Bluff-Catching with Combos: The Value-to-Bluff Ratio
Marry pot odds to combo counting: compute the equity the price demands, count the bettor's value combos and bluff combos, and call exactly when the bluff fraction clears the bar. One worked call, one worked fold, and the miscounts that flip them.
Assumptions: All examples use a 6-max online cash game at 100 big blinds effective with no rake, unless a different stack depth or format is stated in the example.
Every river bluff-catch is the same question wearing different clothes: is this bet a bluff often enough to pay off at this price? Both halves of that question are numbers you already know how to produce. The price half comes from pot odds — the bet size fixes exactly how often you must win. The frequency half comes from combinatorics — count the value combos and the bluff combos in his betting range, with removal applied, and divide. When the bluff fraction beats the required equity, you call; when it doesn't, you fold; and your hand's actual cards barely matter beyond beating the bluffs. This lesson works one complete call, one complete fold, and then shows you how fragile the whole machine is to a single miscounted class — which is exactly why you count the big classes first.
The price sets the bar
You're in the big blind with T♣T♦. An aggressive CO reg opened to $3, you called ($6.50 pot), and the flop came A♠9♦5♥. He c-bet a full pot-size $6.50; with two unders to the ace but a hand far too good to fold to one bet, you called ($19.50). The turn 2♣ went check-check. The river is the 8♦, you check, and he bets $19.50 — full pot.
Price first, always, because it tells you what you're shopping for. You must call $19.50 to win the $39 out there (his bet plus the pot): 19.5 ÷ 58.5 = 33.3% required equity. Against a polarized pot-size bet — he has it or he doesn't, and your tens beat nothing he values — required equity is required bluff frequency: his betting range must be at least one-third bluffs, or your call burns money. That's the bar. Now count whether he clears it.
Count the value
Board: A♠9♦5♥2♣8♦. Your read on this reg, from this line — pot flop bet, turn check-back, big river bet: he value-bets every ace-king and ace-queen he opened, plus the two sets he flopped and decided to trap with on the turn. (AJ and weaker aces check back this river or never arrive in his CO opening range this way; assumptions like these are the input to every count, and we'll stress-test them at the end.)
Apply removal — A♠, 9♦, 5♥ are on the board, and your T♣T♦ matters more than it looks:
- AK: 3 aces × 4 kings = 12 combos
- AQ: 3 aces × 4 queens = 12 combos
- 99: the 9♦ leaves three nines → 3 combos
- 55: the 5♥ leaves three fives → 3 combos
Value total: 30 combos. Every one of them beats your tens.
Count the bluffs
What does this line bluff? His flop pot-bet bluffs were broadway air and backdoors; the ones that kept bluffing potential to the river are queen-jack and jack-ten — the air he arrived with that never improved past jack-high on this runout, with no showdown value and every incentive to fire when checked to twice. Count them:
- QJ: no Q or J is visible anywhere — 16 combos (4 suited + 12 offsuit), all jack-high or queen-high at showdown, all beaten by your tens.
- JTs: here's your hand doing hidden work. JTs is 4 combos fresh — but you hold the T♣ and T♦, leaving only J♥T♥ and J♠T♠: 2 combos. Your bluff-catcher blocks his bluffs, which is the bad kind of removal when you want to call. Notice it, count it, don't round it away.
Bluff total: 18 combos.
Compare and call
His river betting range is 30 value + 18 bluffs = 48 combos. Bluff fraction: 18 ÷ 48 = 37.5%. The bar was 33.3%. He's over it — call. As a sanity check, a full equity simulation of T♣T♦ against exactly that 48-combo range on this board returns 37%, matching the count (your tens win against precisely the bluffs and nothing else, so equity equals bluff fraction). The margin is 4 points of equity on a $58.50 pot — roughly $2.50 of EV per call. Not a triumph; a correctly priced decision you'll repeat hundreds of times.
The fold: same price, value-heavy line
Now change the line, not the price. You flat J♠J♦ in the CO against a solid UTG open ($2.50; pot $6.50 after the blinds fold), and the board runs out Q♦8♠4♣ — K♣ — 2♠ while he bets every street: $4 on the flop (you call; $14.50), $9.50 on the turn (you call; $33.50), and $33.50 — full pot — on the river. Same bar as before: 33.5 ÷ 100.5 = 33.3%.
