TrapMetrics3,387 winners from 8,198 runners
The record behind the product
Stronger claims. The numbers beside them.
Every figure here comes from our own settled record, with its race count, its window and its date. The closing market beats every rating, ours included, and that result is shown too. Snapshot 1 November 2026, refreshed monthly.
Straight forecast / since 4 August 2026
Measured against Timeform. And the Racing Post.
The model’s most likely pair landed the straight forecast more often than either published forecast, on the same races.
10.5%TrapMetrics model
straight-forecast hit rate
straight-forecast hit rate
9,167 settled races from 4 August to 31 October 2026. A straight forecast is the first two home in the right order.
Straight forecast / same races
Combine all three opinions10.7%
The three-source blend landed more than any source alone. That is why the racecard shows a consensus forecast.
A hit rate is not a return. The forecast pool keeps about 28% of every pound staked, and no filter we have tested beats it.
As issued / Top-pick accuracy
31.5% versus 28.9%.
The TrapMetrics top pick against the Timeform forecast favourite, on the ratings we published each morning.
That is 2.6 percentage points, with a 95% interval of +1.6 to +3.5, on 8,492 races and 47,519 runners over 91 racing days.
The backtest below, scoring the current model over 18,914 earlier races, gives 32.1% against 28.6%. The live record is the one that counts; the backtest is the longer sample.
The Betfair favourite scored higher than both. BSP is only known at the off, so it is a benchmark, not a price you could have taken in the morning.
Probability quality
How the whole set of chances scored.
Log-loss rewards putting probability on the winner and punishes confident mistakes. Lower is better.
| Source | Log-loss | Top pick won |
|---|---|---|
| Betfair SP | 1.490 | 37.0% |
| TrapMetrics Rating | 1.579 | 32.1% |
| Timeform forecast | 1.630 | 28.6% |
Sources & definitions
How each figure is measured.
Every number on this page, with the races it was counted over, the window, and what it can and cannot tell you.
Top pick and log-loss
Races from 18 May 2026, the first day the Timeform forecast is in our capture, to 31 October 2026, none of which the model trained on. The ratings were produced by scoring the current model over those races, so this is a backtest: the model was retrained on 28 July and the stored ratings before 1 August are not the ones published at the time. Complete fields only: every runner that ran has a rating, a Timeform forecast and a Betfair SP. Each source’s chances are normalised within the race. When joint favourites share top spot, a winner among them scores a fair share. 18,914 races, 103,580 runners.
The live record, as issued
8,492 races over 91 racing days from 1 August 2026 to 31 October 2026, using the ratings exactly as published each morning before racing. Top pick 31.5%, Timeform forecast favourite 28.9%, Betfair SP favourite 36.3%: 2.6 points ahead of the forecast, 95% interval +1.6 to +3.5, resampled by day. 13 weeks is a short record and we will keep publishing it as it grows.
Straight forecasts
The morning ledger from 4 August, when the Racing Post forecast joined the capture, to 31 October 2026: 9,167 settled races. Each source’s win chances are expanded into ordered pairs with the same correction the racecard uses, and its most likely pair is scored against the first two home. The blend averages the model, Timeform and the Racing Post.
Rating sources & definitions
Every settled runner with a TMR at or above the threshold, from 1 January to 31 October 2026. Not one runner per race, and the higher groups sit inside the lower ones: 115+ won 41.3% of 8,198, 120+ won 50.3% of 1,962, 125+ won 62.6% of 454. 115 is the scale’s built-in “twice the average chance” mark; 120 and 125 were looked at after the fact, so read them as description, not as a rule.
The System Builder record
1,549,046 runs with a result, in 273,704 races since June 2021: the table every Builder rule is tested against, settled at Betfair SP after commission.
Refreshed monthly
This snapshot is dated 1 November 2026. The figures are recomputed at the start of every month by the same queries, so if a number moves, this page moves with it.
What these figures do not say
A hit rate or a strike rate is not a return. The closing market is more accurate than every rating here, ours included. Nothing on this page is a promise about the next race.
Your next step
Find your place in the research.
Try the tools, read the evidence and choose the access that fits your questions.
View access optionsOpen a real race
Register before accounts open and Research is £19.99 a month, locked in.