Bluffing in poker: frequency, price, and the opponent who folds

Bluffing in poker fails in two different ways, so split the word. A pure bluff is a hand that is almost certainly behind and will not improve. You are buying the pot. If they call, you lose the bet. A semi-bluff is a draw or a weak hand with outs. You can win immediately when they fold, and you can still hit a pair, a flush, or a straight when they call. The semi-bluff is the workhorse. The pure bluff is the specialist tool on the river, when no more cards are coming.

The Poker guides hub is the index. Poker pot odds is the price of a call. This page is the price of a bluff, which is the mirror. GTO poker strategy is where balanced frequencies live once the basic formula feels small. You do not need a solver to stop bluffing the calling station.

A bluff is not a lie you tell with your face. Poker tells is a separate subject and a weaker one. If your plan requires a facial expression, you do not have a plan. You have a bet size and a hand they might fold.

Worked example 1. The pot is $200 on the river. You bet $100 with a missed draw. You risk $100 to win $200. That is 2-to-1. The formula for a pure bluff is success rate greater than bet / (pot + bet), which is 100 / (200 + 100) = 1/3. You need them to fold more than one time in three. If they fold 34 percent of the time and call 66 percent, and you always lose when called, the bluff is a thin winner before rake. If they fold less than that, the bluff loses money. Fold equity is not optional. It is the whole result.

Write it once and keep it. For a pure bluff, required fold frequency is the amount you risk divided by the amount you risk plus the pot you win. Risk is your bet. The pot you win does not include your bet, because you do not win your own chips back as profit. You win what was already there, including their earlier money.

If you bet the size of the pot, you risk P to win P. That price is 1-to-1, so a pure bluff needs folds more than half the time. If you bet half the pot, you risk half of P to win P. That is the same 2-to-1 shape as the $100 bet into a $200 pot, and the required fold rate is again one-third. A smaller bet needs fewer folds and applies less pressure. A bigger bet needs more folds. That trade is the sizing decision.

Semi-bluffs lower the bar because you sometimes win when called. Count outs honestly, and remember that a second caller makes the pure-bluff fraction harsher. Rake makes the required fold rate a little higher than the textbook fraction.

A single bluff is a cash question: will this opponent fold often enough. A range is a ratio question: of the hands you bet this way, how many are value and how many are bluffs. If you bet the size of the pot, a caller who is getting 2-to-1 needs to win one time in three to break even. If your betting range is two value hands for each bluff, they win one-third of the time when they call and cannot exploit you by always calling or always folding. That is the toy version of balance. GTO poker strategy adds blockers, board coverage, and the fact that real ranges are messy.

You do not need that toy ratio against a recreational player who never folds a pair. Bluff less. You do not need it against a player who folds everything but the nuts. Bluff more. The ratio is the defense against an opponent who is paying attention. The exploitative adjustment is the profit against an opponent who is not.

Pick the bluffs that block calling hands when you can, and bluff with hands that have little showdown value. Size bluffs the same way you size value. A special bluff size is a translation guide for the table.

Bluff when the story is believable. A dry ace-high board favors the preflop raiser. A bet from that player can represent the ace. A low connected board that clearly helped the caller is a worse place to fire a second barrel just because you are “being aggressive.” Aggression without the range is a donation with better posture.

The best opponent owns a folding button. Tight players fold too much on scary rivers. Fit-or-fold players fold when they miss, so a continuation bet on a dry flop prints money until they adapt. Calling stations, gamblers who came to see a showdown, and anyone who has announced they never fold a pair are the wrong audience. Believe them.

Position helps because you see the check. A check from a straightforward player on a scary card is a green light only if their check really means weakness. Some checks are traps. One trap does not end betting. It does mean you stop auto-firing.

Worked example 2. You raised preflop and the big blind called. The flop is king-seven-two rainbow. You bet half pot with a missed suited connector, a semi-bluff with two overcards and a backdoor. They call. The turn is a blank. They check. You give up. The call said they have a piece, the blank did not add a new story, and a second barrel now needs them to fold a pair. This opponent has not folded a pair all night. The formula still wants folds. You will not get them. Checking back is the bluff you did not fire, which is how the frequency stays honest.

Count your bluffs for a session if you want the truth. Many players think they bluff constantly and actually bluff once a night, always in a giant pot, always as a pure bluff on the river. That is the expensive shape. Small semi-bluffs on dry flops, given up on bad turns, are the cheap shape.

A practical cap: if you cannot name the fold you are targeting, you are not bluffing. You are betting because the silence felt awkward. Awkward is not a frequency. Against unknown players, bluff less than the balanced toy ratio until you see a fold. Against nits who over-fold, bluff more and value-bet thinner, because they also call too little with medium hands. Against stations, delete the pure bluffs and value-bet every pair that can get called by worse.

Rake and stack depth change the price. Once the bet is most of the stack, the decision is push or give up. Use the chips actually at risk.

Read it on the river especially, where the pure bluff has no outs left.

• Name the better hands you want to fold. If you cannot, check.
• Compute risk and pot. For a 2-to-1 pot, you need folds more than one-third of the time if the bluff has no equity.
• If it is a semi-bluff, say how often you improve when called, and do not double-count.
• Check the opponent. Stations do not pay you for a story.
• Check the board. Your range should contain the hand you are pretending to have.
• Use a size you also use for value.

Then bet once, with a plan for the next street. A bluff without a turn plan becomes two bluffs, and the second one is usually the loser. Giving up is part of bluffing in poker. Players who never give up are not aggressive. They are the customers.

A bluff does not change the cards. It changes whether the cards are shown. If they call and you are behind, you lose. No story fixes that. A bluff also does not work in a game with no decision to fold. Coinflip pays the player who picked the side that landed. There is no fold equity, because there is no second action. Poker’s extra layer is the fold. That layer is profitable only when someone sometimes uses it.

Do not bluff multiway just to “take the pot away” when two players have already called a street. You need both to fold. The pure-bluff fraction gets harder as soon as a second caller exists. Semi-bluff multiway only when the price and the outs are real, which is rarer than the movies suggest.

And do not confuse a bluff with a tell war. If you want faces, read the tells page and then come back to the fraction. The fraction is what keeps the lights on. One-third is the worked answer for a 2-to-1 pot with no equity. A real opponent may simply decline to fold.