Blockers in a river decision: count what disappears

An ace in your hand can remove strong hands from an opponent’s range. That sounds useful when you want a fold. The same ace can remove missed draws that would have folded, too. A river blocker earns its meaning from the hands it removes and the decision you are studying. Naming the card without naming those hands leaves the most important part of the argument unfinished.

This lesson uses a small fictional Hold’em ledger on a fixed river board. You can inspect every listed hand, remove the combinations made impossible by your cards, and compare two action branches. The responses and equal weights are supplied assumptions. The exercise counts their consequences; it does not solve the hand or choose a bluff, a call or a bet size.

Yuna: “You said the ace blocks calls. Which calls disappear?” Ace: “Some ace-king hands.” Yuna: “Write those down. Then check whether the ace also takes away hands that were going to fold.”

Start with a decision and a range

A blocker is a known card that makes certain opposing combinations impossible. If you hold As, the opponent cannot hold As Js. That removal is exact. Whether As Js belongs in the opponent’s river range, and whether it bets, calls or folds, needs a separate assumption. A correct removal calculation can still describe an unsuitable range.

The order of the action matters. When you consider betting, you want to know how the opponent responds to that bet. When you face a bet, you want to know which hands produced the bet. A hand that folds to your bet is not automatically a hand that bluffs after you check. It might check back instead. Keep those two decision trees separate.

A card can remove a useful target in one line and an unhelpful target in another. It can also remove hands from several groups at once. Identify which combinations leave which branch, then ask how much weight they had there.

Read exact cards before hand names

Ac Kd and Ad Kc are two different combinations, although both have the hand name ace-king. Our notation uses A, K, Q, J and T for ace through ten; c, d, h and s mean clubs, diamonds, hearts and spades. Each two-card combination is one pair of physical cards. The order of those two cards does not create another combination.

The fixed board is Ks 9s 7d 3c 2h. The king of spades is already on the table, so it cannot appear in anyone’s private hand. Before your cards are entered, four aces and three remaining kings make twelve ace-king combinations. The exercise starts after that board-card removal. It never counts an opponent holding the Ks again.

Holding Ac removes Ac Kc, Ac Kd and Ac Kh from that twelve-hand group. Holding Ac Kd removes combinations containing either card. The Ac Kd combination overlaps both removal reasons and disappears once. Three ace-based removals plus four king-based removals share one combination, so six distinct AK combinations are removed and six remain.

That is the counting foundation. For a fuller treatment of exact combinations and shared-card exclusions, see range combinations and card removal. Here we use that arithmetic to inspect a river response model. A count changes the model’s composition; the strategic meaning comes from the model and the action branch.

Use a ledger with its assumptions exposed

Our invented starting inventory has eighteen exact hands. Twelve are the AK group. Three missed-spade-draw hands form a bluff class: As Js, Qs Js and Js Ts. Three other missed-spade-draw hands form a passive class: As Ts, As 8s and Qs Ts. Every listed hand is compatible with the five board cards. Each receives one unit of starting weight.

The board has two spades. Holding two more spades does not make a flush, and there is no next card to complete one. None of the six missed-draw fixtures has a pair, straight or flush on this board. Those observations describe the fixture cards. They do not prove that every missed draw should bluff or that every AK hand should call a particular bet.

We assign two different response maps to this same starting inventory. If you bet, every remaining AK combination calls and all remaining missed draws fold. If you check, every remaining AK combination bets, the bluff class bets, and the passive class checks back. These are fixed teaching inputs. No solver produced them, and the lesson does not claim they describe an actual opponent.

The names value class and bluff class describe those assigned roles in the toy betting branch. The workbench accepts any legal two-card holding but does not evaluate its showdown strength or relabel the classes against it. Some entered hands can beat or tie hands in the ledger. Treat the output as an inventory experiment, not a verified value-versus-bluff solution for your holding.

