Postflop Bet Sizing Mastery: When to Go Small, Medium, or Large
Your bet size tells a story — learn how to choose optimal sizing based on board texture, range advantage, and your strategic goals.
The Wrong Way to Think About Bet Sizing
Most mid-stakes players treat bet sizing like a thermostat with two settings: half pot when unsure, pot when strong. They c-bet 50% on every flop, fire 60% on every turn, shove the river when they hit. The hand history reads like a metronome.
This is the single biggest leak separating a 3bb/100 winner from an 8bb/100 crusher at NL200+. Your bet size is not just a number — it is a sentence. It tells your opponent what range you're representing and what hands you can hold. When you size identically across all situations, you're speaking one language while every solver-trained reg in the pool speaks five.
The right way is range-based, not hand-based. You don't pick a sizing because your hand is strong or weak. You pick it because the strategic situation — board, ranges, position, SPR — calls for it. Then you put your entire range into that sizing, and your individual holding goes along for the ride.
Principle: Bet size is a property of your range, not of your hand. Pick the sizing that maximizes EV for the whole range.
The Five Sizing Buckets
Every postflop bet falls into one of five buckets. Each one sends a different message about your range polarity — the spread between the strongest and weakest hands you'd take that line with.
| Sizing | % of Pot | Message Sent | Range Shape |
|---|---|---|---|
| Block bet | 20–25% | "Marginal — please don't raise" | Capped, condensed |
| Small | 33% | "My whole range is fine here" | Merged, wide |
| Medium | 50–66% | "Real hand or real draw" | Mixed |
| Large | 75–100% | "Strong, or bluff with equity" | Polarized |
| Overbet | 125%+ | "Nuts or air — call at your peril" | Extremely polarized |
Block bet (20–25%) — river sizing for marginal showdown value. Denies villain the chance to bet big and force you off. Gets called by worse pairs; rarely raised without two pair or better.
Small (33%) — the range bet sizing. Use on flops where you have clear range advantage but no nut advantage — A-high or K-high dry boards where you have all the top pair combos and villain has none. You bet your entire range, denying equity to villain's underpairs.
Medium (50–66%) — the default. Use when your range is somewhat polarized but not extremely so — turn cards that don't drastically change ranges, or flops where you have both range and nut advantage but villain still has reasonable equity.
Large (75–100%) — a polarized sizing. Your range is split between very strong hands and bluffs with good equity, and you need fold equity from villain's medium-strength holdings.
Overbet (125%+) — the most polarized sizing in poker. Correct when you have a massive nut advantage and villain's range is capped — for example, when you c-bet, turn bricks, but villain has called twice. Villain rarely has a set; you can have all the sets, two pairs, and overpairs.
Principle: The wider the gap between your strongest and weakest hands in a line, the bigger your bet. Condensed ranges bet small or check; polarized ranges bet large or overbet.
The Pot-Odds Math You Actually Need
As a bluff with 0% equity, your bet needs to make villain fold at the breakeven rate or it loses money:
breakeven_fold_rate = bet / (pot + bet)
caller_pot_odds = bet / (pot + 2 * bet)
| Bet Size | Breakeven Fold % (0-equity bluff) | Pot Odds Offered to Caller |
|---|---|---|
| 25% pot | 20% | 16.7% |
| 33% pot | 25% | 20% |
| 50% pot | 33% | 25% |
| 66% pot | 40% | 28.5% |
| 75% pot | 43% | 30% |
| 100% pot | 50% | 33% |
| 150% pot | 60% | 37.5% |
| 200% pot | 67% | 40% |
A small bet only needs villain to fold 20–25% of the time to print as a pure bluff — that's why range bets at 33% are so resilient. An overbet needs 60%+, which only works when villain's range is genuinely capped.
Board Texture Decision Tree
| Texture | Example | Value Sizing | Bluff Sizing | Frequency Mix |
|---|---|---|---|---|
| Dry, ace-high | A♠7♣2♦ | 25–33% | 25–33% | 80%+ bet, merged |
| Dry, low | 8♠4♣2♦ | 33–50% | 33% | 60% bet, merged |
| Semi-wet | K♠9♥6♣ | 50–66% | 50% | 70% bet, mixed |
| Wet connected | J♥T♥8♣ | 66–100% | 66–100% | 50% bet, polarized |
| Monotone | K♥9♥4♥ | 33% | 33% | 40% bet, condensed |
| Paired | Q♣Q♦5♥ | 33% | 33% | 70%+ bet, merged |
| High-card connected | A♣K♥Q♦ | 50–75% | 75% | 50% bet, polarized |
Monotone boards demand small sizing — both ranges are condensed. Big bets fold out hands that beat you and get called by hands that crush you.
