I spent two weeks asking ChatGPT to predict 50 sports results — with real match data fed in every time. It hit 52%. One prediction short of the break-even threshold at -110. Not profitable. But not random noise either.
Here's the full experiment, where it fell apart, and the one scenario where accuracy jumped to 60%.
The Experiment
I gave ChatGPT real match data for each prediction — not just "who wins tonight?" I fed it head-to-head records, recent form, home/away splits, and injury reports pulled from official team pages. Then I logged every single result.
50 matches. Premier League, NBA, NHL, and a handful of international football fixtures. The question wasn't whether AI is smart. The question was whether it can beat a coin flip often enough to matter if you're betting.
The Raw Numbers
ChatGPT called 26 out of 50 matches correctly. That's 52%. Sounds decent until you do the math — at -110, you need to hit 52.4% to break even. ChatGPT landed one prediction short.
I ran the same matches through Gemini and Claude for comparison. Results were nearly identical — all three models hovered in the 50–54% range depending on sport and sample size.
ChatGPT was strongest picking heavy favourites. When one team was clearly dominant on paper, it called it correctly about 62% of the time. The problem? Those outcomes are already priced in. Betting the favourite at -200 doesn't move the needle even at 62%.
Where It Falls Apart
The model struggled most with three types of matches:
- Mid-table Premier League clashes — too many variables, neither team dominates on paper
- Back-to-back NHL games — fatigue isn't baked into the training data the way it needs to be
- Late injury news — ChatGPT's knowledge cutoff means it won't know a starting goalkeeper is out the morning of the game unless you paste the news in yourself
That last one is the biggest issue. Sports predictions live and die on real-time information. A model without current injury data is working with fundamentally incomplete inputs — and the market already has that information priced in.
What Happens When You Feed It Everything
For the last 15 matches, I changed the approach. Instead of just asking "who wins?", I pasted in full match previews from reliable sources, injury bulletins, and line movement data from sharp sportsbooks.
The accuracy jumped to 60% on those 15 matches.
That's still not a winning record at -110, but it's a meaningful difference. ChatGPT isn't a prediction engine — it's an analysis engine. Give it quality inputs and it synthesizes reasonably well. Ask it to conjure predictions from thin air and you'll get coin-flip results.
The World Cup 2026 Predictions Test
For fun, I asked four different AI tools to predict the 2026 World Cup winner:
- ChatGPT: Argentina
- Gemini: Netherlands
- Claude: France
- Perplexity: Spain
Four different answers. No consensus. These models are drawing from training data, not crystal balls — they're essentially telling you who was good, not who will be good.
Is ChatGPT Useful for Betting at All?
Yes, but not as a tipster. The ceiling for general AI tools in match prediction sits around 55–60% when you feed them excellent real-time data. That's interesting. It's not profitable at standard vig, but it's not random noise either.
Where it earns its place: building quick comparison tables across multiple matches, arguing the opposite case to your own lean, or stress-testing a model's assumptions. What it can't do is generate edges you don't already have a hypothesis for.
Frequently Asked Questions
Can ChatGPT predict soccer / football matches?
Our testing showed approximately 52% accuracy across football matches — marginally below break-even. Performance improved to around 57–60% on matches with clearer form differentials when current injury and lineup data was provided manually.
Is ChatGPT better than a tipster for betting?
Not on raw pick accuracy. Where ChatGPT has an edge is analysis speed and no financial incentive tied to its picks. A human tipster who sells subscriptions has a different relationship with their track record than an AI does.
Which AI is best for sports predictions?
In our testing, ChatGPT, Gemini, and Claude all performed in a similar 50–54% range on general predictions. Purpose-built tools like Rithmm, which let you build sport-specific models, outperformed general chatbots when used correctly.
Bottom Line
ChatGPT at 52% won't make you money betting sports. It will lose you money slowly, which is somehow worse than losing it fast.
The honest answer to "can ChatGPT predict sports results" is: barely, and only if you do most of the work first. Think of it as a research assistant, not a prediction engine — and make sure the data you feed it is current.