Free tool
Poker variance calculator
A win rate is an average, and you never play the average. This shows you twenty versions of the same year, played by a player with exactly your edge.
Your game
Your numbers, or the defaults if you do not know them yet.
2 to 8 is a real winner
80 to 100 for NLHE 6-max
Twenty versions of the same year
Twenty sample runs of exactly the game you described.
The straight line is your expectation. The shaded bands hold the middle 70% and 95% of outcomes. Every wandering line is one possible run of the exact same game, played by a player with exactly your edge.
Expected
5,000 bb
Chance of being down
3.9%
Unlucky 5% run
318.2 bb
Lucky 5% run
9,681.8 bb
Why good players quit games they were beating
Because a losing month is not evidence of anything. At a standard deviation of 90bb per 100 hands, the noise over any sample a human actually plays is larger than the edge. A winner sees losing stretches constantly, and a loser sees winning stretches constantly, and neither can tell which they are from the graph alone.
The practical consequence is that results are the worst available feedback on your play, and they are the only feedback most players ever look at.
What to look at instead
Decisions. Whether the call had the equity the price demanded, whether the bet size made sense for the range, whether the fold gave up too much. Those are checkable on a single hand, and they do not need 100,000 of them to mean something.
That is what a session debrief does: it reads the hands that decided a session and prices the mistakes in big blinds, so the feedback comes from what you did rather than from how the cards fell.
Common questions
How many hands do I need before my win rate means anything?
More than almost anyone plays before drawing conclusions. At a 90bb/100 standard deviation, a genuine 5bb/100 winner is still down after 10,000 hands roughly a third of the time, and after 100,000 hands roughly one time in twenty. Run those numbers above and the shape of the problem is obvious.
What standard deviation should I use?
Around 80 to 100bb/100 for 6-max no-limit hold'em, lower for full ring and tighter styles, higher for loose games and for pot-limit Omaha, where 130 and up is normal. Your tracker reports your own figure if you have one. If you do not, 90 is a fair default for NLHE.
Does a downswing mean I am playing badly?
It is not evidence either way on its own, which is exactly the problem. Variance at these sample sizes swamps the signal, so the only reliable way to tell a downswing from a leak is to look at the decisions rather than the results. That is a different job from this calculator.
Is this simulation or maths?
Both. The expectation, the bands and the chance of being down are computed from the normal approximation, exactly. The twenty wandering lines are simulated runs drawn from that same distribution, seeded on your inputs so the picture does not change every time you look at it.
Stop reading the graph. Read the decisions.
A debrief reads the session you actually played: the hands that decided it, reviewed street by street, and your leaks priced in big blinds. The first one is free and complete.
No card. No subscription. $3 a session after that, and credits never expire.