In flat racing, every horse is assigned a starting stall — the "draw". On some courses and distances, where you start genuinely affects your chance. On most, it barely matters. Knowing the difference is the whole skill.
Why draw bias exists
It comes down to geometry and ground. A sharp bend soon after the start can favour low (inner) draws that save ground; a long run to the first turn can neutralise it. Sometimes a strip of faster or slower ground down one side of the track gives runners there an edge. Sprint races, with no time to recover from a poor position, show the strongest effects.
Why it's smaller than punters think
Draw bias is real but frequently overstated. Many "biases" are small samples — a handful of races where the well-drawn horses also happened to be the best horses. Pace, jockey tactics and plain ability usually swamp the draw. Betting heavily on stall number alone is a quick way to lose.
How to weigh it
- Treat it as a tiebreaker, not a headline factor.
- Trust it most in big-field sprints on known biased tracks.
- Be sceptical of dramatic claims from small samples.
How a model handles it
Rather than apply a blanket rule, our model learns the draw's effect by course, distance and field size from decades of races — and lets it count only where the data supports it. It's one input among many, alongside ratings and going. See the data behind the model.