Why Your Poker App's Win Rate Is Probably Wrong

Your poker app is probably showing you a misleading number, and it is building it from your own data.

If it computes a win rate from the hands you chose to log, it is computing a win rate on a biased sample. We decided PokerInk would refuse to do that, even though the number would look professional.

Logged hands are a biased sample

Online, bb/100 works because the software logs every single hand dealt. The denominator is complete. Nothing about which hands enter the database depends on how interesting they were.

Live is the opposite. In a typical session you are dealt hundreds of hands and you will log a handful. Which handful? The big pots. The tough river spots. The bad beats. The hero calls. Nobody journals the 73 offsuit they folded under the gun, and nobody ever will.

That selection is not random, and it correlates directly with the thing you are trying to measure. Big logged pots mean big logged swings, in both directions. A win rate computed over those hands is a win rate of your most memorable moments. It might flatter you. It might terrify you. Either way it is not your win rate.

No amount of math fixes the denominator

This is not a precision problem you can average away. More sessions of selectively logged hands produce a bigger biased sample, not a less biased one. The error is structural: the denominator, hands played, is unknown, and the numerator was hand-picked by human memory under emotion.

A bb/100 printed over live logged hands is a polished-looking number with no statistical meaning behind it.

The honest denominator: the session

There is one thing a live player records completely and reliably: the session. You know what you bought in for, what you cashed out, and how long you played. Those three facts are a complete sample of your results, with nothing selected and nothing forgotten.

So that is what PokerInk computes money stats from:

• Dollars per hour, from buy-ins, cash-outs, and hours played.
• Percentage of sessions won, an honest look at how often you book a win.
• Cash P/L and tournament ROI, kept separate, because they are different games with different variance.

Your logged hands still matter enormously. They are where the coaching happens: leak detection, street-by-street review, and drills all run on the hands you record. Hands are for studying decisions. Sessions are for measuring money. PokerInk keeps those two jobs separate on purpose.

Small samples swing hard

The second half of honesty is sample size. Poker variance is large enough that a strong player can lose over a month and a weak player can win over one. A 30-session sample is closer to a weighted coin flip than a verdict.

That is why every money stat in PokerInk is captioned with the sample it was computed from. A $/hour over 12 sessions reads differently than the same number over 200 sessions, and the app says so instead of letting one number impersonate certainty.

What this buys you

Numbers you can act on. If your sessions-won percentage is fine but your $/hour is poor, your losing sessions are too big, which is a stop-loss and tilt conversation. If your leak profile says you are calling rivers too light, that is a study conversation. Honest inputs make the difference between those diagnoses visible.

The flattering fake number would have been easier to ship. It would also have made every downstream feature quietly wrong.

Track honest results from your next session

PokerInk is free to start on iPhone, Android, and the web. See pricing.