Value Betting Explained: Finding Prices Above the True Chance

Value betting means backing outcomes whose odds are bigger than their real chance of happening. That is the whole idea, and it has nothing to do with whether a particular bet wins. A bet at 2.20 on something that happens half the time is a good bet even when it loses, and a bet at 1.40 on something that happens 65% of the time is a poor one even when it wins.

The hard part is the word real. Nobody knows the true probability of a football match or a tennis set, the bookmaker included. Value betting is the habit of estimating that probability carefully, comparing it with the price, betting only when the gap is in your favour, and keeping records honest enough to show whether your estimates are any good. Nothing about it is guaranteed. The bookmaker's margin is hard to overcome, and many people who try never do. This guide shows the arithmetic so you can judge for yourself.

Value is about price, not about who wins

Start with a coin toss. Heads and tails each have a 50% chance, so the fair decimal price is 2.00. If someone offers 2.10 on heads, a £10 bet wins £11 half the time and loses £10 half the time. Its expected value (EV) is (0.50 × £11) - (0.50 × £10) = £5.50 - £5.00 = +£0.50, a 5% edge per bet. At 1.90 the same sum gives (0.50 × £9) - (0.50 × £10) = -£0.50, a 5% disadvantage. The coin is identical in both cases. Only the price changed.

The general formula with decimal odds is: EV per unit staked = (probability × odds) - 1. A positive number means value. A negative number means you are paying the bookmaker for the bet.

This is also why value has nothing to do with backing favourites or underdogs. Suppose a team is 1.40 to win, an implied 71.4%, and you rate their chance at 75%. A £10 bet has EV = (0.75 × £4) - (0.25 × £10) = £3.00 - £2.50 = +£0.50. An underdog at 5.00 (implied 20%) that you rate at 17% has EV = (0.17 × £40) - (0.83 × £10) = £6.80 - £8.30 = -£1.50. The short-priced favourite is the value bet here, and the long shot that looks generous is not. Our moneyline guide has more on how favourites and underdogs are priced.

Implied probability and the bookmaker's margin

Every price converts to an implied probability. In decimal odds it is 1 ÷ odds, so 2.10 is 47.6% and 1.80 is 55.6%. In American odds, a plus price such as +150 is 100 ÷ (150 + 100) = 40%, and a minus price such as -150 is 150 ÷ (150 + 100) = 60%. The odds guide covers fractional conversions as well.

Add up the implied probabilities of every outcome in a market and the total comes to more than 100%. Take a football (soccer) 1X2 market priced at 2.10 home, 3.40 draw and 3.75 away:

• Home 2.10: 47.62%
• Draw 3.40: 29.41%
• Away 3.75: 26.67%
• Total: 103.70%, a margin of 3.7%

To estimate the market's fair view, divide each figure by the total. Home becomes 47.62 ÷ 1.037 = 45.92%, the draw 28.36% and away 25.72%, which convert back to fair odds of roughly 2.18, 3.53 and 3.89. This proportional method is a simplification. Bookmakers tend to load more margin onto long shots than onto favourites, a pattern known as the favourite-longshot bias, so the fair price of an outsider is often a little longer than the proportional method suggests.

Keep the two numbers apart. The fair figure of 45.92% tells you what the market thinks. The raw figure of 47.62% is the bar you have to clear, because you are paid at 2.10, not at 2.18. To find value on the home side at 2.10, you need to rate the home win above 47.62%.

Estimating your own price

Value betting depends on having a number of your own before you look at the odds. These are the common ways to get one:

• Ratings or models built from past results, such as expected goals (xG) in football or serve and return points won in tennis.
• Starting from a price in a market that tends to be accurate, such as a busy betting exchange, and adjusting for information you believe it has not absorbed yet, such as a confirmed injury.
• Comparing prices across many bookmakers and treating the consensus, with the margin removed, as a baseline.
• Specialist knowledge of a narrow market, such as a lower league or a smaller sport, where prices get less attention.

Whatever the method, write the number down first. Say your estimate for a home win is 50%, which is a fair price of 2.00. A bookmaker offering 2.20 gives an edge of (0.50 × 2.20) - 1 = 10%. At 2.05 the edge is only 2.5%, small enough to disappear with a small error in your estimate.

Be honest about that error. If you rate a team at 50% but could just as easily justify 45%, a price of 2.10 may not be value at all: 0.45 × 2.10 = 0.945, a 5.5% disadvantage. Many value bettors set a minimum edge before they will bet for exactly this reason, since their own probabilities are estimates, not facts.

