How accurate are our probabilities?

Our whole approach rests on one thing: when we say a horse has a 30% chance, it should win about 30% of the time. Here's the proof, measured against real results, published openly.

↑ Close to reality
±0.5pp
average gap between predicted and actual

Across 41,351 runners the model has assessed, its predicted win probabilities land within 0.5 percentage points of what actually happened, on average.

Close to reality: across 41,351 runners, our chances land within 0.5 points of what happened on average. Our highest ratings (35%+) run about 3 points high, and we show that below.

Predicted chance vs actual win rate

Every horse the model assigns a probability to, grouped by predicted chance, compared to how often those horses actually won. The closer "Actual" is to "Predicted", the more trustworthy the number.

Predicted Chance Predicted Avg Actual Win Rate Difference Runners Accuracy
0–5% 3.3% 3.9% +0.6pp 9774
5–10% 7.3% 7.5% +0.2pp 14588
10–15% 12.2% 12.4% +0.2pp 8095
15–20% 17.2% 17.7% +0.5pp 3762
20–25% 22.3% 23.5% +1.2pp 2023
25–30% 27.4% 27.5% +0.1pp 1083
30–35% 32.5% 31.1% -1.4pp 877
35%+ 42.5% 39.5% -3.0pp 1149

pp = percentage points. Green = within 3pp · Amber = within 7pp · Red = needs attention. Based on 41,351 runners across 137 racing days, since 10 Jun 2026.

Accuracy by race type

Whether the probabilities hold up across different kinds of race. Predicted is the model's average chance; actual is how often those horses won.

Flat Handicap
Predicted10.6%
Actual10.4%
Difference -0.2pp
20746 runners · see the races →
Novice / Maiden
Predicted11.6%
Actual12.0%
Difference +0.5pp
8868 runners · see the races →
Jump Handicap
Predicted11.0%
Actual12.6%
Difference +1.5pp
4910 runners · see the races →
Other
Predicted11.4%
Actual11.9%
Difference +0.4pp
4900 runners · see the races →
Group
Predicted11.6%
Actual11.7%
Difference +0.1pp
1127 runners · see the races →
Listed
Predicted11.3%
Actual11.1%
Difference -0.2pp
800 runners · see the races →

Why we show you this

Most racing products show you their winners. We show you how accurate our numbers are, including where they're not perfect.

A well-calibrated probability is the foundation of finding value. If the model says a horse has a 25% chance and the market is offering odds that imply 15%, that gap is only meaningful if the 25% is real. This page is our evidence that it is, updated continuously from settled results.

If the numbers ever drift, you'll see it here first. That's the deal.

That record is published every day, wins and misses alike. Members get the same numbers on every UK and Irish race before it runs.

Founder members pay only £14.99 a month, locked for life, against £24.99 a month once the founder places go. Starts with 7 days free.

Start 7-day free trial

Days 1 to 7 are free. You're charged £14.99 when the 7 days end, unless you cancel before then. We'll email you 2 to 3 days before you're charged, and you can cancel any time from your account page.

Had a free trial before? You'll be charged £14.99 today.

Free account first, no card needed · see pricing