Poker Variance: Understanding and Handling Swings
By the FoldLess team · · 5 min read
Poker variance is the gap between your actual results and the results your skill level should produce on average. It is measured by standard deviation, which sits around 80 to 100 bb/100 in 6-max cash games. In practical terms: a winning player at 5 bb/100 still has roughly a 29% chance of being in the red after 10,000 hands.
That number is surprising, and it explains why so many players misjudge their own level. Understanding variance won't make it go away, but it does let you tell a bad run apart from a real problem in your game.
What is poker variance?
In statistics, variance measures how spread out a set of values is around its average. In poker, the average is your win rate (bb/100 in cash games, ROI in tournaments) and the spread is your swings.
Two simple examples:
- AA versus KK all-in preflop wins about 82% of the time. So you lose that hand close to one time in 5 without making any mistake.
- AK versus QQ is a coin flip at roughly 43% to 57%. Over 10 of these matchups, losing 7 or 8 is still a common outcome.
Variance is thousands of these situations piling up. It cuts both ways: an upswing can hide leaks just as surely as a downswing can make you doubt a solid game.
Standard deviation and win rate: the pair that matters
Standard deviation is the square root of variance, expressed in the same unit as your winnings. The higher it is relative to your win rate, the noisier your short-term results. An aggressive style with lots of big pots raises your standard deviation; a high win rate reduces its relative impact.
How many hands do you need to know your true skill level?
Far more than most people think. Take a 6-max cash game player with a true win rate of 5 bb/100 and a standard deviation of 90 bb/100. Here is what the math gives for different sample sizes:
| Sample | Expected winnings | 95% range | Risk of being a loser |
|---|---|---|---|
| 10,000 hands | +500 bb | −1,300 to +2,300 bb | ≈ 29% |
| 50,000 hands | +2,500 bb | −1,500 to +6,500 bb | ≈ 11% |
| 100,000 hands | +5,000 bb | −700 to +10,700 bb | ≈ 4% |
So after 100,000 hands, this winning player can still show an observed win rate anywhere between −0.7 and +10.7 bb/100. That's the law of large numbers: the observed average converges toward the true average, but slowly.
The practical takeaway: a graph over 5,000 hands says almost nothing about your skill level. It does, however, tell you a great deal about your decisions, provided you review them one by one.
Variance by format
Not every format exposes you to the same swings. The more your winnings are concentrated in a few rare big scores, the higher the variance.
| Format | Variance level | Main source | Commonly recommended bankroll |
|---|---|---|---|
| 6-max cash game | Moderate | Big all-in pots, coolers | 30 to 50 buy-ins |
| Spins / Expressos | High | Hyper-turbo format, multiplier draw | 100 to 150 buy-ins |
| MTTs | Very high | Winnings concentrated at final tables | 200 to 300 buy-ins |
In MTTs, a winning player can go several hundred tournaments without a meaningful score. In Spins, short stacks and the fast structure mean far more preflop all-ins with close to 50% equity. These bankroll ranges are common guidelines, to be adjusted to your win rate and your risk tolerance.
Downswing or leak: how do you tell the difference?
This is the key question, and raw results can't answer it. You have to look at the quality of your decisions.
Signs of a "clean" downswing
- Your all-ins went in with favorable equity, but the results aren't following.
- You're running into unavoidable coolers (set over set, KK into AA).
- Your stats are stable compared with your winning stretches.
Signs of a real problem in your game
- Your stats are drifting: you're opening wider, calling more on the river, 3-betting less.
- The same losing spots keep coming up, such as defending your blinds too wide.
- You make different decisions after a big loss, a sign that tilt has set in.
The reliable method is to review your hands away from the tables, with a clear head, asking yourself "would I have played this hand the same way if I were winning?". Comparing your opens against reference preflop ranges gives you a first objective filter.
How to limit the impact of variance
You can't eliminate variance, but you can reduce the damage.
- Size your bankroll properly: it's the only real shock absorber. Moving down in stakes after losing 10 to 15 buy-ins isn't an admission of failure, it's risk management.
- Judge decisions, not results: a good call that loses is still a good call.
- Set session limits: a 3 buy-in stop-loss, for example, mostly protects your ability to think clearly.
- Increase your edge: the higher your win rate, the shorter and shallower your downswings.
- Talk with other players: posting your questionable hands on the forum keeps you from getting trapped in your own bias.
One honest note: no method guarantees winnings. If poker is weighing on your budget or your state of mind, taking a break is the best decision. Gambling is for adults aged 18 and over only.
Key takeaways
- Poker variance is the normal gap between actual and expected results; it works in both directions.
- With a 5 bb/100 win rate, you're still losing about 29% of the time over 10,000 hands and 4% over 100,000.
- MTTs and Spins are significantly higher variance than cash games and call for a deeper bankroll.
- Only analyzing your decisions can separate a downswing from a leak.
Conclusion
Variance is neither an excuse nor a curse: it's the background noise of poker. The only lever you control is the quality of your decisions, and that work happens after the session. With FoldLess, review your hands post-session, spot your leaks and track your bankroll to find out whether your swings are coming from the cards or from your game.
FAQ
What is poker variance?
It's the gap between your actual results and your expected results given your skill level. It is measured by standard deviation, typically around 80 to 100 bb/100 in 6-max cash games, and it explains short-term swings.
How long can a downswing last?
There is no hard limit. A winning player at 5 bb/100 still has roughly an 11% chance of being a loser over 50,000 hands, and losing stretches of several hundred tournaments do happen in MTTs.
How many hands do you need to know your true win rate?
At least 100,000 cash game hands for a usable estimate, and even then the margin of error is still about 5 to 6 bb/100 either way. Below that, you're better off judging your decisions than your results.
How do I know if I'm on a downswing or just playing badly?
Review your hands away from the tables: if your all-ins are going in with favorable equity and your stats are stable, it's variance. If your stats are drifting or the same losing spots keep coming up, it's a leak.
Can you reduce variance in poker?
You can't eliminate it, but you can limit its impact: a bankroll suited to your format, session limits, moving down in stakes when needed, and working on your game to raise your win rate, which shortens downswings.
Review your hands after each session, spot your leaks and track your bankroll with FoldLess.
Download FoldLess