Postflop in Three-bet Pots

The ratio is computed, not chosen

Blinds 1 and 2, everybody 100 big blinds deep, which is 200 chips. Somebody opens for the maximum and somebody three-bets for the maximum. In pot-limit both of those numbers are forced:

• Maximum open: to 7.
• Maximum three-bet: to 24.
• Blinds fold, the opener calls. Flop pot 49, stacks behind 176.

That is a stack-to-pot ratio of 3.59, and it is the smallest ratio this stack depth can produce with one raise and one re-raise. You cannot make it smaller by betting bigger, because 24 was already the maximum. This is the single most important structural fact about three-bet pots in this game: the tree they create is short, and it is short by rule rather than by choice.

Rule of the game: a raise may not exceed the pot after your call. That is what caps the open at 7 and the three-bet at 24. In no-limit the three-bettor picks the size and therefore picks the ratio; here the ratio picks itself.

Every amount above comes from js/core/plo-pot.js, the legal-raise calculator checked against three independent public rule pages (docs/M2-SOURCES.md).

In a three-bet pot, work out the ratio before the flop comes, because it is knowable before the flop comes. Then plan the whole hand around a tree that is at most two bets deep.

Any single-raised-and-re-raised pot at this depth. Deeper stacks push the ratio up proportionally; a fourth raise pushes it near the mechanical band.

The one thing this does not license is treating every three-bet pot as identical. If the three-bet was smaller than 24 — and most are — the pot is smaller and the ratio is higher, sometimes far higher. Compute it; do not assume it.

A short tree means the decision is on the flop

At a ratio of 3.59 the whole hand is 2 bets long. That is not a metaphor: bet the maximum on the flop, get called, and the money is in on the turn. So the flop is not the first of three decisions — it is the decision.

Two lines split the postflop world in this camp. Below 1.5 is the mechanical band: the hand plays itself, there is nothing to plan, and the only question is whether you are getting the whole stack in. Between 1.5 and 4 is the planned band, which is where almost every maximum three-bet pot lands. Above that you have real multi-street decisions again.

What the ratio asks of your hand, if the stack goes in:

• Ratio one: you need 33.3%.
• Ratio 3.59: 43.9%.
• Ratio four: 44.4%. Ratio ten: 47.6%.

Notice how flat that gets. Past a ratio of about four, more depth barely raises the bar — the requirement is crawling towards a coin flip and never gets there. Almost all of the movement happens in the region three-bet pots live in.

In the planned band, choose on the flop between committing and folding, and let the size follow from that choice rather than the other way round.

Heads up, ratio between 1.5 and 4, on a flop you have not already been priced out of.

The bands are not a hand-strength table, and the mechanical one especially is not. Below 1.5 a hand that is only a draw still has to be judged as a draw — the same ratio that makes a made hand automatic can make a big draw automatic too, and the two are getting in for completely different reasons. Our rule layer handles that case separately for exactly this reason.

What we do not know about three-bet pots

Everything on the previous two pages came out of a pot calculator. Ask the next obvious question — who has the better hands in a three-bet pot, and how should that change the flop bet — and this course goes quiet, on purpose.

Answering it needs to know what a three-bettor holds and what a caller holds after the flop comes down. That is postflop range data. This project has none. Not a rough version, not a borrowed version: none.

So this course gives you no range-advantage numbers and no sizing table that depends on them. Our sizing rule refuses to return an answer at all unless you tell it who has the edge — it will not guess, and it will not default to "the three-bettor". Building the data honestly has no date on it, and the postflop range work it rests on does not exist in this project; a number invented before it does would be a decoration, and this camp does not ship decorations.

In a three-bet pot, plan from the two things you can actually compute: the ratio, which is 3.59 here, and what your specific hand is worth against hands that would put a stack in.

Always in this camp, and especially here, where the pot is 49 with 176 behind and the temptation to substitute a slogan for a calculation is at its highest.

The gap does not mean range advantage is unreal — it is real, and on some flops it is the dominant effect. It means we have not measured it, and those are different sentences. Anyone quoting you a frequency for this spot should be able to say how they measured it; if the answer is a solver run on a range somebody assumed, that assumption is the whole result.

What to take away

• The ratio is forced. A maximum three-bet to 24 leaves a pot of 49 with 176 behind — a ratio of 3.59, and you cannot make it smaller.
• A short tree moves the decision forward. At that ratio the hand is 2 bets long, so the flop is where committing or folding is decided.
• Know what the stack asks for. Getting it in at 3.59 needs 43.9%; the bands are 1.5 and 4.
• Range advantage gets no number from us. We have no postflop range data, our sizing rule refuses to answer without it, and inventing one would be worse than saying so.

That is the end of this camp. Sixteen lessons, and every percentage in them can be traced back to a command, a seed, or an enumeration in this repository; every threshold is one of the rules of this course, not an equilibrium solution. Where we could not trace a number, we wrote down that we could not — those pages are worth rereading, because the gaps are the honest map of what this trainer does and does not know.