Poker stats, and what a good one looks like
There are about forty numbers in a poker tracker. A handful will change how you play, and the rest are slices of those. This is the handful, in the order worth learning them.
Trackers show you everything they can compute, which is the wrong default. Most of those numbers are slices of the same handful of decisions, and staring at forty of them is a good way to change nothing.
Six cover almost all of it. Learn them in this order, because each one only makes sense once you have the one before it.
| Stat | What it answers | Reliable after |
|---|---|---|
| VPIP | How many hands you play | ~1,000 hands |
| PFR | How many you raise | ~1,000 hands |
| The gap between them | How many you call with | ~1,000 hands |
| WWSF | How often you win after the flop | ~5,000 hands |
| WTSD | How often you reach showdown | ~5,000 hands |
| Red line | Where your money comes from | ~20,000 hands |
Once those make sense, four more are worth having. Each answers a narrower question.
| Stat | What it answers |
|---|---|
| C-bet | Whether you follow through on the flop, and whether you should |
| Aggression factor | How you behave after the flop, with a formula worth knowing the flaw in |
| 3-bet percentage | How often you re-raise before the flop, seat by seat |
| All-in EV | Whether a bad stretch was luck, or was not |
What a winning player looks like
| Stat | Six-handed | Full ring, nine seats |
|---|---|---|
| VPIP | 22 to 26 | 15 to 20 |
| PFR | 18 to 22 | 12 to 17 |
| Gap between them | under 5 | under 5 |
| WTSD | 24 to 28 | 22 to 26 |
| W$SD | 50 to 54 | 50 to 54 |
Sitting inside these ranges doesn't make you a winner and sitting outside them doesn't make you a loser. What they're good for is spotting that you're a long way out, which is worth a look.
The three readings that find real leaks
A wide gap between VPIP and PFR. Above ten points means you're entering a lot of pots by calling, and calling has only one way to win. This is the most common shape among losing players by a distance.
High WTSD with low W$SD. You're paying to be shown the winner. Almost always river calls with a hand that beats a bluff, against someone who wasn't bluffing.
A flat line across the seats. Every one of these numbers should climb steeply from the first seat to the button. If yours doesn't, you're playing the same way from everywhere, and position is the largest free edge in the game.
What to ignore for now
Trackers will happily show you fold-to-flop-check-raise-in-3bet-pots. That number needs tens of thousands of hands before it means anything, and by then you'll have found bigger problems. 4-bet percentage is the clearest example: it comes up about once in 240 hands, so most players will never have the sample to read it.
The same goes for opponent stats after fifty hands. A read of 30 at that sample is consistent with a true number anywhere from 20 to 45. You know almost nothing, and acting on it is worse than acting on nothing.
The guides
- 01VPIP in Poker: What It Means and What a Good Number Looks LikeVPIP is how often you put money in before the flop by choice. What it counts, what a winning player's number looks like at your table size, and what the gap to your PFR says about your game.5 min
- 02PFR in Poker: Preflop Raise Percentage, and What Yours Should BePFR is how often you raise before the flop. What it counts, what winning players show at each table size, how it should climb across the seats, and why it matters more than VPIP.5 min
- 03WTSD in Poker: Went to Showdown, and What a Good Number IsWTSD is how often you reach showdown after seeing a flop. What counts, what winning players show, and how to read it next to W$SD to tell a calling station from a player who folds too much.5 min
- 04Red Line Poker: What Your Non-Showdown Winnings Actually MeanYour red line is falling. Before you start bluffing more, work out whether it's even unusual. What the line counts, why the blinds drag it down for everyone, and four checks to find the real leak.5 min
- 05C-Bet in Poker: What the Stat Means and Why One Number LiesContinuation betting, what your c-bet percentage counts, and why a single figure can't tell you whether you're doing it well. Split by board, position and how many players, with the numbers that matter.5 min
- 06Aggression Factor in Poker: What AF Misses, and Why AFq Beats ItHow aggression factor is calculated, the blind spot built into the formula, and why aggression frequency is the number to read instead. With the figures winning players show on each street.4 min
- 07WWSF in Poker: Won When Saw Flop, and Why It Beats Every Other Postflop StatWhat WWSF counts, the number winning players show, and why one outcome stat tells you more about your postflop game than any measure of how you behaved.4 min
- 08All-In EV in Poker: What the Adjusted Line Really ShowsWhat the all-in adjusted line on your graph measures, the small slice of luck it actually covers, and why the gap between it and your winnings explains less than most players think.4 min
- 093-Bet Percentage: What Yours Should Be, and Why the Old Numbers Are WrongWhat a 3-bet percentage counts, why the figure that was correct a decade ago is now too tight, the spread across seats, and the companion stat that costs more than this one.4 min
- 104-Bet in Poker: What It Means, and Why Your Stat Is Probably NoiseWhat a 4-bet is, what a cold 4-bet is, the frequencies winning players show, and the sample size problem that makes this the least readable number in your database.4 min
Reading is the slow half. Zayrion Analytics is the other half.
Frequently asked questions
What are good poker stats?
At a six-handed cash table: VPIP around 22 to 26, PFR around 18 to 22, the gap between them under five, WTSD around 24 to 28 and W$SD around 50 to 54. Full ring runs tighter on the first two. Treat these as the range most winners fall in rather than a target to hit.
Which poker stats should I learn first?
VPIP and PFR, because they're the fastest to become meaningful and they describe the decisions you make most often. Once those two look sane, WTSD is the next one, since it catches the habit that costs most money at low stakes.
How many poker stats do I actually need?
Six is plenty and three is enough to start. Trackers show dozens because they can, and most of them are slices of the same few decisions. A player who properly understands VPIP, PFR and WTSD is ahead of one who glances at forty numbers.
Which stat finds the most money at low stakes?
The gap between VPIP and PFR, followed by how often you fold to a flop bet. Both point at the same habit: entering pots by calling and then giving up on them. That pattern costs more at small stakes than anything happening on the river.
Do I need a HUD to use these stats?
No. A heads-up display shows stats on the table while you play, and several sites ban it. Every number here can be read after the session from your own hand histories, which is allowed everywhere and is the better way to study anyway.
How many hands before my stats mean anything?
It depends on how often the decision comes up. VPIP and PFR settle in about a thousand hands. WTSD needs a few thousand. The red line needs around twenty thousand, and your win rate needs more than most players ever play.
Educational only. No real-money play. 18+. On Google Play now; iPhone coming soon.