The short answer: in American racing the favourite wins about two races in five, more often than in Britain and Ireland. The longer answer, which matters if you bet, is that backing every US favourite still loses at the odds, and the famous tracks are the worst place to do it.
US racing is run mostly on dirt, in smaller fields, and it is more formful than the big-field European handicap. So you would expect favourites to win more often here, and they do. We took every US result we have captured over a rolling year, from October 2025 to September 2026, with a clear starting-price favourite: 21,902 races. Here is what the data says.
So the American favourite is a better bet than the British one for hitting the target, but the market prices that in and then some. Winning four times in ten is not the same as making money: at the odds on offer, the blind favourite backer is still handing over roughly 4p in every pound.
Split by surface and the pattern holds: favourites win a little more often on the main-track dirt than on turf, though both lose at SP.
| Surface | Races | Win rate | ROI at SP |
|---|---|---|---|
| Dirt | 17,129 | 41.3% | -4.2% |
| Turf | 4,773 | 38.6% | -3.7% |
As in Britain, field size does most of the work. In a short-field US dirt race the favourite is close to a coin-flip; pack the field and the edge drains away.
| Field size | Races | Win rate | ROI at SP |
|---|---|---|---|
| Small (4 to 6 runners) | 7,585 | 45.9% | -4.0% |
| Medium (7 to 8) | 8,294 | 39.5% | -5.1% |
| Large (9 to 10) | 4,289 | 35.9% | -3.9% |
| Big field (11+) | 1,617 | 34.4% | +0.7% |
This is the part most punters get wrong. You might assume the elite circuits, Keeneland and Saratoga, are where the best horses win as expected. The opposite is true: they draw the deepest, most competitive fields in the country, so the favourite is beaten more often there than almost anywhere. The best strike rates come from the smaller regional tracks. See the full 35-track ranking. See also favourites by field size.
| Track | Races | Fav win rate | ROI at SP |
|---|---|---|---|
| Will Rogers Downs | 237 | 47.7% | +0.9% |
| Penn National | 675 | 46.2% | +2.5% |
| Finger Lakes | 692 | 45.7% | +0.4% |
| Delaware Park | 584 | 44.3% | +2.0% |
| Saratoga | 524 | 36.3% | -6.7% |
| Keeneland | 303 | 35.0% | -13.5% |
| Turfway Park | 527 | 33.0% | -3.6% |
A handful of small tracks (Penn National, Delaware Park) even show a small profit to blind favourite backing, but those are modest samples over a single year, so treat the plus signs with caution. The durable lesson is the shape: short prices in deep fields at the marquee meetings are where favourites are most over-bet.
The favourite is not a strategy, it is the market's opinion, and in the US that opinion is sharp. Across nearly 22,000 races the win rate and the payout cancel out to a small loss. If you want an edge you cannot get it from "who is the favourite"; you have to find the races where the price is wrong. That is exactly why we do not rate US racing with a model: we do not hold the full-field form to do it honestly. What we do publish is the raw, per-track picture so you can see where the crowd is right and where it is not.
The Overlay is a free, data-only US racing service, never a tipster. Our US track angles page has the favourite strike rate and level-stakes ROI for every track, sortable, alongside par times, the major-race calendar and live cards and results.
Open the US Racing HubWe used every US race we captured from 1 October 2025 to 24 September 2026 that had a settled result and a clear starting-price favourite (the single shortest-priced runner). Win rate is favourite wins divided by races. ROI is level-stakes return at starting price: stake 1 point on the favourite in every race and we measure profit or loss per point. Track rows use a minimum of 200 races. Figures are historical and for information only, never a guarantee.