Value is when a horse’s real chance of winning is better than the chance its price says it has. That’s the whole idea. A horse at 4.0 is priced as a 25% chance. If you honestly think it wins one time in three, about 33%, you’ve found value. If you think it wins one time in five, you haven’t, however good the horse looks.
Plenty of people lose at the races by picking the horse they think will win and backing it. The picks aren’t the problem… the question is. Picking winners and making a profit are two different jobs.
The short version
- Value is real chance versus the chance the price implies.
- Odds convert to a probability: divide 1 by the decimal odds.
- Backing favourites blind doesn’t find it. The market has already done that work.
- Write your own price before you look at the board.
- Value bets lose more often than favourites. That’s normal.
How do odds turn into a probability?
Divide 1 by the decimal odds. That gives the market’s implied chance.
| Decimal odds | Implied chance |
|---|---|
| 1.5 | 66.7% |
| 3.0 | 33.3% |
| 3.75 | 26.7% |
| 6.0 | 16.7% |
Every price on the board is the market telling you what it thinks a horse’s chance is. That’s the only job odds do for this purpose.
Does backing the favourite find value?
No. The market is good at ordering horses. In Race Advisor’s UK data from 2013 to 2026, around 1.5 million runners, the favourite wins about 34% of the time. Favourites win more often than second favourites, second favourites more often than thirds, and so on down the card.
The question is whether the prices are wrong. The A/E ratio tests that. A/E stands for actual over expected: the number of horses that actually won, divided by the number the prices said should have won. Above 1.00 means more winners than the odds implied. Below 1.00 means fewer.
At Betfair SP (the exchange’s starting price), the favourite, the second favourite and the third favourite come out at about 1.00, 0.99 and 0.99, each across roughly 160,000 runners. That’s very close to fair. The market is doing its job, so picking a rank off the board gives you no edge. The A/E ratio explained shows how to work the number out.
These are past results, and they tell you how prices have behaved, not how they will behave. They are also before exchange commission.
So what question should I ask instead?
Stop asking which horse will win. Thousands of people are trying to answer that, and the price already reflects most of what they know. Ask whether this price is wrong. Far fewer people do that.
What does a value bet look like?
Say a chase has a favourite at 3.0, a 33.3% implied chance. The horse you rate third is 6.0, a 16.7% implied chance. If your honest number for that horse is 24%, you have value: 24% against 16.7%.
That horse still loses most of the time. At 24% it loses about three races in four. It makes sense only if your 24% is right, and that is the hard part. It’s easy to overrate the horses you like. A rating you’ve never tested is a guess, not an edge. If you want to test one, how to test a betting system live covers it.
How do I build a value habit without a spreadsheet?
You need three small habits, and a pen will do.
- Write your price first. Put a number next to your selection before you look at the board. If you say 4.0 and the market says 6.0, look closer. If you say 4.0 and the market says 2.8, the horse may be the likeliest winner but it isn’t the bet.
- Compare to Betfair SP, not the morning price. Bookmaker prices carry a margin, which eats the gap you’re looking for. Betfair SP is the cleaner yardstick, though commission still applies.
- Expect to lose more often. A 6.0 shot that wins 25% of the time loses 75% of the time. You need to be able to sit through that. Bankroll management is about surviving the losing runs.
One more check. A tip is not the same thing as a value bet. If you’re looking at a tipped horse, compare today’s price with the price when the tip was first given. If it has shortened, some of the value may already be gone.
What if I want help seeing the gaps?
Race Advisor’s racecards show the Race Predictor’s estimated win chance for each runner, next to a value figure comparing it with the market price. It doesn’t tell you what to back. You can set its number against the price and decide for yourself, and treat any disagreement as something to test, not a bet to place. I haven’t put a hit rate on it here, and I’d rather not guess one.
What this doesn’t tell you
It doesn’t tell you how to rate a horse accurately. That’s the hard part, and it takes data and time. It doesn’t promise a profit either. Every figure here is from past UK races, before commission, and prices can change. Whether your own numbers beat the market is only something a proper sample will show you.
