Expected Value Explained for Sports Bettors

Expected value quantifies whether a bet is worth making on average, over time -- not whether it will win. Here's the formula and a worked example.

Last updated 2026-09-17

Expected value (EV) is the single most important concept in value betting, and it's also one of the most commonly misunderstood. Getting it right changes how you evaluate every bet you place — not by predicting what will happen, but by quantifying whether a bet is worth making at all, on average, over time.

The basic formula

Expected value for a bet is calculated as:

EV = (Probability of winning × Amount won per win) − (Probability of losing × Amount staked)

If a bet has a 50% true probability of winning, pays out at odds implying only 45% (meaning the payout is larger than a "fair" 50/50 bet would offer), the bet has positive expected value — even though it will still lose roughly half the time.

A worked example

Suppose the true probability of an outcome is 50%, and a bookmaker offers decimal odds of 2.30 (implying roughly 43.5%). On a ₦10,000 stake:

• If you win (50% of the time): profit of ₦13,000 (₦23,000 payout minus ₦10,000 stake)
• If you lose (50% of the time): loss of ₦10,000

EV = (0.50 × 13,000) − (0.50 × 10,000) = 6,500 − 5,000 = +₦1,500

That positive EV means that, on average, repeating this exact bet many times would net you roughly ₦1,500 per ₦10,000 staked — even though any single instance is a coin flip that could go either way.

Why positive EV bets still lose, often

This is the part beginners struggle with emotionally: a bet with genuinely positive expected value can still lose more often than it wins, especially when the true probability is under 50%. A bet with a 30% true win probability and strong value can easily lose seven times out of ten — and still be the mathematically correct bet to have made, because the size of the win when it does hit more than compensates over enough repetitions.

Judging a single bet's quality by whether it won or lost is a common and costly mistake. The bet was either +EV or −EV at the time it was placed, based on the information available then — the outcome afterward doesn't retroactively change whether it was a good decision.

Why sample size matters so much

Expected value is a statement about averages over repetition, not a guarantee about any individual outcome. A handful of bets — even a few dozen — can easily show results that look nothing like the underlying EV, purely due to normal variance. This is why serious value bettors track results over hundreds of bets before drawing conclusions about whether their process is actually working, rather than judging based on a short losing (or winning) streak.

Applying this practically

Before placing a value bet, the useful question isn't "do I think this will win?" — it's "is the price being offered better than the true probability justifies?" If the answer is yes, and you can estimate that true probability reasonably reliably (typically by benchmarking against a sharp bookmaker's price), the bet is worth making regardless of your gut feeling about the specific match. Expected value is what turns value betting from guesswork into a repeatable, quantifiable process.