River play: thin value and disciplined bluff-catching

There are no more cards to come, so equity is binary: you have the best hand or you don't. That changes everything.

• No protection. You never bet "to deny equity" on the river — there's nothing to deny. Every chip you put in must either (a) get called by a worse hand (value) or (b) fold out a better hand (bluff). If a bet does neither, it's a leak.
• Fold equity is purely psychological now. Your bluff works only if villain *folds a hand that beats you*. So bluffs must target a range that actually contains beatable-but-foldable hands.
• Pot odds are exact. Facing a bet of B into pot P, you risk B to win P + B. Your break-even call% is B / (P + 2B).

Two numbers to memorize cold: - Half-pot bet → you need 25% equity to call (you're getting 3:1). - Full-pot bet → you need 33% (2:1). - 2x-pot overbet → you need 40%.

Most river leaks are people calling at 25%-needed with a hand that's only good 15% of the time, or betting two-thirds pot for value with a hand that's only ahead of the calling range 40% of the time.

Thin value means betting a medium-strength hand that beats *just enough* of villain's calling range to profit. The classic mistake is checking these hands "to keep the pot small" and giving up obvious value.

The test: if you bet and get called, are you ahead more than 50% of the time against the hands that call? If yes, bet. You do NOT need to beat villain's whole range — only the part that calls.

• Sizing for thin value is small. Use 25-40% pot, not 75%. A small bet keeps villain's weak pairs and ace-highs in; a big bet folds them and only gets called by hands that beat you (turning your value bet into a bluff that paid the price of a value bet).
• Example. Single-raised pot, you hold A♦9♦ on 9♣ 6♠ 2♥ J♣ 4♦. You have second pair, decent kicker. Villain check-calls flop and checks turn and river. Bet 30% pot. Worse pairs (T9, 98, 9x with weak kicker, missed gutshots that turned into a sliver of showdown value) will pay; better hands (J9, 99, sets) would mostly have raised earlier or will only call.
• Counter-example — DON'T bet. Same board but villain has been the aggressor (bet flop and turn). Now their range is *condensed to hands that beat second pair*. Checking back is correct; betting only folds out worse and gets called by better.

Heuristic: value-bet thin against capped, passive ranges; check back against ranges that have shown strength.

Pick the bet size *first* based on board texture and range, then choose which hands belong at that size. Mixing this up is the #1 postflop sizing leak.

• Small (25-40% pot): thin value + give-up bluffs. Use on static boards where ranges are wide and capped (e.g. K72r turn-checked-through). Lots of marginal made hands to milk.
• Big (66-80% pot): your polarized range — strong value (sets, two pair, straights) and your best bluffs (busted draws with blockers). Use on dynamic, draw-heavy runouts where you can credibly hold the nuts.
• Overbet (110-150% pot): nut-advantage spots — you hold the top of the range and villain almost can't. Example: you 3-bet preflop and the board runs A♥ K♠ 4♦ 7♣ A♣; you rep AK/AA/A4s and villain's flatting range is full of KQ/JJ/AQ that can't stand an overbet but hates folding.

Key blocker discipline for bluffs: - Hold a blocker to the nuts you're repping; unblock the hands you want to fold. Bluffing a missed flush on a 9♠8♠5♣2♠Q♠ board with A♠x is great — you block the made flush. Bluffing with A♣7♣ (no spade) is much worse. - A river overbet bluff with zero blockers to villain's continue range is almost always -EV. If you can't name the hand you're blocking, don't fire.

A bluff-catcher is a hand that beats *only* villain's bluffs and loses to *all* their value. With one, your call is purely a function of how often villain is bluffing vs. the price.

The rule: call if villain bluffs more often than your required equity. - Half-pot bet: you need to be right 25% → call if villain is bluffing ≥25% of the time. - Full-pot bet: you need 33%. - Overbet (1.5x): you need ~43% — villain has to be bluffing nearly half the time, which most players simply don't do.

How to estimate bluff frequency at the table: - Count the missed draws. On J♥T♥4♣8♥2♣, how many flush/straight draws are in villain's range that bricked? If lots, calling goes up. - Player type beats theory. Most live/low-stakes players under-bluff rivers, especially with big sizing. Against a passive player firing 80% pot on a dry river, fold your bluff-catchers and feel good. Against a maniac who barrels every missed draw, call light. - Blockers help your call too. Holding a card that blocks villain's value (e.g. you hold A♥ on a flush board, blocking the nut flush) makes their range more bluff-heavy — a green light to call.

Concrete spot: pot is 20bb, villain shoves 20bb (full pot) on a Q♣7♦3♠5♦K♠ river. You hold A♣Q♦ (one pair). You need 33%. Does this villain shove a bluff one time in three here? Against a tight player who'd shove KQ/sets/two pair and rarely bluffs a brick — fold. Against a player who turns busted AdJd/9d8d into a shove — call.

OOP on the river you have three options and most players default badly to one of them.

• Check-call with bluff-catchers and thin value you can't bet for value (because only better calls). This is your largest river category OOP. Don't lead these — you'll fold out worse and get raised by better.
• Check-raise only with a deliberately built range: nutted value + a few blocker bluffs. A river check-raise from a weak player is almost always the nuts — so when *you* do it, you must be balanced or villain over-folds (great) but never pay you off. Pick spots where villain bets a lot and is capped.
• Donk-lead (bet into the aggressor) is rarely right but shines when the river card massively favors your range over theirs. Example: you flat from the big blind, board is 8♠7♠2♣ 2♦ and the river is 5♣ completing 64/96/T9 straights you have and the preflop raiser doesn't. A small lead (33% pot) prints.

The leak to kill: leading out small with a hand that wanted a check-call. If your hand beats bluffs but not value, *let villain bet*. Checking realizes more value because villain bluffs their air for you, whereas leading lets them fold the air and raise their value.

Run this every river. It collapses the whole article into a repeatable loop.

1. What's my hand's job? Value (beats worse that calls), bluff-catcher (beats only bluffs), or bluff (beats nothing, needs folds)? Name it before acting.
2. If betting for value — thin or thick? Beat >50% of the *calling* range → bet. Size small (30%) for thin/capped, big (70%) for polarized, overbet only with nut advantage.
3. If bluffing — what do I block and unblock? Must block villain's value and unblock their folds. No blocker, no bluff. Size up so the bluff matches your value bets at that price.
4. If facing a bet — what's the price and their bluff rate? Compute required equity (25% / 33% / 43%). Call only if their *actual* bluff frequency exceeds it. When unsure against a passive player, the default is fold; against a maniac, call.

Discipline summary: most money on the river is lost calling too much against players who don't bluff enough, and won by betting thin against players who call too much. Be the latter.