Updated
Draw bias is when the starting stall a horse breaks from materially affects its chance of winning — usually because the track turns sharply or the field races toward one favoured strip of ground. It's measured with Impact Value (IV): a group's win rate divided by the average for that course and distance. The textbook case is Chester's 5-6f sprints, where over three recent seasons low-drawn horses won at IV 1.38 versus 0.69 for high stalls.
Tight, turning tracks show the clearest effects: Chester (low stalls strongly favoured at most trips), Beverley's stiff 5f (low favoured), and several all-weather tracks at sprint trips. Galloping, straight courses like Newbury show little. Racing Alpha publishes the full three-year heat map for every GB and Irish flat course free at /draw-bias, with sample sizes shown.
Three standard stats. Impact Value (IV) = the stall group's win rate ÷ the average win rate at that course and trip (1.00 = neutral). A/E = actual wins ÷ market-expected wins, which asks whether the bias is already in the price. PRB = percentage of rivals beaten, which uses every finishing position rather than just wins. A real bias shows up consistently across all three.
Usually not — and this is the part most draw guides skip. The market knows about famous biases, so favoured stalls are often over-bet. At Chester sprints, for example, low-drawn horses won far more races (IV 1.38) but returned an A/E of 0.91 — meaning the market over-priced the advantage. A bias can be real and still offer no value.
Yes. Rail movements shift the favoured strip on straight courses, and soft ground can move the best ground from one side to the other. Treat historical draw stats as a baseline, then check how today's races at the meeting are actually being run.
Racing Alpha is for information and entertainment for adults aged 18 and over. We do not accept bets. Past performance does not predict future results. BeGambleAware.org