Poker HUD Stats Explained: The Six Numbers That Matter

Six numbers do almost all the work: VPIP, PFR, 3-bet, fold to c-bet, aggression factor and went to showdown. Learn those and you can classify any regular inside a few orbits and pick the adjustment that beats them. Everything else on a default layout is decoration until the six are automatic. Each one names a specific leak, and each needs a different number of hands before you should believe it.

Start with six stats, not sixty

A fresh HUD install will happily show you thirty statistics per player. You cannot read thirty numbers while a timer runs, so you read none of them, and the whole thing becomes expensive wallpaper.

The fix is to strip the layout back to six, learn what each one is actually counting, then add more only when the six have become instant. It helps to understand what a HUD actually shows before you start trusting the display: it is a running tally of your opponent's past decisions, nothing more.

Stat	What it counts	Rough sample before it means much	Leak it exposes
VPIP	Hands played voluntarily preflop	200-300	Playing too many or too few hands
PFR	Hands raised preflop	300-500	Passive limping and flat-calling
3-bet	Reraises before the flop	1,000-2,000	No preflop pressure
Fold to c-bet	Flop folds facing the raiser's bet	1,000-2,000	Giving up too fast, or never folding
AF	Bets and raises divided by calls	1,000	Passivity after the flop
WTSD	Hands reaching showdown	2,000-3,000	Calling down far too light

VPIP and PFR: the shape of a player

VPIP is the share of hands where money goes in by choice — your blind posting does not count, but defending it does. PFR is the share where you raise. Together they describe how wide someone plays and how aggressively they play that width.

The interesting part is the distance between them. A 24/21 player raises nearly everything they enter with; a 38/6 player is limping and calling their way into pots and hoping to hit. Reading the gap between VPIP and PFR tells you more than either number alone.

Profile	Typical VPIP/PFR	How to play against them
Solid regular	24/20	Avoid marginal spots, attack their blinds selectively
Nit	15/12	Steal relentlessly, believe their raises
Calling station	40/7	Value bet thin, never bluff more than once
Loose-aggressive	32/27	Widen your calling range, let them barrel into you

Two practical rules follow. Against a large gap, bet your good hands and delete bluffs from your plan. Against a tight pair of numbers, open wider from the BTN and CO and fold to their aggression sooner.

3-bet: who is actually applying pressure

Overall 3-bet percentage tells you whether an opponent fights before the flop. Most winning 6-max regulars land in mid single figures. Under about 3% means they reraise premiums only, so when they do it you can fold QQ-and-worse-shaped hands and feel fine about it.

The companion number is fold to 3-bet. Above roughly 70% you can reraise their steals with hands that have no business getting there, because taking the pot down immediately is the whole plan. Below 45% you need a real hand, because you are getting called and playing a bloated pot out of position.

Watch the pairing rather than either figure alone. A player who opens the BTN wide and folds to three-quarters of 3-bets is handing you a pot every time you reraise, and at 5bb a go that adds up faster than any postflop edge you are likely to find against them.

Split the stat by position as soon as your sample allows it. A single overall 3-bet number blends loose BB defence with the tight reraises someone makes against UTG, and those two situations demand opposite responses. If your HUD supports positional popups, the BTN and BB figures are the two that change your decisions most often.

Fold to c-bet: the most exploitable number on the table

This is the stat with the fastest payback. Fold to flop c-bet says how often an opponent surrenders the flop after calling preflop.

The theoretical anchor is minimum defence frequency — MDF = pot ÷ (pot + bet) — which means against a two-thirds-pot bet a player should continue with around 60% of their range. Plenty of low-stakes players fold well past that.

If an opponent folds to more than about 65% of flop c-bets, bet every flop against them. The board texture barely matters; the fold does.

The reverse case is just as valuable. Against someone who folds under 40%, cut your bluffs entirely and bet only when you would be happy to hear a raise. Continuation betting a dry ace-high board with 72o into a player who never folds is how you donate 3bb at a time.

AF and WTSD: what happens after the flop

Aggression factor divides bets and raises by calls. Below 1.5 you are looking at a caller; above 3 you are looking at someone who bets and raises rather than calls, which changes how you read their turn action.

Went to showdown is the share of hands where a player who saw the flop reached the river with cards to show. Somewhere around a quarter to a third is normal. The pairing of the two is where the money is.

• High WTSD, low AF — a station. They call three streets with second pair. Value bet everything, bluff nothing.
• Low WTSD, high AF — a barreller who gives up when called. Call down lighter and let them bluff into you.
• High WTSD, high AF — a genuine problem. Tighten your ranges and take flops in position.
• Low WTSD, low AF — a fit-or-fold player. Bet flop and turn and take the pot.

These four boxes cover most of a low-stakes pool. The remaining information lives in your own memory, which is why opponent notes that pair with stats beat stats on their own: the HUD tells you how often, your notes tell you which hands.

Sample size: when a number becomes real

Every stat is a sample, and small samples lie confidently. VPIP updates every single hand, so it settles quickly. Fold to c-bet only updates when someone calls preflop and then faces a flop bet, which might be one hand in eight.

That ratio is the whole rule. The rarer the situation, the longer you wait before acting on the number. River stats are the extreme case: a fold-to-river-bet figure needs many thousands of hands before it stops jumping around by twenty points, which is why almost nobody should be making river decisions from one.

A player showing 0% fold to 3-bet over 12 hands has simply not folded to the two 3-bets they faced. Treating that as a read and reraising them with J5s is not exploitation, it is noise-chasing. Use a variance and sample size calculator on your own results and the point lands hard: if your own win rate needs 100,000 hands to confirm, an opponent's turn-fold percentage is not settled after 300.

The same maths explains why your own graph swings so violently, and why judging a strategy over a weekend is hopeless. Thinking in win rates in big blinds per 100 rather than session results keeps both problems in proportion.

Using stats without being used by them

Three habits separate players who profit from a HUD from players who just decorate their table.

1. Colour-code the extremes. You want the outliers to catch your eye without reading anything. Everyone in the middle is a regular and gets your default strategy.
2. Check stats before you act, not after. Glancing at fold to c-bet after you have already bet is a review exercise, not a decision.
3. Let a live read override a number. A stat is an average of the past. If someone has visibly tilted after losing a big pot, the last 30 seconds beat the last 3,000 hands.

The six numbers are not a strategy. They are a shortcut to the question that matters: which mistake is this opponent making, and which of my hands profits most from it. Answer that in the first orbit and you are already ahead of everyone reading thirty statistics and acting on none.