VPIP, PFR, 3-bet: the six numbers that describe your game, and how to read them

Poker statistics explained the way most guides skip: six numbers, read together, tell you far more than any one of them alone. VPIP and PFR get most of the attention because they load first on a HUD, but 3-bet frequency, fold-to-3-bet, went-to-showdown percentage and bb/100 fill in what those two leave out. None of these numbers requires a subscription or a live read. They all come out of a hand history you already have.

The goal here is not to memorize a "correct" number for each stat. It is to know what a number outside the normal range is actually telling you, and which of the six to check first when a session felt off.

What VPIP and PFR actually tell you about your own game

VPIP stands for Voluntarily Put money In Pot: the percentage of hands where you called, bet or raised before the flop, excluding forced blinds. PFR, Preflop Raise, is the percentage of hands where you entered with a raise rather than a call. Neither number means much alone. Read together, they describe your preflop style in one glance.

As poker coach Chipy explains, this pairing digitizes something players used to do by eye: "we generally created four broad categories of play" long before tracking software existed, sorting opponents into tight or loose, passive or aggressive. A high VPIP with a low PFR is the modern equivalent of what players used to call a calling station: loose entries, few raises. A high VPIP with a high PFR is a LAG (loose-aggressive). A low VPIP with a high PFR is the classic TAG (tight-aggressive) shape most winning cash-game regulars aim for.

Benchmark numbers exist, and they come with real ranges rather than a single target. PokerCopilot puts a sensible full-ring VPIP between 15% and 20% for a player prioritizing tight, selective hand selection at micro stakes. Pokerati's March 2026 breakdown gives a wider window for six-max cash regulars, typically 20% to 28% depending on position awareness and table size. Both sources agree on the direction: full ring sits tighter than six-max, and tournament VPIP usually starts tighter than cash before loosening as blinds climb.

What counts as a "voluntary" pot in VPIP?

Only preflop action counts, and only action you chose. Posting a blind or an ante does not count as voluntary, even though the money is in the pot. Calling, betting or raising before the flop does, regardless of what happens afterward. This is why a player can have a low VPIP and still see plenty of flops from the big blind: those are closed-out blind hands, not voluntary entries, and a stat that blends them with your open-raises from under the gun will always undersell how tight your real entries are. Reviewing opening ranges by position directly from your hand history fixes this blind spot that a single blended VPIP cannot.

3-bet percentage and fold-to-3-bet: reading your aggression

A VPIP and PFR pair already separates loose from tight and passive from aggressive. 3-bet percentage adds a second layer: how often you re-raise preflop when someone else has already opened. Fold-to-3-bet percentage is the mirror image, showing how often you fold your own open when someone re-raises you. Together they describe the aggression that VPIP and PFR alone cannot capture, since both of those only look at first action, not at how a hand escalates.

A 3-bet frequency that sits noticeably above your VPIP relative to how often you get the chance is a signal you are re-raising wider than your calling range, which is the standard modern approach: fewer flats, more 3-bets or folds. A fold-to-3-bet number above roughly 60% across a large sample tends to mean your opens are getting exploited, because observant opponents will simply re-raise you more once they notice you fold too easily. A leak this specific is worth isolating before it costs another session, which is exactly the kind of number a full session review built to find leaks is designed to surface.

What's a reasonable 3-bet range from each seat?

There is no single correct percentage because position changes the math completely: a 3-bet from the big blind against a late-position open faces a very different price and a very different opponent range than a 3-bet from under the gun against a button open. What stays constant is the relationship between the number and your fold-to-3-bet: if you 3-bet rarely and fold to 3-bets often, opponents can profitably attack every pot you open. If you 3-bet often and never fold to one, you are paying to find out who has it every single time. Reading the two together, hand by hand, from your own history beats memorizing a single target percentage for either one.

