Expected Value (EV) in Betting Explained
Expected value is the single most important concept in sports betting. It tells you, in precise mathematical terms, whether a bet is profitable or not — not whether it will win, but whether it is worth placing. Every serious bettor, professional gambler, and sharp sports investor thinks in terms of EV. This guide explains what EV is, how to calculate it on any bet, what positive and negative EV mean in practice, and why consistently finding +EV bets is the only sustainable route to long-term profitability.
What Is Expected Value?
Expected value (EV) is the average profit or loss you would expect from a bet if you placed it an infinite number of times. A positive EV bet gains money on average. A negative EV bet loses money on average — regardless of whether it wins or loses on any single occasion.
EV comes from probability theory and is used across finance, insurance, poker, casino gambling and sports betting to evaluate decisions under uncertainty. The core insight is this: the result of a single bet tells you almost nothing about whether placing it was a good decision. What matters is whether the price you received was better or worse than the true probability of the outcome.
A coin flip at evens (2.00 decimal) is zero EV — you will break even on average. If someone offers you 2.20 on heads in a fair coin flip, that's positive EV — you're getting paid more than the 50% probability warrants. If they offer you 1.80 on heads, that's negative EV — you're being underpaid for a 50% chance. This is the entire logic of EV, applied to any bet in any sport.
Understanding EV reframes how you think about every betting decision. The question stops being "will this bet win?" and becomes "am I getting the right price for what I think will happen?" Those are completely different questions — and only one of them leads to long-term profitability.
The EV Formula
Expected value is calculated by weighting each possible outcome by its probability and summing the results:
EV = (Probability of Winning × Profit if Win) − (Probability of Losing × Stake)
Or equivalently, using decimal odds:
EV = (Your estimated probability × Decimal odds) − 1
A result above 0 is positive EV. A result below 0 is negative EV. A result of exactly 0 is break-even.
Breaking Down the Formula
Your Estimated Probability
This is your own assessment of how likely the outcome is — expressed as a decimal between 0 and 1 (or as a percentage). This is the key input you are responsible for. Getting this right — having a probability estimate that's more accurate than the bookmaker's implied odds — is where the edge comes from.
The EV Result
The result of the formula is expressed as a proportion of your stake. An EV of +0.08 means you expect to gain 8p for every £1 staked, on average, over a large number of identical bets. An EV of −0.05 means you expect to lose 5p per £1 staked.
Positive EV vs Negative EV
The bet is worth placing
A positive EV bet is one where the odds offered are higher than the true probability warrants. You are being overpaid for the risk you are taking. Over a large enough sample, +EV bets make money. Individual results vary — you can lose a +EV bet — but the mathematical expectation is profitable. All profitable bettors are systematically finding and placing +EV bets.
The bet loses money on average
A negative EV bet is one where the odds offered are lower than the true probability warrants. You are being underpaid for the probability you are backing. Over a large sample, −EV bets lose money. This includes the vast majority of bets placed at standard bookmakers on well-priced markets — the bookmaker's margin ensures most bets are slightly negative EV by construction.
A positive EV bet can lose. A negative EV bet can win. In the short run, results are dominated by variance (luck). EV only predicts average outcomes over a large number of bets. A single result — or even a run of 50 results — can easily deviate from the expected value due to variance. This is why evaluating your betting by profit/loss over a small sample is unreliable — and why tracking EV over time is more informative than tracking win rate alone.
Worked Examples Across Different Bets
Example 1 — A Football Match Result
Your model estimates Man City have a 62% chance of winning (probability = 0.62). The bookmaker is offering 1.80 decimal odds.
EV = 1.116 − 1 = +0.116
This is a positive EV bet — for every £100 staked, you expect to gain £11.60 on average.
EV = (0.62 × 1.55) − 1 = 0.961 − 1 = −0.039
At 1.55, the same bet becomes negative EV — you'd lose £3.90 per £100 staked on average, even though you still think City are 62% to win. The outcome didn't change; the price did.
Example 2 — A Coin Flip with Biased Odds
Bookmaker offers 2.10 on heads (equivalent to fractional 11/10 or American +110).
+5% EV per bet. Place this bet 1,000 times at £10 stake:
Expected profit = 1,000 × £10 × 0.05 = £500 expected profit.
Example 3 — An Accumulator
If each leg of a 5-fold accumulator has an EV of −0.05 (i.e. the bookmaker has a 5% margin on each leg), the combined EV of the accumulator is not −0.05 — it's compounded across all legs.
Combined EV = (1 − 0.05)^5 − 1 = 0.774 − 1 = −22.6% EV
A 5-fold acca with just 5% margin per leg costs you 22.6% of your stake in expected value — not 5%. This is why accumulators, while exciting, are particularly poor EV bets. The margin compounds with every leg added.