Your jacks are a pure bluff-catcher again: every value hand beats you, and you beat every busted draw. But count what a triple-barrel from a UTG range contains on this board, with the Q♦, K♣ and your two jacks removed:
| Class | Combos | Notes |
|---|---|---|
| AA | 6 | untouched |
| KK | 3 | K♣ on board |
| 3 | Q♦ on board | |
| AK | 12 | 4 aces × 3 kings — top pair, top kicker |
| KQ | 9 | 3 kings × 3 queens — top two pair |
| 88 | 3 | set |
| 44 | 3 | set |
| Value total |
And the bluffs? This board offered UTG's range almost nothing to semi-bluff: no flush draw ever existed, and the only one-card straight draws his opening range held were JTs (gutshot, then open-ended when the king arrived) and T9s (gutshot at a jack). Both missed — and you hold two of the jacks that JTs needs: your J♠ and J♦ kill the spade and diamond combos, leaving JTs at 2 combos (J♥T♥ and J♣T♣), while T9s survives at 4. Bluff total: 6 combos.
Bluff fraction: 6 ÷ 45 = 13.3%, against a 33.3% bar. Fold, and it isn't close — the simulation against the full 45-combo range returns the same 13%. For the call to be right, his bluffs would need to triple.
Set the two examples side by side and the method's honesty shows. Nothing about your hand strength changed meaningfully — tens and jacks are both one-pair bluff-catchers. Nothing about the price changed at all. What changed is that one line crossed a wet-enough runout to arrive with 18 bluff combos, and the other crossed a desert and arrived with 6. You didn't decide to be a station in hand one and a nit in hand two; the counts decided.
How wrong can the count be?
This is an estimate built on range assumptions, and you should know exactly how much load each assumption carries. Re-run hand one with single-class errors:
- He doesn't bluff QJo, only QJs. Bluffs drop from 18 to 6 (QJs 4 + JTs 2): 6 ÷ 36 = 16.7% — your clear call becomes a clear fold. One assumption about one offsuit hand class swung the answer by 21 points, because offsuit classes are 12 combos at a time.
- He also value-bets AJ. Add roughly 12 combos of value: 18 ÷ 60 = 30% — call flips to fold.
- He gives up with QJ entirely and finds some other 8-combo bluff class instead. Bluffs 8, value 30: 8 ÷ 38 = 21% — fold.
The pattern: big classes move the answer, small ones don't. JTs being 2 versus 4 combos shifted the fraction by about two points; QJo being in or out shifted it by twenty-one. So when you count under time pressure, count the largest classes first and most carefully — offsuit broadways (12s), unpaired value hands (12s and 16s) — and let the 2-and-3-combo classes round. If the decision is close enough that a 2-combo class flips it, it's close enough that either action is fine; if a 12-combo class flips it, the assumption behind that class is the decision, and that's where your attention belongs.
One more habit transfers from this lesson to everything downstream: state the price as a bluff-fraction bar before counting. Pot-size bet: 33%. Half pot: 25%. Two-thirds: 28.6%. If you know the bar first, you often don't need a precise count — a triple-barrel on a board that never had draws can't reach a 33% bar no matter how you shade the assumptions, and a pot bet from a range with three live busted-draw classes usually clears 25% easily. The full count is for the spots in between, and now you can run it end to end: price, value census, bluff census, divide, act.
Choosing the right bluff-catcher
Once the range is close to the bar, your exact hand matters through removal, not through pride. A bluff-catcher wants to block value and avoid blocking bluffs. In the TT example, holding two tens was awkward because it removed JTs bluffs; if you instead held a pair that blocked no missed straight draws, the bluff count could be slightly higher and the call easier. In the JJ example, your jacks performed the same bad service by cutting JTs from four suited combos to two.
The good blocker is the opposite. If a river value range is full of top pair and two pair, holding a card from those value classes matters. On A♠9♦5♥2♣8♦, a hand like A blocker plus a bluff-catcher is rare because ace-high usually has showdown logic of its own, but the principle still applies: cards that remove AK and AQ make calling better; cards that remove QJ and JT make calling worse. Do not say "I have blockers" as a blanket reason. Name which side of the ratio they block.
This is why two hands with identical absolute strength can have different river decisions. A pair of tens and a pair of sevens may both lose to every value hand and beat every bluff, but if the tens remove the missed straight draws and the sevens do not, the sevens can be the cleaner call. The count tells you the range's baseline bluff fraction; blocker quality tells you which bluff-catchers realize that fraction best.