Starting class	Exact inventory	If you bet	If you check
AK / value class (12)	Every ace with Kc, Kd or Kh	Calls	Bets
Bluff class (3)	As Js; Qs Js; Js Ts	Folds	Bets
Passive draw class (3)	As Ts; As 8s; Qs Ts	Folds	Checks back

Change the ace's suit and inspect both losses

Enter Ac Qh and inspect the branch where you bet. The Ac removes three AK combinations. Neither card occurs in a missed-draw combination, so all six possible folds survive. Nine calls and six folds remain, giving fifteen equally weighted hands. The folding share is 6/15, or 40%, within this listed response model.

Now change only the ace to As and keep Qh. You remove three AK combinations again. You also remove As Js from the bluff class and As Ts plus As 8s from the passive class. Only three missed draws remain. The branch where you bet now has nine calls and three folds, so the folding share is 3/12, or 25%.

Both holdings block the same number of listed calls. The spade ace also blocks three of the hands that were assigned to fold. That is why the folding share falls despite the appealing story about blocking ace-king. Comparing only the number of strong hands removed would miss the change on the other side of the response ledger.

Ace: “Both aces remove three calls, so they should do the same job.” Yuna: “Count the folds left behind.” Ace: “Six with Ac, three with As. I stopped halfway through the receipt.”

Neither share establishes that you should bet. Ac Qh and As Qh can have showdown value against some missed draws. Checking is therefore not assumed to be worth zero. The exercise supplies no bet size, no full opponent range, no expected-value comparison and no account of how another size changes the responses. A better folding share is one model property, not an action verdict.

Switch the branch when you face a bet

With Ac Qh, the branch after you check contains nine AK bets and three bluff-class bets. The passive class checks back and does not belong in the range of bets you face. Under the assigned equal weights, the bluff-class share of that betting branch is 3/12, or 25%. The denominator is twelve bets, rather than the fifteen hands remaining before an action is observed.

With As Qh, the betting branch contains nine AK bets and two bluff-class bets. The remaining passive hand still checks back. The bluff-class share is 2/11, about 18.18%. This number concerns the makeup of stipulated bets. It is not the chance that your entered hand wins if it calls, because the widget does not compare the final poker hands or ties.

A bluff-catching review needs the price of a call and the hands the catcher actually beats. A class labeled bluff may contain hands that beat a very weak catcher, and a nominal value hand may tie or lose to a stronger entered holding. The labels help us track assigned actions. They cannot replace a showdown comparison when the question becomes whether calling earns money.

This separation prevents a common shortcut: taking every missed draw that folds to a bet and treating it as a bluff when checked to. Our passive class is an explicit counterexample. Its hands fold in the first branch and check back in the second. The difference is an input to inspect, not a frequency the exercise has discovered.

Test a prediction against the full receipt

Before submitting a holding, predict which classes it touches. Ac Qh should remove only AK hands. As Qh should remove hands from all three classes. Js Td should remove the entire three-hand bluff class and leave the AK and passive classes intact. A low holding such as 4c 5d should remove none of the eighteen listed combinations.

Remove your cards from the response ledger

Fixed board: Ks 9s 7d 3c 2h. Eighteen equally weighted exact hands start in three assigned classes: 12 AK hands, three bluff-class missed draws and three passive missed draws. If you bet, AK calls and all six draws fold. If you check, AK and the bluff class bet; passive draws check back. These responses are invented inputs, not recommendations.

Enter two distinct cards not on the board, such as Ac Qh. T means ten; c clubs, d diamonds, h hearts, s spades. The ledger keeps its response labels fixed and does not evaluate your hand, a bet size or expected value.

Ac Qh removes three AK hands and no missed draws. Nine calls and six folds remain if you bet, a 40% equal-weight folding share. If you check and face a bet, nine value-class hands and three bluff-class hands remain, a 25% bluff-class share of that betting branch.

As Qh removes three AK hands, AsJs from the bluff class, and AsTs/As8s from the passive class. If you bet, nine calls and three folds remain, a 25% folding share. The same number of blocked calls can come with fewer available folds. These counts do not choose your action.