EV vs Winning — Why a Losing Bet Can Be Correct
This is one of the most important — and counterintuitive — concepts in betting. The correctness of a bet is determined by whether it was positive EV at the time of placing. Whether it wins or loses is a separate question determined by the actual outcome.
EV = (0.25 × 5.00) − 1 = 1.25 − 1 = +0.25. This is a strong +25% EV bet.
The underdog loses — as it will 75% of the time.
Was the bet wrong? No. The bet was correct. The outcome was unfavourable. These are different things.
EV = (0.60 × 1.30) − 1 = 0.78 − 1 = −0.22. This is −22% EV.
The favourite wins — as it often will.
Was the bet correct? No. It was a wrong bet that happened to win. Over time, bets like this lose money.
Over a large enough sample — hundreds or thousands of bets — EV predicts your actual profit with increasing accuracy. In the short run (tens of bets), variance dominates and results can deviate wildly from EV in either direction. This is why professional bettors track EV across their entire portfolio and don't evaluate individual bets by their outcomes.
Sample size determines how closely results track EVHow Bookmaker Margin Creates Negative EV
Every bookmaker builds a margin into their odds — meaning the combined implied probabilities of all outcomes in a market add up to more than 100%. This excess is the bookmaker's profit margin (also called the overround or vig), and it is the mechanism that makes every standard bet slightly negative EV by default.
A fair coin flip should be priced at 2.00 / 2.00 (50% + 50% = 100%).
A bookmaker prices it at 1.90 / 1.90 (52.6% + 52.6% = 105.3% implied).
EV on either side = (0.50 × 1.90) − 1 = 0.95 − 1 = −0.05 (−5% EV)
That 5% is the bookmaker's margin — built into both sides. Even if you pick the right side every time, the price is wrong by 5% per bet. This is why simply betting on who you think will win is a losing strategy long-term — the price is always slightly below fair value.
To achieve positive EV, you must either:
Have a More Accurate Probability Estimate
Your probability estimate for the outcome must be significantly higher than the bookmaker's implied probability — enough to overcome their margin. If you believe something has a 60% chance and the bookmaker's odds imply 55%, the margin is closed and the bet is +EV.
Find Mispriced Markets
Bookmakers occasionally misprice specific selections — particularly in less liquid markets, around breaking news, or when lines haven't been updated. These mispriced selections offer positive EV without requiring your model to be more accurate than the bookmaker's overall market — just that specific price is wrong.
Line Shop for the Best Price
If you can consistently find prices 5–10% better than average across bookmakers, you are reducing or eliminating the bookmaker's built-in margin. A bet priced at −4% EV at one bookmaker might be +2% EV at another offering a better price on the same selection. Getting the best available price is a direct improvement to your EV on every bet.
Use Odds Boosts Strategically
When a bookmaker promotes a selection at a vastly inflated price — say 3.50 on something the market prices at 2.20 — that boost is explicit, guaranteed positive EV. Unlike hunting for mispriced markets, the edge is visible and quantifiable before you bet. Incorporating bookmaker promotions into your process alongside standard line shopping is one of the fastest routes to +EV without any probability modelling.
How to Find Positive EV Bets
Finding +EV bets is the core skill of profitable sports betting. It requires either a better probability model than the bookmaker, identifying market inefficiencies, or both:
Form your view of the true probability of an outcome using data, form, team news, statistics — before you look at any bookmaker's price. Once you have a probability estimate, you can calculate whether any offered price is positive or negative EV. Looking at the odds first anchors your probability estimate to the bookmaker's view, which undermines the process.
Your estimate first, always — then check the priceConvert the offered decimal odds to implied probability: Implied prob = 1 ÷ Decimal odds. If your estimate (say 55%) is higher than the implied probability (say 48%), the bet is +EV. If your estimate is lower, the bet is −EV. The difference between your estimate and the implied probability, scaled by the odds, is your edge.
Pinnacle (where available) and betting exchanges are known for sharp, efficient pricing. If a selection is priced at 2.40 on Betfair Exchange but 2.80 at a softer bookmaker, the softer bookmaker's price likely represents positive EV — without you needing to form your own probability estimate at all. This is the basis of arbitrage betting and line shopping for value.
Sharp market prices are a calibrated EV benchmarkThe most efficient markets — Premier League match odds on a Saturday afternoon — are priced by bookmakers using enormous data resources and decades of model refinement. Finding +EV there requires a genuinely superior model. Less liquid markets — lower league football, niche tennis tournaments, player props — are less efficiently priced. Specialised knowledge of a specific market is often more valuable than broad knowledge across many markets.
