Your results so far
What your tracker currently shows.
Your total sample. Live players: ~25–30 hands an hour.
~100 is typical for online 6-max No-Limit Hold’em.
The verdict
Are you a proven winner?
Not proven yet
Your 95% range still crosses zero, so the data can’t yet say you’re a winner — your true rate could plausibly be a loss. This is exactly the gap most trackers paper over.
- 95% range for your true rate−3.8 to +13.8
- Chance you’re a long-term winner87%
- Hands to prove it at this rate~154,000
“Proven” here means a 95% confidence interval that clears zero — the standard bar for calling a result significant. It’s a floor, not a ceiling: clearing it says you’re a winner, not by how much.
How this is calculated
Your win rate is an average, so it carries a margin of error of SE = SD / √(hands / 100). We build the 95% confidence interval as rate ± 1.96 × SE and call you a proven winner only when its lower bound is above zero (a proven loser when its upper bound is below zero). The “chance you’re a long-term winner” is the normal probability that your true rate exceeds zero, Φ(rate / SE).
The hands-to-prove figure solves that same inequality for the sample size: hands = 100 × (1.96 × SD / rate)². It assumes your observed rate holds — if your true edge is smaller, it’ll take more. It all assumes roughly normal results and a known standard deviation, so treat it as an honest guide, not gospel.
Winning on the graph vs proven a winner
Almost everyone who’s up on the year feels like a winning player. But “up” and “proven” are different claims. Over a small sample, a true break-even player is up about as often as they’re down — so a positive graph is the expected outcome of luck, not evidence against it. The only way to separate the two is to ask whether your edge is large enough, relative to your swings, to rule luck out.
That’s exactly what the test above does. It checks whether your 95% confidence range has cleared zero. Until it does, the honest answer to “am I a winner?” is not proven yet — no matter how good the headline looks.
For a typical 2.5 bb/100 winner with a standard deviation of 100, proving it at 95% takes well over half a million hands. That’s not a flaw in your play — it’s how much luck hides in poker. Read why the number is so large.
Let StackWise tell you the moment it’s real.
StackWise tracks the confidence on your win rate as your sample grows and only calls it once the data clears the bar — no flattering green numbers in the meantime. Free to start, no account.
Sources & method
A two-sided 95% confidence interval on the mean, rate ± 1.96 × SD / √(hands ÷ 100), with significance when the interval clears zero. Sample-size reasoning from Poker Trainer, SplitSuit / Sweeney, and Primedope.