDS-2026). Each resolves YES only if that game wins Game of the Year at the 2026 Game Awards, and the market closes following the show. The nominee slate itself is not announced until mid-November 2026, so every price here is on the full eligible field.

How were these model estimates produced? Six AI models scored the entire field without seeing any market prices, working only from each game's release status and critical reception fetched at runtime, then read each other's anonymized reasoning in a second round and revised. Their per-game numbers were blended into one estimate per game. These are model estimates, not predictions of fact.

The Bottom Line

The market and the models agree on the shape of the 2026 Game of the Year race: GTA 6 is the favorite, and it is not close. Where the panel parts ways with the crowd is on how much that favorite is worth. At 72 cents the market prices GTA 6 as if arrival is a formality; the models, weighing a twice-delayed game landing one day inside its eligibility window and a jury that has passed over a 97-Metacritic Rockstar epic before, fade it to the mid-50s and hand the difference to Resident Evil Requiem, the one contender that has already shipped, already reviewed at the top of the year, and carries no calendar risk at all. Read it against the caveat the models kept: a GTA 6 that makes its date and reviews as expected still wins going away. The fade is a bet on risk the market is under-charging, not a call that Rockstar loses. If this method is your thing, our 2026 Ballon d'Or model verdict and the full graded scoreboard score every field the same price-blind way.

Hottest Prediction Markets Right Now

• 2028 Democratic presidential nominee · $166M traded
• Rocket Classic Winner · $124M traded
• 2027 Pro Football Champion · $51M traded
• Pro Baseball Champion · $40M traded
• Bitcoin price at the end of 2026 · $30M traded
• 2027 Pro Basketball Champion · $19M traded

Every market above links to our full AI model verdict; browse them all on the OddsShopper prediction markets hub, and see how every settled call actually scored on the full graded scoreboard.

Verdicts generated August 3, 2026, over two rounds. Prices as of August 3, 2026. Panel: Claude Opus, Claude Sonnet, Claude Fable, GLM, Kimi, and DeepSeek, price-blind. These are model estimates, not predictions of fact and not financial advice.