Value betting explained: how to find value bets and expected value

Every price implies a probability (1 ÷ odds, see implied probability). A bet has value when the real chance is higher than the price implies.

Value = your probability × decimal odds

• Above 1: positive value (the price is too generous).
• Exactly 1: a fair price.
• Below 1: negative value (the normal case, because of the margin).

Notice the last row. Value is not about favourites or outsiders. A 4.50 shot that you rate at 25% is a better bet than a 1.80 "banker" you rate at 50%, even though it loses three times out of four.

The hard part is not the maths. It's the estimate. If you think a team has a 50% chance just because you like them, the formula will happily tell you every bet is value. Most bettors lose over time precisely because the bookmaker's estimate, plus its margin, is better than theirs.

Expected value tells you what a bet is worth on average, in money.

EV = (probability of winning × profit if it wins) − (probability of losing × stake)

A shortcut for decimal odds: EV = stake × (probability × odds − 1).

Now the same bet at 1.80: £10 × (0.50 × 1.80 − 1) = £10 × (−0.10) = −£1.00. Same match, same opinion, a different price, and the bet goes from good to bad. That's why price shopping matters: compare the prices of licensed bookmakers before you bet, and see choosing a bookmaker for how to read their margins.

Most people can't out-predict the market from scratch. A more practical approach is to estimate the fair price from the market itself and look for a price that beats it.

1. Take a sharp, low-margin 1X2 market.
2. Strip out the margin (normalise the implied probabilities).
3. Compare the fair probability with the best price on offer elsewhere.

Here is the same comparison on a real match from today's schedule (implied chance against the fair, margin-free chance):

This method has limits. The fair price is only as good as the market you take it from, and a 4.5% edge from one price is a small number that can disappear with news you haven't seen (an injury, a rotated line-up). Still, it's the most honest starting point for a beginner: you're not claiming to know better than everyone, only that one price is out of line with the rest.

A positive EV is a long-run average. In the short run, luck dominates.

So even with a genuine 10% edge, roughly one in five runs of 100 bets ends in a loss, and a range from about −£120 to +£320 (two standard deviations either side of +£100) is perfectly normal. Real edges are usually much smaller than 10%, which means you need hundreds or thousands of bets before results say anything about skill.

Two practical lessons follow:

• Judge the process, not the last ten results. A losing week doesn't prove you were wrong, and a winning week doesn't prove you were right.
• Stake small. Variance is survivable only if no single bet can hurt your bankroll. See bankroll management.

Because results are so noisy, many bettors track closing line value: did you get a better price than the final price before kick-off? The closing price includes all the information available at the start (team news, market money), so it's the best public estimate of the true chance.

One bet that beats the close means little. But if, over a few hundred bets, your prices are consistently higher than the closing prices, that's strong evidence you're finding real value, whatever your profit says so far. If your prices are usually worse than the close, a winning run is probably luck.

The Kelly criterion sizes a bet according to your edge:

Kelly stake (% of bankroll) = (b × p − q) ÷ b, where b = odds − 1, p = your win probability, q = 1 − p.

Full Kelly assumes your probability is exactly right, which it never is. Overestimate your edge and full Kelly overbets badly. Most careful bettors use a quarter or half Kelly, or simply flat stakes of 1–2% of the bankroll. Try your own numbers here:

Our predictions start from the same logic as this guide: estimate a fair probability from the data and from margin-free market prices, then flag a pick only when the available price is above that fair price. Every settled pick goes into a public record with the odds we quoted, so you can check the hit rate, the profit and the losing runs yourself rather than take our word for it.

• Confusing value with "likely to win". A 1.30 favourite can be terrible value; a 6.00 outsider can be good value.
• Using your gut as the probability. If you can't explain where your 50% comes from, the EV figure is meaningless.
• Judging an edge on a small sample. Twenty bets tell you almost nothing. Track hundreds, and track CLV.
• Ignoring the margin. In a 7–8% margin market you need a far bigger edge than in a 3% market.
• Overstaking after a good run. Your edge doesn't grow because you won; keep stakes in line with your bankroll.
• Chasing a "value" price into a market you don't understand. Odd prices in obscure markets are often odd for a reason.

The honest summary: value betting is the only judgement-based approach that can be profitable in theory, but most people who try it don't have a real edge, and even those who do face long losing runs. Treat it as a disciplined way to bet less badly, not as a promise of profit.