Poker Bet Sizing: Why How Often Matters More Than How Much

Ask a table how much to bet and you will get a number. A third. Half. Two-thirds, because someone read that two-thirds is standard. The number is always about this hand on this board, and it is almost always the wrong thing to be thinking about.

Here is what the solver actually does, taken from our own c-bet charts — 105 solved flops across seven common matchups.

The size barely moves. The frequency moves enormously.

Across all 42 single-raised-pot flops in that library, the main c-bet size is 1.4bb — on every single board. Ace-high and dry, low and connected, monotone, paired: same number.

What changes is how often that bet gets made:

• Q♥J♥T♠ — c-bets 81% of the range
• A♦8♣3♥ — 78%
• 9♠4♦2♣ — 58%
• 5♥4♠3♦ — 49%
• A♥K♦4♦ — 45%

That is a range of 45% to 81% in frequency, with the headline size pinned at one number. If you have been agonising over whether to make it a third or a half, you have been turning a dial the solver mostly leaves alone.

But there are two dials, not one

The 1.4bb figure is the main size, and every one of these spots also has a larger option. This is where the interesting part lives, because the mix between them is not fixed at all. Counting how much of all the betting goes to each size:

• A♦8♣3♥ rainbow — 78% c-bet, and 100% of it small. The large size is not used at all.
• 5♥4♠3♦ rainbow — 49% c-bet, 85% small and 15% large.
• Q♥J♥T♠ two-tone — 81% c-bet, and 76% of it at the larger size.

So QJT and A83 both get bet around 80% of the time, and they are not remotely the same bet. One is a wide, cheap stab. The other is a big bet with a wide range behind it. Reading only the c-bet percentage would tell you those two boards play alike. They do not.

What each dial is actually responding to

Frequency responds to who the board favours

A♦8♣3♥ is a preflop raiser's board. You hold far more strong aces than the big blind does, they folded most of their ace-high, and almost nothing on that flop connects with the hands they defended. You can bet nearly your whole range for a small price because they cannot fight back credibly.

5♥4♠3♦ is the opposite. It hits a big blind's calling range — the suited connectors, the small pairs, the gapper junk they defended cheaply — far harder than it hits a button opening range full of broadways and big aces. So the frequency collapses to 49%. Betting often into a board that favours your opponent is how you donate.

Size responds to how polarised you are

Small bets work when your advantage is range-wide but thin. On A83 you are ahead nearly everywhere and nowhere by very much — you do not have many hands that want a big pot, and you do not need one to deny equity to king-high. A cheap bet with everything extracts the maximum.

The larger size appears when the board creates real separation between the top and bottom of your range. Q♥J♥T♠ makes straights, sets, two pair and monster draws — and it also leaves you holding plenty of complete air. That is a polarised range, and polarised ranges want a bigger bet: the strong hands charge draws that genuinely have equity, and the bluffs get maximum fold pressure. Betting 1.4bb into that texture would be charging a flush draw almost nothing.

This is the whole idea in one line: frequency is about who the board favours, size is about how spread out your own range is on it.

Pot geometry sets the baseline

The other thing that moves size is the pot you are already in. In our 3-bet pot charts, the main size jumps to 5.4bb — and on 8♥7♥3♠, 76% of the betting goes to the larger option, the same pattern as QJT.

That is not a different philosophy. It is the same fraction of a much bigger pot, with far less room behind. In a 3-bet pot the stack-to-pot ratio is low enough that a couple of bets gets the money in, so each one has to be larger to stay on schedule. If you want the arithmetic of getting stacks in smoothly over a known number of streets, our geometric sizing calculator does it.

What to actually do at the table

You do not need to memorise 105 flops. Three questions get you most of the way:

1. Whose range does this board favour? That sets how often you bet. Ace-high and broadway-heavy boards favour the raiser. Low and connected boards favour the caller.
2. How spread out is my range here? Thin edge everywhere means small. Nuts-and-air means big.
3. How much is already in, and how deep are we? Low SPR pushes every size up.

A workable default for low stakes: small and often on dry, high-card boards; less often and bigger on connected, high boards; and give up more than feels comfortable on low connected ones. That is a crude version of what the charts show, and it will beat a fixed two-thirds with everything.

The mistakes this fixes

One size for everything. The most common leak in small games. A single sizing means your range is never polarised when it should be and never cheap when it could be, and it makes you trivially readable to anyone paying attention.

Sizing by hand strength instead of by range. Betting big with strong hands and small with weak ones is the leak that never gets punished at 1/2 and never stops being a leak. The size is a property of your whole range in that spot, not of the two cards you happen to hold.

Betting the same frequency everywhere. If you c-bet 75% on every flop, half those boards are ones where the solver is folding or checking most of its range. The frequency dial is the one doing the real work, and it is the one most players never touch.

Confusing size with aggression. A 1.4bb bet made 81% of the time applies far more pressure over a session than a pot-sized bet made 30% of the time. Small is not passive.

Check the numbers yourself

Every figure above is on the site and clickable. The c-bet charts show the per-hand strategy for each of those 105 flops, including which hands take the larger size and how often — hover any cell in the grid. Comparing two boards side by side for five minutes will teach you more about sizing than any rule of thumb, because you will see the frequencies move and the reason will be obvious once you know what to look for.