A value bet is a bet whose odds are higher than they should be given the true probability of the event. In other words, the bookmaker has underestimated the outcome's chances. If your probability estimate is right, it is a positive expected value bet: over the long run it returns more on average than it costs. Below is what value means in practice, how to find it and why it is harder than it looks.
What value means in plain terms
Every set of odds is a price. From the price you can get the probability the bookmaker is implying: 1 / odds. Odds of 2.50 correspond to a probability of 40%.
If you believe the event will really happen 45% of the time, a price of 2.50 is good for you: fair odds at 45% are about 2.22, and the bookmaker is offering more. That is value. If your estimate is 35%, the same 2.50 means overpaying for the event.
The key idea: a value bet is not a bet on something that will definitely happen. It is a bet at a price above fair value. A favourite at 1.30 can be poor value, and an underdog at 6.00 can be good value.
Positive expected value: how to calculate it
The expected value of a one-unit bet is:
EV = probability × odds − 1
- with a probability of 45% and odds of 2.50: 0.45 × 2.50 − 1 = +0.125, i.e. +12.5% of the stake on average;
- with a probability of 35% at the same odds: 0.35 × 2.50 − 1 = −0.125.
Positive EV does not mean the bet will win. It means that if you repeat such bets many times, the result will be positive on average. Any single bet still wins or loses, and losing streaks happen even with correct estimates. How to get a probability from odds and remove the bookmaker margin is explained in detail in probability from odds.
How to find value: where your probability comes from
The whole idea rests on one thing: your probability estimate has to be more accurate than the market's. Where to get it:
- Your own statistical calculation. Team form, chance quality, line-ups. It takes time and discipline.
- An independent model. Ready-made football predictions on xG.Football show an estimated probability next to bookmaker odds, which makes them easy to compare.
- Comparing prices. If one bookmaker is noticeably off the others on a single outcome, it is worth a closer look. It is not proof of value on its own.
Before trusting someone else's probability, check how well it has matched reality in the past. The log of past predictions and their accuracy is open on the prediction statistics page. If a model claims more on average than actually comes true, its "value" will be imaginary.
Why finding value is harder than it seems
- Margin. The bookmaker builds a commission into every price. For a bet to be value, your edge has to be bigger than the margin.
- Estimate accuracy. The gap between 45% and 40% is the gap between value and overpaying. Being five percentage points off is very easy.
- Line movement. Odds shift as team news comes in. Value spotted in the morning may be gone by the evening.
- Variance. Even with a real edge, the result of a hundred bets can be negative.
Common mistakes
- Treating any bet on an underdog that "feels underrated" as value.
- Comparing your probability with the odds without accounting for the margin.
- Drawing conclusions from a dozen bets.
- Raising stakes after a losing run to "win it back". How to avoid that is covered in bankroll management.
The short version
A value bet is a price above fair value, not a guaranteed win. Finding value requires your own probability estimate that is more accurate than the market's, and an honest check of that estimate over the long run. Without that, "value" is just an ordinary negative expected value bet.
| Question | Answer |
|---|---|
| What is a value bet? | A bet whose odds are higher than fair given the true probability of the event. |
| How do I calculate the expected value of a bet? | EV = probability × odds − 1. If the result is above zero, the bet has positive expected value. |
| Does a value bet guarantee a win? | No. Positive expected value only shows over the long run; any single bet can lose. |
| How do I find value? | You need your own probability estimate that is more accurate than the market's, compared with the margin-free probability from the odds. |