How accurate is Grok 4.3?
We put Grok (currently Grok 4.3) on real games and grade every call in public. It gets the identical line-blind data packet — no betting line, no web search — locks its forecast, and is scored against the final result. Graded since June 2026.
Frozen, no longer running. The current Grok flagship is Grok 4.5.
MLB · picking winners · 168 games
MLB · run totals (over/under) · 163 games
UFC · fights · 13 graded
Head-to-head
Over 168 graded MLB games, Grok 4.3 is 92–76 (55% winner accuracy), with a Brier score of 0.246. The naive "pick the home team" baseline over the same games lands at 50%.
Scored as a 1-unit bet at the closing market price, its picks have returned -1.3% per game — the honest, vig-included number.
And it's not only calling winners. On run totals it's 57% against the over/under — each market graded on its own.
On UFC, Grok 4.3 is 11–2 (85% accuracy), and calls the method of victory — KO/TKO, submission or decision — right 46% of the time.
Grok 4.3's calls, graded
Best MLB calls · confident & right
Best UFC calls · confident & right
Every call links to the game so you can check it. "Confident" = the probability the model put on its own pick. No cherry-picking — these are simply its boldest right and wrong calls on record.
How Grok is graded
One prompt, frozen. Every model — Grok and the rest of the field — gets the exact same prompt. Not just the same data: the same instructions, the same output schema, the same wording. We wrote that prompt once and we don't touch it. No per-model tuning, no prompt tweaks mid-season, no coaxing a better answer out of one model than another. Changing the prompt would change the test, so the only thing that varies between models is the model.
Line-blind. The model never sees the betting line. It produces its own win probabilities from the data alone, so we're measuring forecasting skill, not an echo of the market.
One packet, no search. It gets the same point-in-time data (ratings, Statcast, bullpens, park/umpire, situational splits) ~3 hours before first pitch, with web search off. It's the model we're measuring.
Every market, graded on its own. From that one forecast we score the moneyline winner and the run total (over/under) on MLB, and the winner and method of victory on UFC — each against the final result, and the money markets against the closing price (ROI).
Graded in public, no do-overs. One set of calls per game and we live with it. See the full leaderboard and the Can AI beat Vegas? essay.
Compare Grok with the field
Same brand, graded head-to-head — see Grok 4.5's record.
GLM 5.2's graded record, line-blind.
Claude Opus 4.8's graded record, line-blind.
Gemini 3.5 Flash's graded record, line-blind.
DeepSeek V4 Pro's graded record, line-blind.
Claude Sonnet 5's graded record, line-blind.
GPT-5.6 Sol Pro's graded record, line-blind.
Frequently asked
Can Grok predict sports?
We grade Grok (Grok 4.3) in public. Across 168 MLB games forecast line-blind — no betting line, no web search — it is 92–76, a 55% winner accuracy, with a Brier score of 0.246. Every call is locked before the game and graded against the final result.
What can Grok predict — just winners?
No — Grok makes a full forecast for every game, and each part is graded separately in public. On MLB it calls the moneyline winner and the run total (over/under). On UFC it calls the winner and the method of victory — KO/TKO, submission or decision. So far its over/under calls are 57% accurate over 163 games. Its UFC method calls are 46% accurate over 13 fights.
Is Grok good at sports betting?
Every pick is also scored as ROI at the closing market price, so the number is honest about the vig. So far Grok 4.3 has returned -1.3% per unit over 168 graded MLB games. Beating the closing line is a high bar — this page tracks how close it gets, updated nightly.
How accurate is Grok at predicting baseball?
Over 168 graded MLB games, Grok 4.3 has picked the winner 55% of the time, with a Brier score of 0.246 (0 is perfect, 0.25 is a coin flip). The 'pick the home team' baseline over the same games is 50%.
Track Grok and the whole field.
Who's hot, who's slipping, where the AIs disagree — free in your inbox every morning.