When significant information breaks — a key player injury, a late team selection, a weather change — some bookmakers adjust their odds faster than others. During the window between the news breaking and the last bookmaker adjusting, the slower books are offering positive EV on the correctly anticipated side. Speed and awareness of breaking news is a genuine +EV source for active bettors.
Information timing creates +EV windows before markets correctEV Over Time — Variance and the Long Run
Understanding the relationship between EV, variance, and sample size is essential for interpreting your own results — and for maintaining the discipline to stick with +EV bets even through losing runs.
What Is Variance in Betting?
Variance is the natural statistical fluctuation around the expected outcome. Even if every bet you place is +EV, you will experience losing streaks — sometimes long ones. A series of 10 consecutive losses on +EV bets is entirely consistent with having a genuine edge. Variance is not a sign your model is wrong; it's a statistical inevitability at any sample size.
How Many Bets to Judge EV?
A sample of 50 bets is almost entirely dominated by variance — you could be a skilled +EV bettor on a losing run or a −EV bettor on a lucky winning run. At 200–300 bets, the signal starts to emerge. At 1,000+ bets, your P&L will closely reflect your true EV. Professional bettors typically require many hundreds of bets in a specific market before drawing strong conclusions about their edge.
Higher Odds Bets Have More Variance
Betting on 5.00 underdogs means each bet wins only 20% of the time. Even with +EV, you'll routinely face runs of 10+ losers. The expected profit per bet may be substantial, but the standard deviation (spread of outcomes) is enormous. Higher-odds +EV betting requires a larger bankroll and more patience before results reflect EV. Lower-odds +EV betting has less variance but smaller edge per bet.
Bankroll Management Is EV's Partner
Having +EV bets is not enough if your bankroll is wiped out by a losing run before the edge materialises. Proper bankroll management — staking a consistent small percentage of your total bankroll per bet — ensures you survive the variance to reach the long run where EV gets to express itself. See our Bankroll Management guide for staking strategies in detail.
Stop evaluating bets by whether they won or lost. Start evaluating bets by whether they were +EV or −EV at the time of placing. A +EV bet that loses was a good decision with a bad outcome. A −EV bet that wins was a bad decision with a good outcome. In the long run, good decisions accumulate profits and bad decisions accumulate losses — regardless of short-term results.
Common Questions
Your probability estimates are calibrated when your actual results match your predicted probabilities over a large sample. If you rate outcomes as 60% probability and they actually win 60% of the time across many bets, your model is well-calibrated. This requires careful record-keeping of your estimates and outcomes. In practice, calibration is hard to assess below 200–300 bets per probability bucket. An easier benchmark is consistently beating the closing line — if you systematically get better prices than the final odds before an event starts, your probability estimates are likely better than the bookmaker's on average.
Yes — in several ways that don't require building a statistical model. Line shopping (always taking the best available price) directly improves EV without any probability estimation. Betting exchanges remove the bookmaker margin entirely, improving EV on every bet by the amount of the margin. Taking advantage of genuine odds boosts (promotional prices that exceed market value) gives explicit +EV opportunities. And specialised knowledge of a niche market — knowing a particular league, sport or competition better than the bookmaker's traders — can generate +EV without formal modelling. Models help, but they're not the only route.
They are the same concept described differently. "Value betting" means finding bets where the odds are higher than the true probability warrants — which is precisely the definition of positive EV. A "value bet" is a +EV bet. The term "value betting" is more commonly used in UK and European betting circles; "EV betting" or "+EV betting" is more common in US and professional contexts. Both refer to identifying and placing bets where the price exceeds the true probability — the mathematical foundation of profitable betting.
No — and this is important. In a market with a bookmaker margin, the sum of implied probabilities exceeds 100%. This means, on average, bets in that market are collectively negative EV. Some individual selections may be mispriced positively, but the market as a whole takes money from bettors. In a perfectly efficient market, no bet would be +EV. Real sports betting markets are imperfectly efficient — particularly in less liquid markets and around breaking information — creating selective +EV opportunities that informed bettors can identify.
Most bettors lose because they consistently place −EV bets — not because +EV opportunities don't exist. The reasons are well-documented: many bettors bet on emotion or team loyalty rather than probability; they bet on popular markets that are efficiently priced with margin built in; they chase losses with larger stakes; they don't track their results seriously; and they don't compare odds across bookmakers. Finding +EV bets requires discipline, a systematic process, and a willingness to bet differently from the recreational crowd. It's achievable, but it requires deliberate effort — most bettors simply don't apply it.
Every extra tick of odds on a +EV bet increases your expected return. Use our live odds comparison to ensure you're always getting the best available price — the simplest way to maximise EV on every bet you place.
View Live Odds Comparison →