WTSD and bb/100: grading the outcome, not just the style

VPIP, PFR, 3-bet and fold-to-3-bet describe how you play. Went-To-Showdown percentage (WTSD) and bb/100 describe what happened as a result. WTSD is the share of hands you saw through to showdown once you reached the river; bb/100 is your win rate expressed in big blinds won per 100 hands, the standard unit for comparing results across different stakes.

A WTSD that runs high alongside a passive VPIP/PFR profile usually means calling down too often with hands that cannot win at showdown. A WTSD that runs low can mean disciplined folding, or it can mean over-folding profitable rivers, and the two look identical in the raw number. The only way to tell them apart is to open the actual hands and check what you folded, which is a review task, not a stat-reading one.

bb/100 is the number every player wants to trust first, and the one that deserves the most patience. A short stretch of a few thousand hands carries enormous variance: a run of missed flushes or cracked aces can swing a genuinely winning player's bb/100 deep into negative territory, and a heater can do the opposite to a losing one. The math behind how many hands it actually takes for a win rate to separate from noise is worked out with real numbers here; the short version is that a few thousand hands is closer to a coin flip's worth of information than a verdict on your game.

Reading the six numbers together

Any one of these six numbers, read alone, invites the wrong conclusion. A low VPIP could mean disciplined selection or could mean folding too much from the blinds. A high bb/100 over a small sample could mean a genuine edge or could mean a hot month. Reading them side by side, against typical six-max cash ranges, narrows down which explanation is more likely before you spend an hour second-guessing a session that was probably just variance.

Where these six numbers actually come from

All six numbers come from the same source: a hand history file, whether a HUD reads it live or a review tool reads it after the session. That distinction matters more than it looks. A HUD calculates VPIP, PFR and the rest from the same raw data a post-session export would use; the only difference is timing.

Some rooms, including GGPoker, WPT Global and CoinPoker, ban real-time HUDs outright, which forces the same six numbers to be computed after the fact, from an exported hand history, instead of live at the table. Studying without a HUD on those rooms is not a downgrade for the purpose of these stats: VPIP and bb/100 measured from a completed session are exactly as accurate as VPIP and bb/100 tracked live, because both draw on the identical hand data.

Where the reliability of the underlying data does matter is the platform generating it. A Reddit user who spent two years and 250,000+ hands analyzing Zynga Poker and Global Poker, two social and play-money apps rather than regulated real-money rooms, filed a federal lawsuit against Zynga over what they describe as hand-scripting, a case since compelled to arbitration. That investigation targets social apps specifically, not the regulated real-money rooms most cash-game and tournament players use, but it is a useful reminder that a stat is only as trustworthy as the hand history it was built from. On a regulated real-money room, your own exported hand history is the one dataset in this whole conversation you can verify yourself.

Sample size still governs everything here. PokerBankrollApp notes that pocket aces arrive roughly once every 221 hands, a small enough frequency that a single losing session tells you almost nothing about your actual aces win rate. Chris Moneymaker's 2003 WSOP run is the extreme version of this: a handful of hands can look like a story, but the six numbers in this article only become reliable once the sample is large enough to drown out that kind of noise. Research on data-driven decision-making broadly, tracked by firms including McKinsey, consistently finds that decisions built on a tracked, systematic dataset outperform decisions built on memory and impression alone; poker's version of that is simply reading enough hands before trusting the number they produce.

None of the six numbers changes anything mid-hand. They are review tools, not live aids, and reading them accurately means pulling up hands after the session, not trying to calculate them at the table.

The verdict: read the six together, trust none of them alone

VPIP and PFR set the frame: tight or loose, passive or aggressive. 3-bet and fold-to-3-bet fill in the aggression a first-action stat misses. WTSD explains what happened at showdown, and bb/100 grades the outcome, provided the sample is large enough to mean something. Reading any single one of these six numbers in isolation is how a player misdiagnoses a variance-driven downswing as a leak, or misses an actual leak because the overall win rate still looks fine.

A leak rarely shows up as one dramatic number. It shows up as two of these six numbers disagreeing with each other, and the only way to catch that is to look at all six from the same stretch of hands, then open the specific hands that explain the gap.