Expected Value (EV) in Poker Explained

Expected value, or EV, is the average result a decision produces if you could make it over and over. A play with positive EV wins chips on average; a negative-EV play loses them, even on the hands it happens to win. That’s the whole definition — and it’s also the quiet engine behind every pot-odds call, bluff, and shove you’ll ever make.

The formula

EV adds up every possible outcome, each weighted by how often it happens:

EV = (chance of winning × amount won) − (chance of losing × amount lost)

More than two outcomes? Add a term for each. The result is one number, in chips or dollars: positive means the play profits on average, negative means it bleeds. Everything after this is just feeding it honest inputs.

Is this call worth it?

You flop a flush draw — 9 outs, about 36% equity to hit by the river. Villain bets $50 into a $100 pot, so you call $50. Hit and you win the $150 already out there; miss and you’re down your $50.

EV = (0.36 × $150) − (0.64 × $50) EV = $54 − $32 = +$22

The call earns +$22 on average. You’ll lose it nearly two times in three, and it’s still correct — folding here throws away twenty-two dollars every time you do it. This is precisely what pot odds measure: a quick way to see the +EV call without grinding the full arithmetic.

When a play can win two ways

EV really opens up once a decision has more than one path to winning. Shove $80 into a $100 pot with a draw worth 30% equity, against an opponent who folds half the time:

• They fold (50%): you take the $100 now → +$100.
• They call, you hit (50% × 30% = 15%): you win their $80 plus the $100 → +$180.
• They call, you miss (50% × 70% = 35%): you lose $80 → −$80.

EV = (0.50 × $100) + (0.15 × $180) + (0.35 × −$80) EV = $50 + $27 − $28 = +$49

A +$49 shove — more than double the passive call above — because folding equity hands you an entire extra winning branch. That’s the math driving aggressive preflop and GTO shoving ranges.

Why the right play still loses

EV is an average, and a single hand is a tiny, noisy sample of it. Make that +$22 call a hundred times and you’ll bank roughly $2,200 — but you could drop the first six in a row and feel like the math betrayed you. It didn’t. That swing is variance, and it’s baked into every edge you have.

The practical anchor is that folding is worth exactly $0 — no more won, no more lost. So every decision reduces to one comparison:

• Is calling worth more than $0? Then it beats folding.
• Is raising worth even more? Then raise instead.

You’re always choosing the highest-EV option among fold, call, and raise — and the biggest number wins, not the flashiest. A +$10 call beats a +$3 shove. And when every action is negative, the fold’s clean zero is the profitable play, which is why strong players fold more often than beginners expect.

Where EV goes wrong is almost always the inputs: overestimating how often you win, ignoring the extra branch that aggression adds, or miscounting your outs. Get the inputs honest and the framework does the rest — compute EV when you can, estimate it when you can’t, then trust volume to pay out the average at the Hold’em tables.