Poker Player Types: the Foundation of Every Exploitative Strategy

Exploitative poker fits in one sentence: you make money by playing differently against different opponents. Which requires knowing who you are actually up against. Without a classification, an adjustment is just a hunch, and a wrong hunch costs more than no adjustment at all.

This guide gives you the full framework: the two axes that define a profile, the six archetypes you will meet, the numeric thresholds that identify them, the sample size you need before trusting them, and above all the counter-strategy that prints against each one.

The two axes that classify everything

Forget the twenty stats available to you. A profile sits on two dimensions.

Axis 1: width. How many hands does this player enter the pot with? That is VPIP (Voluntarily Put money In Pot), the percentage of hands played voluntarily, excluding free blinds.

Axis 2: aggression. When they enter, do they take the initiative? That is PFR (Preflop Raise) before the flop, then AF or aggression frequency after it.

Cross the two and you get four quadrants, which practice splits into six archetypes.

The gap between VPIP and PFR is almost as informative as either stat alone. A 24/20 player (gap of 4) has an almost non-existent calling range: they raise or they fold. A 38/9 player (gap of 29) limps and calls everything: they are handing you money postflop.

The six archetypes, in numbers

Reference thresholds for 6-max cash. In full ring, subtract 4 to 6 points of VPIP. In tournaments, read them against stack depth.

The distinction between the calling station and the passive fish matters in practice: the passive fish still folds to pressure, the calling station does not. The first can be bluffed, the second never.

How many hands before you trust a stat

This is the quiet, expensive mistake: reading a 12% 3-bet over 25 hands and starting to 4-bet light. That stat does not exist yet.

Field rule: the rarer the situation a stat describes, the more hands it needs to be true. A fold-to-4-bet read over 500 hands may rest on three occurrences.

While you wait for volume, you can still classify: a player who limps twice in ten hands and calls a c-bet with second pair is already categorised, sample or no sample.

Counter-strategies, profile by profile

Against the nit

They only call with the goods, so their calls announce their hand.

• Steal their blinds relentlessly. Widen your opening range on the button and cutoff against them.
• Fold marginal hands to their aggression. Their river raise is pure value.
• C-bet and give up. They fold to c-bets a lot and never check-raise as a bluff.
• Do not value bet thin. If they call three streets, you are beaten more often than you think.

Against the TAG

The hardest profile, because they make few mistakes. Your edge comes from position and situations, not from a gross imbalance.

• Play pots in position, avoid out-of-position confrontations.
• Attack an inflated c-bet frequency with check-raises on textures that belong to your range.
• Respect their 3-bet, but attack their fold to 4-bet if it climbs above 60%.
• Hunt their automatic spots. Most TAGs play the river in very predictable ways.

Against the LAG

They put you under constant pressure. The answer is not to match them, it is to let them overextend.

• Widen your calls, not your bluffs. Their range is wide, your bluff catchers gain value.
• 3-bet wider for value, they call and they 4-bet bluff.
• Let them bet for you with your strong hands. Check-calling and check-raising beat leading.
• Avoid frequency wars out of position with no plan for the river.

Against the maniac

• Stop bluffing entirely. They call or they raise, they never fold out of fear.
• Wait for a hand and play it to the end, letting them build the pot.
• Bankroll and variance: pots against a maniac are huge and often lost. That is correct over the long run.

Against the calling station

The most profitable profile at the table, provided you do not play against yourself.

• Zero bluffs. Not on the flop, not on the turn, and above all not on the river.
• Value bet extremely wide. Second pair with a decent kicker becomes a two-street value bet.
• Raise the sizing. They call 75% pot almost as often as they call 40%.
• Do respect their raise. When a station raises, it is the nuts.

Against the passive fish

• Isolate them in position the moment they limp, with a wide range: that is the whole subject of the iso-raise against a limper.
• C-bet almost every time, they check-fold constantly.
• Cut river bluffs, where even a passive fish eventually calls with a pair.

The one-page adjustment table

Reading profiles in tournaments

In a tournament, the same VPIP means different things at different depths. At 15 big blinds, a 30% VPIP is a sound push-fold strategy. At 100 big blinds, it is a LAG.

Two reading adjustments are mandatory:

1. Segment by depth. A player's early-tournament stats do not describe their final table game.
2. Factor in ICM pressure. A player who suddenly tightens near the bubble has not changed profile, they have changed incentive. That is exploitable, but only temporarily.

Building these profiles automatically in GrindLab

Classifying by hand works, but it does not scale to several thousand opponents. That is exactly what the tracker automates.

• The opponent profile aggregates every hand you have played against a given player and shows their VPIP, PFR, 3-bet and tendencies, hand after hand.
• The live HUD puts those three numbers straight on the table, which makes classification instant during the session.
• Player types let you tag an opponent once and get the label back the next time they sit down.
• The Lab villain tracker brings those stats into the moment you build their range on a specific hand, tying the profile to a real decision.

The full loop: classify during the session, verify the classification in review, and adjust the range you assign them in the Lab.

Classic profiling mistakes

1. Classifying off one spectacular hand. A player who shows you a three-barrel bluff is not necessarily a LAG, they may just have had that spot once.

2. Never reclassifying. Players change: they learn, they tilt, they go from two tables to eight. A six-month-old profile is a hypothesis, not data.

3. Confusing tight with good. A nit is tight and losing. Preflop discipline does not compensate for the absence of postflop aggression.

4. Applying the adjustment on the wrong street. Against a station, the adjustment is postflop. Against a nit, it is preflop. Swap the two and the edge disappears.

5. Forgetting your own profile. Regulars classify you too. If you are read as a nit, your value bets will never get paid.

Key takeaways

• Two stats are enough to classify: VPIP for width, PFR for aggression.
• The VPIP minus PFR gap instantly reveals the size of the calling range.
• Every stat has a reliability threshold: 50 hands for VPIP, 1000 for WTSD.
• Against loose passives, value betting replaces bluffing. Against tight players, stealing replaces confrontation.
• In tournaments, always read a profile relative to stack depth.
• A profile gets reclassified, it is not carved in stone.

GrindLab builds every opponent's profile automatically from your imported hands, shows their VPIP, PFR and 3-bet live on the table, and lets you carry those tendencies into the Lab to set their range on a specific hand. Try it free during the beta →