Are You Running Bad or Playing Bad? Understanding Variance in Poker

The question every poker player asks

You've had a bad month. Down 15 buy-ins. The question burns: am I running bad, or am I playing bad?

The honest answer is usually "some of both." But poker math can actually help you separate the two — and that separation matters more than most players realize.

What variance actually is

Variance is the statistical measure of how far your results spread from your expected value (EV). In poker terms: it's the gap between what you "should" have won and what you actually won.

A player winning at 5bb/100 doesn't win 5bb every 100 hands. Some stretches they win 20bb/100. Some stretches they lose 15bb/100. Over thousands of hands, these swings average out toward the true win rate. But "thousands of hands" means more than people think.

The numbers that matter

Let's put some real numbers on it. For a typical NL200 6-max player:

• Standard deviation: roughly 75-100bb/100 hands (depending on play style)
• Win rate: let's say 5bb/100

With a standard deviation of 85bb/100 and a win rate of 5bb/100, here's what's possible over different sample sizes:

After 10,000 hands: Your results could reasonably range from -45bb/100 to +55bb/100. That's right — a winning player can appear to be a massive loser over 10,000 hands and it's statistically normal.

After 50,000 hands: The range narrows to roughly -7bb/100 to +17bb/100. You're starting to see signal through noise, but it's still noisy.

After 200,000 hands: Now we're talking. The range tightens to about 0bb/100 to +10bb/100. Your true win rate is emerging.

The takeaway: if you've played fewer than 50,000 hands at a stake, your results tell you very little about your actual ability at that level.

Standard deviations and what they mean

When someone says "I'm running 2 sigma below EV," here's what they mean:

• Within 1 standard deviation (1σ): 68% of outcomes fall here. This is "normal" variance. If your results are within 1σ of your expected value, nothing unusual is happening.
• Between 1σ and 2σ: 27% of outcomes. You're running somewhat hot or cold, but it's not rare. Happens to everyone regularly.
• Beyond 2σ: 5% of outcomes. This is genuinely unusual. If you're beyond 2σ in either direction over a meaningful sample, something noteworthy is happening — either extreme luck or a win rate that's different from what you assumed.
• Beyond 3σ: Less than 0.3% of outcomes. This is the "something is probably wrong with my assumptions" territory. Either you're incredibly unlucky, or your actual win rate is significantly different from what you think it is.

How to check where you stand

The basic calculation is straightforward:

1. Estimate your win rate in bb/100 (use your best guess or your lifetime results at the stake)
2. Know your sample size in hands played
3. Calculate your expected profit: win rate × (hands / 100)
4. Compare to your actual profit
5. Calculate how many standard deviations away you are: (actual - expected) / (standard deviation × √(hands / 100))

The result tells you how much of your outcome is explained by variance vs. your underlying win rate.

We built a variance calculator right into the StackTrack landing page — try it with your numbers and see where you fall on the bell curve.

Why this matters for your game

Understanding variance changes how you think about poker:

Stop results-oriented thinking

If you lost 10 buy-ins this month but you're within 1σ of your expected value, your results are normal. You don't need to overhaul your game — you need to keep playing. Conversely, if you won 20 buy-ins but you're running 2σ above EV, don't assume you've suddenly cracked the game.

Make better bankroll decisions

Variance is why bankroll management exists. If you understand the realistic range of outcomes, you can set buy-in requirements that protect you through normal downswings instead of panicking when they happen.

Know when to investigate

If your results are more than 2σ below your expected win rate over 50,000+ hands, it might not be variance. Your game may have leaks, the player pool may have gotten tougher, or your estimated win rate may have been too optimistic. That's when it's time to review, study, and adjust.

Manage tilt

Knowing that your downswing is statistically normal doesn't make it feel good, but it does make it easier to stay disciplined. "I'm running 1.5σ below EV" is more useful than "I'm running like garbage and everything is broken." One is a data point. The other is tilt fuel.

The bottom line

Variance is not your enemy. It's a feature of the game that creates the conditions for your edge to exist. If poker had no variance, bad players would quit immediately and there'd be no money to win.

Your job is to understand it, plan for it, and keep playing through it. Track your sessions, check your numbers against expected variance, and let the math tell you when to worry — and when to just keep grinding.