Closing line value

Results take hundreds of bets to mean much, so many bettors also track a faster signal: closing line value (CLV). The closing line is the last price available before the event starts, when the market has absorbed the most information. If you regularly take prices bigger than the closing price, that is evidence your estimates are ahead of the market.

The calculation is simple. You back a team at 2.20 on Monday and by kick-off on Saturday the price is 2.00. Your CLV is 2.20 ÷ 2.00 - 1 = 10%. If the price had drifted to 2.40 instead, your CLV would be 2.20 ÷ 2.40 - 1 = -8.3%. For a cleaner measure, remove the margin from the closing market first, using the proportional method above, and compare your price with the fair closing price rather than the raw one.

CLV is not a guarantee. A single bet that beat the closing line can still lose, and a closing line can be wrong. Over a few hundred bets, though, a consistently positive CLV is one of the more reliable signs that you are finding value, and a consistently negative one usually means you are not, whatever your short-term profit says.

Variance: why value bettors still lose for long stretches

Value is a long-run idea, and the long run is longer than most people expect. Imagine 100 level-stakes bets of £10 at 3.00, each with a true 36% chance of winning. Each bet has EV = (0.36 × £20) - (0.64 × £10) = £7.20 - £6.40 = +£0.80, so the expected profit is £80 on £1,000 staked, an 8% return. That is a strong edge by any standard.

Now look at the spread of outcomes. Each bet either wins £20 or loses £10, a gap of £30, so the standard deviation per bet is £30 × √(0.36 × 0.64) = £14.40. Over 100 bets it is £14.40 × 10 = £144. Roughly 95% of the time the result lands within two standard deviations of the expected £80, which covers everything from a £208 loss to a £368 profit. A bettor with a genuine 8% edge can quite easily be down after 100 bets. Even ten straight losers, at 0.64 to the power of 10, has about a 1.2% chance in any given run of ten bets, which sounds rare until you have placed several hundred.

This is why staking matters as much as selection. Level stakes of 1% to 2% of a dedicated betting bankroll are a common approach. Some bettors use the Kelly criterion, which sets the stake at (b × p - q) ÷ b, where b is the decimal odds minus one, p the win probability and q = 1 - p. For the example above that is (2 × 0.36 - 0.64) ÷ 2 = 4% of the bankroll. Because p is only an estimate, full Kelly is aggressive, and many who use it stake a fraction such as a quarter or half of it. None of this changes the basic rule: never raise your stakes to win back losses.

Keeping records

Without records, value betting is guesswork with extra steps. A spreadsheet is enough. Log each bet with:

• Date, sport, competition and market
• Your estimated probability, noted before you saw the odds
• Odds taken and where you placed the bet
• Stake
• Closing odds
• Result and profit or loss

From that you can calculate the numbers that matter. Return on investment (ROI) is total profit divided by total staked: £75 profit from 250 bets of £10, or £2,500 staked, is a 3% ROI. Average CLV shows whether you are beating the market. Comparing your estimated probabilities with how often those outcomes actually happened tells you whether you are over- or under-confident. Split the figures by sport and market too, because a profit in one area can hide a steady loss in another.

Records also keep you honest about the time and money going into betting, which is part of betting responsibly. Our responsible gambling page covers limit tools and where to find support.

Where beginners go wrong

The most common mistake is treating value as a feeling. A price that looks big is not value unless you have a probability that says so, written down before you saw the odds. The second is judging a method on a small sample: 30 winning bets can be luck, and 30 losing ones can be bad luck on a sound approach.

Accumulators are another trap. If every leg of a four-leg acca comes from a market with a 5% margin, the acca pays only about 82% of its fair odds, because 1 ÷ 1.05 to the power of 4 is 0.823. The margins multiply along with the prices, so each leg needs a real edge just to stand still, as the parlay and accumulator guide shows.

Finally, beginners tend to ignore the practical side. Bookmakers can restrict the stakes of customers who consistently beat their prices, so a strategy that depends on one bookmaker's odds may not last. Prices also move quickly after team news, which means the value you spotted an hour ago may already be gone.

Before you place a value bet

Use the same routine every time, so your records measure your method rather than your mood.

• Write down your probability before you look at the odds.
• Convert the price to implied probability and check that your number is higher.
• Calculate the edge: (your probability × decimal odds) - 1.
• Check that the edge survives if your estimate is a few points too high.
• Compare prices across bookmakers and take the biggest one available to you.
• Make sure the operator is licensed where you live. Laws differ by country and, in the US and India, by state, and some Indian states ban online betting.
• Stake a fixed, small share of your bankroll, log the bet, and add the closing odds once the event starts.