What is Expected Value (EV) in Sports Betting?

Expected value is the only number that tells you whether a bet is worth taking, independent of whether it wins or loses. Positive EV means the math is on your side. Negative EV means it is not. Over enough bets, EV is the only thing that matters.

Here is the formula, a worked example with real numbers, and why a bet can be the right call even when it loses.

The formula

Expected value on a sports bet is calculated as:

EV = (probability of winning x amount won) - (probability of losing x amount lost)

Both probabilities are expressed as decimals between 0 and 1. The win amount is what the bet pays in profit (not including the returned stake). The loss amount is the stake itself.

A simpler way to read it: EV is a weighted average of the two outcomes, weighted by how often each outcome happens in the long run.

Worked example at -110

Say you are looking at a standard -110 line. The implied probability the market is offering is 52.38%, meaning the sportsbook is pricing this side as if it wins 52.38% of the time. You think the true probability is 53%. That sounds tiny, but watch the math.

For a $100 bet at -110:

• If you win, you profit $90.91 (because $110 risked returns $100 plus your stake)

Wait, let me get the math right. -110 means you risk $110 to win $100. For a $100 stake at -110, you risk $100 and win $90.91. So:

• Win amount per $100 staked: $90.91
• Loss amount per $100 staked: $100

Now apply the formula with your 53% probability estimate:

EV = (0.53 x $90.91) - (0.47 x $100) EV = $48.18 - $47.00 EV = +$1.18 per $100 wagered

That is a 1.18% edge on every dollar bet. Over 1,000 such bets, you would expect to win roughly $1,180.

Why +EV bets still lose

Here is the part most casual bettors get wrong. A bet with positive expected value will still lose almost half the time at standard prices. Run the same -110 example: even with a 53% true win rate, you are losing 47 of every 100 bets. The wins simply cover the losses with a small surplus.

That surplus is the edge. It only becomes visible over a large sample. Over 50 bets, variance dominates, you can run 22-28 with positive EV and feel like the system is broken. Over 500 bets, the math starts to assert itself. Over 5,000 bets, your actual return converges on EV.

This is why bettors who chase short-term results blow up disciplined strategies. They take a +EV approach, run cold for two weeks, conclude it does not work, and switch to a -EV approach that happens to be running hot. The next month flips. They lose either way because they never let the long-run average play out.

Why -EV bets sometimes win

The same logic in reverse. A -EV bet (like most parlays at standard juice, most live in-game flyers, most longshot futures) wins often enough that bettors who take them get false confirmation. A 4-leg parlay with -8% EV will hit roughly 25% of the time at typical prices. One in four feels like the strategy works.

It does not. Over 100 such parlays, the bettor is down a meaningful share of their bankroll. The wins are visible and screenshot-worthy. The losses are quietly rebought tomorrow.

How to estimate true probability

EV math is only as good as your probability estimate. Three honest ways to anchor it:

1. Model the market. Build a simple regression on whatever inputs matter for the sport (rest, injuries, pace, weather). The output is a probability you can compare to the line.

2. Use base rates. Historical hit rates on similar situations (home dogs after a road loss, totals in cold-weather games) give you a starting point that is more honest than a gut number.

3. Triangulate against multiple books. If three sharp books agree the line should be -118 and your local book is at -105, the implied probability gap is real edge. The market itself is doing your estimation.

The trap is overconfidence. A bettor who pencils in 60% on every bet they like is not modeling, they are rationalizing. Discipline shows up in the willingness to mark a bet at 53% even when the team is your favorite.

Putting it together

Three checks before any bet:

1. What does the line imply the probability is?
2. What is your honest estimate of the true probability?
3. Is the gap large enough to clear breakeven plus a margin for error?

If all three resolve cleanly in your favor, the bet is +EV and worth taking. If any of them is sloppy, the bet is variance, not edge.

Over a thousand bets, EV is the only thing that survives. Win rate is variance. Streaks are variance. Bankroll changes in any single month are variance. EV is the trend underneath it all.

Every pick on the daily card is screened on calculated EV before it ships. The track record shows what that edge looks like over time.

Frequently asked questions

What does expected value mean in sports betting?

Expected value (EV) is the average amount you would win or lose per bet if you placed the same bet under the same conditions thousands of times. Positive EV means you profit on average; negative EV means you lose on average.

How do you calculate expected value on a bet?

EV = (probability of winning x amount won per win) - (probability of losing x amount lost per loss). Use decimal numbers for probabilities and dollar values for stakes and payouts.

Can a positive EV bet still lose?

Yes, on any single bet. EV is a long-run average. A bet with +5% EV will lose roughly 47% of the time at -110 lines. The edge only shows up over hundreds of similar bets.

How do I estimate true probability versus the market?

Build a model, use historical base rates, or compare lines across multiple sportsbooks to triangulate a fair price. The implied probability of the offered odds tells you the market's view; your job is to estimate something better.

Is positive expected value the same as winning the bet?

No. A positive EV bet has a higher payout than its true probability justifies. Whether it wins depends on the actual game outcome plus variance. You can have a 90% EV-positive bet and still lose because the underdog covered.