Use Ac Kd as the double-counting check. Six AK combinations disappear, leaving six AK hands and all six missed draws. If you bet, six calls and six folds remain, a 50% folding share. In the branch where you face a bet, six value-class bets and three bluff-class bets remain. The passive hands are excluded from that denominator, giving a one-third bluff-class share.

Then try Js Td. Because every bluff-class fixture contains Js, all three disappear. If you bet, twelve calls and three passive folds remain, giving a 20% folding share. If you check and face a bet under our response map, only AK bets remain. This is a count consequence of the invented inventory. It does not claim that a real opponent never finds another bluff.

The receipt lists removed and remaining combinations in every class. Compare those lists before the percentages. Editing a card or switching branches clears the result. Board cards, repeated cards and invalid tokens receive an error. Reset restores Ac Qh and the branch where you bet.

A count needs weights before it becomes a prediction

Equal weights make this experiment easy to audit. Earlier actions can leave different amounts of different combinations in a real range model. A hand that rarely follows the observed line should not automatically receive the same starting mass as one that follows it often.

For a separate weighted example, keep the nine calls and six folds remaining with Ac Qh. Give each call one unit of mass and each fold half a unit. The calls total nine units; the folds total three. The weighted folding share becomes 3/12, or 25%, although six physical folding combinations still exist. Changing weights changes probability mass, not the cards removed.

Arrival weight and response frequency are separate inputs. A missed draw may reach the river rarely and always fold to your chosen size. Another may reach it frequently and sometimes call. To estimate response mass, account for both how the combination got there and what it does after the specified action. The workbench deliberately freezes both so the removal effect can be seen clearly.

Displayed shares are fractions of the stipulated equal-weight inventory, rounded for display. They are not live reads, population frequencies or solver mixing frequencies. Save uncertain weights alongside the count and compare explicit alternatives rather than treating an unsupported weight of one as evidence.

Review the route to the river

A blocker can remove a hand from the relevant river range only if that hand was assigned some chance of arriving. A suited combination that usually raises an earlier street in your model may have little weight after an observed call. Listing every missed draw at full weight can make a blocker look more influential than the action history supports.

Keep the preflop positions, earlier bets and responses, board order, effective stacks and final price in the hand record. They help define which range you intend to study. The five river cards alone cannot reveal which missed draws reached the line or which one-pair hands continue to the chosen size. A blocker story should follow that record rather than fill its gaps.

Bet size can change the group a hand belongs to. A combination assigned to call a small bet might fold to a larger one under a different response model. Holding a card that removes that combination then has a different effect on the folding branch. Recalculate the proposed response map when the action changes; keeping the old labels would compare two different decisions as if they were one.

This is a single-opponent river model. An extra player or an earlier street needs a new inventory and action structure. One removed combination does not describe every player, future board card or later decision.

Yuna: “What would change your conclusion?” Ace: “Whether those missed draws reach the river, and which ones actually bet after a check.” Yuna: “Keep those assumptions next to the count. The receipt should show what you know and what you supplied.”

Keep a review note that can be challenged

A useful note can say: on Ks 9s 7d 3c 2h, Ac Qh removes three of twelve AK combinations and none of six missed draws. Under the assigned bet-response map and equal weights, nine calls and six folds remain. The 40% folding share depends on that map. Checking value and bet size were not modeled. Every part of that note can be checked or revised.

For As Qh, add the three removed missed draws by name and distinguish their classes. For the facing-a-bet view, state that the passive class was excluded because its assigned action is a check-back. This keeps the denominator understandable. A note that says only “ace blocker, good bluff” hides both the unavailable folds and the missing decision inputs.

Identify the assumption most likely to change the comparison: which draws reach the river, which ones bluff, or whether AK calls the chosen size. Test it in a new stated model. The receipt establishes physical removal; the broader decision still needs a range, responses and a price.

Return to Info & Tools for related off-table lessons. These fictional examples are for adults studying poker and do not provide a live betting recommendation.