Most bettors can tell you their strike rate to the decimal place. Almost none can tell you their edge. That matters, because strike rate on its own will fool you. It says backing favourites is “working” right up to the point your bank stops growing and you cannot work out why.
Edge ties together the two things that decide whether a bet pays: how often you win, and what you are paid when you do. Drop either one and you cannot tell skill from busyness.
Here is how to work it out, step by step, with real figures from 13 years of racing.
How do you calculate your edge in horse racing betting?
Multiply your win rate by your average winning price (the decimal price including your stake, after Betfair commission), then take away one. Edge = win rate × average winning price − 1. A positive answer means your bets have paid more than they cost. A negative one means they have not.
What “edge” actually means
Edge is your profit per £1 staked, averaged over a lot of bets. An edge of 5% means that for every £1 you stake you get back £1.05 on average. An edge of 0% is break-even. A negative edge means you are paying for the entertainment.
Two numbers decide it, and only two:
- How often you win, written as a decimal. Winning 25 bets in 100 is 0.25.
- What you are paid when you win, as a decimal price that includes your stake, after commission if you bet on the exchange.
A 50% win rate at an average winning price of 1.50 loses money (0.5 × 1.50 = 0.75, so an edge of -25%). A 15% win rate at an average winning price of 9.00 is a healthy edge (0.15 × 9.00 = 1.35, so +35%). That is why strike rate alone tells you so little.
Getting your two numbers right
Win rate. Count the bets you placed under one clear rule, count the winners, and divide. Be honest about sample size. Twenty bets tells you very little. A couple of hundred gives you a first working number, and a thousand a steadier one, though how much any count can tell you depends on your strike rate and the prices you take. Long-priced bets need more.
Average winning price after commission. This is where many people trip up. They write down the price they took and forget that Betfair keeps a slice of winnings. To fix that, take the price, subtract one to get the profit per £1, multiply by one minus the commission rate, then add one back.
At 2% commission, a Betfair SP (the Betfair starting price) of 6.00 works out as (6.00 − 1) × 0.98 + 1 = 5.90.
That step only applies on the exchange. With a bookmaker, the price you took is the price you keep.
The formula, worked through
Here is the sum on a real signal. Take the Betfair SP favourite across 167,257 GB and Irish races between 1 July 2013 and 14 October 2026. It won 34.4% of the time. Worked back from the return figures, its average winning price after 2% commission comes out at about 2.87.
- Win rate as a decimal: 0.344
- Average winning price after commission: about 2.87
- Multiply: 0.344 × 2.87 = about 0.987
- Take away one: about -0.013, or roughly -1.3%
The database figure is -1.4%. The small gap is rounding in the 2.87. The point is the method: two numbers in, one honest answer out.
A slightly negative answer for the favourite is not a disaster. It means the market prices the favourite fairly, and what you lose is mostly the commission.
What the simple signals look like
The same sum on three obvious signals, at Betfair SP, level stakes, after 2% commission. The average winning prices are worked back from each signal’s return.
| Signal | Races | Win rate | Average winning price (approx.) | Edge |
|---|---|---|---|---|
| Betfair SP favourite | 167,257 | 34.4% | 2.87 | -1.4% |
| VDW Rank 1 (top of our VDW ratings) | 153,920 | 21.6% | 4.45 | -3.8% |
| Race Predictor top horse | 167,314 | 25.5% | 3.92 | 0.0% |
Notice the pattern. A higher win rate comes with a shorter price, and a lower win rate with a longer one. These three broad, obvious signals all land close to zero, which is what you would expect in a well-priced market. These are past results, a description of what happened rather than a promise about the next 13 years.
That is the useful lesson. A good-looking win rate is not an edge, and the obvious signals will not hand you one for free. A real edge is a small positive number, found by testing your own rules.
Using your edge number honestly
Three habits keep the number honest.
Insist on size. An edge of 15% on 60 bets is mostly noise. The same figure across 600 bets under one clear rule is worth attention. If you cannot state your sample size, you have a hunch.
Track price honestly. Use the price you actually received, not the one you screenshotted. On the exchange, take off commission. Prices that vanished before you could bet do not count.
Pair it with A/E. Edge and A/E are two views of the same idea. Edge answers “how much did I make per £1?”. A/E (actual wins divided by the wins the odds imply, explained here) answers “did I win more than the odds suggested?”. At Betfair SP, an A/E of 1.00 lands slightly below zero after commission, as the favourite’s -1.4% shows. Use A/E to spot a promising signal, and edge to see whether it survives commission.
Edge is the test of whether your analysis is real. What value means in racing comes down to this same sum: probability and price, put together honestly.
Putting it into practice
Pick one clear rule, write down the price you would take, and start counting on today’s race cards. After 200 bets you will have a first working number, and after a thousand a steadier one. Expect swings along the way, including losing runs. Keep stakes small enough that you can collect the sample without being forced to drop the rule halfway. If you want to check the rule is built on solid ground, our betting systems guide shows how to turn a hunch into a testable rule.
A small positive edge, measured honestly and bet with sensible stakes, is a perfectly good place to aim. You do not need a big number. You need a true one.
