Expected Value (EV) in Poker, Explained Simply

By PokerCraft Lab Editorial · Published Mar 8, 2026

Expected value, or EV, is the single number that tells you whether a poker decision makes money or loses it over the long run. Master this one idea and you stop judging your play by whether you won the hand — and start judging it by whether the decision was correct.

What expected value actually means

EV is the average result of a decision if you could repeat it thousands of times. Some of those times you win, some you lose, but the weighted average lands on one number. If that number is positive, the play prints money in the long run. If it’s negative, it bleeds chips no matter how good last night felt.

The formula is plain arithmetic:

EV = (chance you win × amount you win) − (chance you lose × amount you lose)

That’s it. No advanced math, no software required at the table. The hard part isn’t the calculation — it’s training yourself to trust the number over the result.

A worked coin-flip example

Start outside poker so the logic is clean. Someone offers you a coin flip: heads, you win $10; tails, you lose $8. Should you take it?

• Win: 50% of the time, +$10
• Lose: 50% of the time, −$8

EV = (0.5 × $10) − (0.5 × $8) = $5 − $4 = +$1 per flip.

Each individual flip still pays you nothing or costs you $8. But the decision is worth +$1 every time you make it. Take that bet a thousand times and you’re up roughly a thousand dollars, even though you lost about half the flips along the way.

Turning a poker call into an EV calculation

Now the same logic at the table. The pot is $100 and your opponent shoves $50, so you must call $50 to win a pot that becomes $150. Suppose you estimate you’ll win this hand 40% of the time.

• Win: 40% of the time, you collect the $150 already out there → +$150
• Lose: 60% of the time, you lose your $50 call → −$50

EV = (0.40 × $150) − (0.60 × $50) = $60 − $30 = +$30.

A positive number means calling is correct. You should make this call every single time the spot comes up, regardless of what the river brings. This is the same engine that drives pot odds — pot odds tell you the break-even percentage you need, and EV tells you the dollar value of being above or below it.

Variance: why a correct call can still lose

Here’s the part that trips people up. That +$30 call will lose 60% of the time. Six out of ten times you make the right decision, you ship your stack to the other player.

That gap between the correct decision and the short-term result is variance — the natural swing of luck around your expected value. EV is the destination; variance is the bumpy road that eventually gets you there. Over one hand, variance dominates and EV is invisible. Over thousands of hands, variance averages out and EV is all that remains.

This is why a player can make flawless decisions and still have a losing week, while a reckless player can run hot and feel like a genius. Results over a small sample tell you almost nothing about decision quality.

Right call, wrong outcome

Imagine you call that shove with the math firmly in your favor and the river bricks. You lose the pot. Did you make a mistake?

No. You made a +$30 decision and got an unlucky result. If you fold “because it lost last time,” you’re letting variance overrule EV — and that’s how winning players slowly turn into losing ones. The discipline is to separate the two:

• A good decision is one with positive expected value.
• A good outcome is winning the pot.

You control the first. You do not control the second. Tilt, in large part, is the failure to keep these apart — punishing a correct play because the cards betrayed it.

Thinking in EV at the table

You won’t compute exact percentages mid-hand, and you don’t need to. What you need is the habit of asking: across all the times this exact situation happens, does this line make money or lose it? Estimate your equity, weigh it against the price, and commit to the positive-EV play even when it stings.

For deeper drilling — running thousands of trials so the long run becomes visible — equity tools like DEEPFOLD let you stress-test spots away from the felt. Bring the conclusions back to the table, trust the math over any single result, and over enough hands the variance fades and your edge is all that’s left.