Expected value, almost always shortened to EV, is the average result of a decision if it were repeated many times over. It is calculated by weighing each possible outcome by how likely it is and how much it wins or loses, then adding those pieces together. A play with positive expected value gains money in the long run; one with negative expected value bleeds it. The aim of strong poker is to keep choosing the positive-EV option, hand after hand, and let the average take care of the rest.
01+EV and -EV decisions
Every decision at the table can be sorted, at least in principle, into one of two camps. A positive-EV decision is one that makes money on average across many repetitions; a negative-EV decision loses money on average. Skill in poker is largely the ability to tell which is which under pressure, with incomplete information, and to keep picking the positive side even when the immediate result is unpleasant.
The crucial point is that the label attaches to the decision, not to the outcome. A call can be clearly positive in EV and still lose the pot, and a reckless gamble can be negative in EV and still win it. Judging plays by their results rather than their expected value is one of the most common and costly mistakes a player can make, because it rewards luck and punishes good process.
The jagged red line is what actually happens hand to hand - it dips below break-even for long stretches. The smooth green line is the expected value of the same correct decision: repeat it enough times and the result is pulled up toward it.
| Situation | Short-term result | EV judgment |
|---|---|---|
| You get it in ahead and lose | Bad outcome | Good decision |
| You call at the wrong price and win | Good outcome | Bad decision |
| You fold when the pot odds are poor | No win that hand | Good discipline |
| You chase without equity, often | Occasional wins | Negative EV over time |
02A worked pot-odds and equity example
The clearest place to see EV at work is a draw facing a bet. Suppose a player is on a flush draw and believes they will complete it and win about one time in five - an equity of roughly twenty percent. The opponent bets, making the total pot four times the size of the call required. The decision comes down to comparing equity with the price being offered.
- Equity. Around twenty percent to hit and win - the share of the time the call pays off.
- Pot odds. Calling one unit to win four means the call needs to succeed only one time in five to break even.
- The verdict. With equity at or above the price the pot is offering, the call carries positive expected value; below it, the call is negative-EV and should be a fold.
This is the engine behind most technical poker decisions: estimate the equity, compare it with the price, and take the play whose expected value is positive. The arithmetic does not promise the hand will be won - it promises that repeating the same correct call will profit over a large enough sample.
A simple framing: pot odds tell a player the price they are being offered, and equity tells them how often the hand will come good. When the equity beats the price, the call is profitable in the long run - whatever the next card happens to be.
Pot odds and equity are the headline test, but a decision earns positive EV from a few things working together:
- Enough equity for the price. Your share of the pot has to beat the odds you are being offered to call.
- Correct pot odds. The size of the bet relative to the pot sets the break-even threshold you measure against.
- Fold equity when betting. The chance an opponent folds adds value to a bet or raise on top of your raw equity.
- Opponent mistakes in the line. Value comes from the errors a play induces later in the hand, not just this street.
- Long-run repeatability. A choice is only truly +EV if it would profit across many identical spots, not once.
- Avoiding result-based thinking. Judge the line by its expected value, never by how the last card happened to fall.
03EV versus actual results over time
Expected value describes the long-run average, and the long run is much longer than it feels. Over a single hand or a single session, actual results scatter widely around their EV, which is why a stretch of perfect decisions can still lose and a stretch of poor ones can still win. That scatter is the subject of poker variance, and it is the reason short-term outcomes are such an unreliable guide to decision quality.
Over a large enough sample, though, results converge on expected value, and a long chain of positive-EV decisions reveals itself as a genuine edge. That convergence is precisely what a measured poker win rate captures, and it is the mechanism behind the broader question of whether poker is profitable at all. Play the positive-EV option, give it enough repetitions, and the average does the rest.
04Why good decisions can still lose
It is worth saying plainly, because it is the hardest part of the game to internalise: a perfectly correct, clearly +EV decision can and will lose, sometimes for a long time. None of that makes the decision wrong.
- Poker outcomes are noisy. Chance decides each individual hand, so the right play often lands on the losing side of its average.
- One trial says almost nothing. A single result can never confirm or refute the quality of a decision.
- All-in outcomes don't prove a line. Getting it in good and losing is the system working as intended, not a mistake.
- Variance punishes strong play short-term. Even a big edge can sit underwater for thousands of hands before it shows.
The discipline is to keep making the +EV choice and let the sample grow. The shape of those swings - how deep and how long they run - is exactly what poker variance measures.