Variance

Variance in poker is the distance between what you earn and what you get. In the short run the luck of the cards decides your result far more strongly than the quality of your play. Good decisions guarantee no good evening, only a good year. Anyone who really feels that plays more calmly, and anyone who denies it gets broken by it sooner or later.

What variance actually is

Every poker decision has an expected value: what it earns on average if you take it a thousand times. One single hand, however, never pays out the average, but one of the possible outcomes. Go all in (betting everything) as an eighty percent favourite and you lose that hand one time in five, and sometimes three times in an evening. There was nothing wrong with your decision then, only with the outcome, and that distinction is the heart of this whole story. Variance is nothing more mysterious than that: the ordinary spread of outcomes around an average, magnified by the amounts at stake at the table. Compare it with a coin that lands on your side sixty percent of the time: over a thousand tosses you win almost certainly, over ten tosses you still lose regularly, and poker consists of endlessly many such runs of ten.

How big the effect really is

Bigger than practically anybody thinks. The swings per hundred hands for a normal cash player are many times larger than his margin, and that means even a clear winner can run around break even or down across tens of thousands of hands. For somebody who plays a few evenings a week that is easily a gloomy month or two, with nothing wrong with their game. The reverse holds too: a losing player can win for weeks and draw exactly the wrong conclusion from it. That immediately explains why poker is so stubbornly addictive and so stubbornly instructive at the same time: in the short run the game's feedback lies to everybody, winner and loser alike.

Why your brain underestimates this

People are built to see patterns, including where there is only noise. Three lost sessions feel like proof that you are doing something wrong, and one winning evening like proof that the new approach works. Add to that the pain of losing, which weighs more heavily psychologically than the pleasure of the same win, and you understand why a neutral month can feel like a disaster. Both conclusions are worthless on that sample. The danger is in what you do with it: anyone who mistakes noise for signal starts tinkering with a working strategy, chases losses or plays more scared than their cards. That is how variance becomes expensive after all, not through the cards but through your reaction to them. The remedy is simple to state and hard to keep up: draw conclusions from large numbers of hands and from the quality of your decisions, never from one week's balance.

Your weapons: buffer and behaviour

Against the cards themselves there is no weapon; against the consequences there is. The first is financial: a bankroll of a good twenty to thirty buy-ins absorbs normal swings without one bad week touching your game, as explained in bankroll management. The second is behaviour: judge yourself on decisions instead of results, and recognise the moment frustration takes the wheel before it costs money. Fixed session lengths, a short review after every playing evening and a hard stopping rule when emotion rises sound dutiful, and are exactly what separates winners from nearly winners. How to train that is in preventing tilt. Finally, bear in mind that variance is your ally too: without the chance of beating a better player for an evening, nobody would sit down against you.