Paired boards favor small sizing — villain rarely has trips. Bet small, get called by worse, fold to the rare check-raise.
Wet connected boards demand large sizing. On J♥T♥8♣ every reasonable hand has equity. A 33% bet gives villain the right price; size up to 75–100% to charge draws and make pair+gutshot a tough call.
Principle: Wet boards demand large sizings; dry, paired, and monotone boards demand small sizings.
Range Advantage and Sizing
Range advantage = your range has higher average equity. Nut advantage = you have more very-strong hands than villain.
- You have both → bet small (range bet) or large (polarize), depending on texture.
- Range advantage but villain has nut advantage → check more, bet small when you do.
- Nut advantage but villain has range advantage → polarize. Bet large with nut hands and best bluffs.
- Villain has both → mostly check.
The classic example: Axx flop, PFR vs caller. You opened, BB flatted. Flop A♥8♣3♦. Your range crushes his. Range bet 25% with your entire continuing range.
Flip it: BB defends, flop 7♣6♣5♦. BB has more nut combos (87, 98, sets) but BTN has more high-card equity. BB should polarize: bet large with nut hands and best bluffs, check the merged middle.
In Position vs Out of Position
Position changes sizing because of one fact: the IP player can always check back and realize equity for free. The OOP player cannot.
IP can use smaller sizings to control pot. Smaller bets keep the pot manageable, and if villain check-raises you can fold cheaply.
OOP must polarize more. Villain can float you with any two cards and realize equity on later streets. To prevent that, OOP must bet larger when they bet at all — and check more often with their medium-strength range. A common OOP mistake: c-betting 50% from the BB on every flop because "that's what the BTN does." The BTN can fire small and still see two cards in position; the BB cannot. From OOP, your sizings should be either bigger or none.
Turn Sizing: The Most Polarizing Street
By the turn, you've already told a story. Both ranges are tighter — the turn betting range is a tighter band of strength than the flop range was. This is why turn sizing escalates.
- Brick turn → continue with similar sizing to flop, maybe slightly larger (60–75%).
- Scare card for villain → size up to 75–100%.
- Scare card for you → check more, bet smaller when you do bet.
- Card that completes draws → polarize hard. Either overbet (75–125%) with nuts and bluffs or check.
The turn is the street where you build the pot for the river. If you want stacks in by the river, turn sizing has to be large enough that a pot-sized river shove is feasible.
River Sizing: Three Categories
The river has no future streets, no draws, no equity to deny. Every river bet is purely about value extraction or fold equity.
| Category | Sizing | Use Case | Required Fold % (as bluff) |
|---|---|---|---|
| Block bet | 20–25% | Marginal showdown value, deny villain a big bet | 17–20% |
| Thin value | 33–50% | Beats villain's calling range, loses to his raising range | 25–33% |
| Polar value/bluff | 75–125% | Nuts or air, want max value or max fold equity | 43–55% |
Block bet: K♥J♣ on A♠K♣7♦5♥2♣ IP vs passive BB, checked through turn, BB checks river. Second pair top kicker beats missed draws, loses to Ax. Bet 20% — gets called by KQ/KT/QJ/JT/small pairs; rarely raised without two pair.
Thin value: A♣Q♣ on Q♥9♠5♣2♦4♣ IP. River bricked. Top pair top kicker beats QJ/QT/bluff-catchers, loses to KQ/99. Bet 40% — small enough QJ/QT call, not so big he folds them.
Polar bluff: A♠5♠ on K♠Q♠8♣3♦4♣ IP. Called flop and turn with the nut flush draw; river bricks but you hold the A♠ blocker. Villain checks. Bet 100%. He can't have the nut flush, range is capped at one-pair. Pot-sized bluff with the nut blocker is one of the highest-EV bluffing spots in NLHE.
Geometric Sizing: Planning the Pot-Build
Geometric sizing means choosing flop and turn sizes so a pot-sized river bet puts both stacks in. It's how solvers think about multi-street value.
Worked example: 100bb single-raised pot
Pot 5.5bb, stacks 97.5bb, SPR 17.7. You flop top set, want stacks in by river.
Flop: pot 5.5, bet 4.1 (75%), call → pot 13.7
Turn: pot 13.7, bet 10.3 (75%), call → pot 34.3
River: pot 34.3, bet 25.7 (75%), call → pot 85.7 (stub remains)
75% per street doesn't get stacks in at SPR 17 — you need at least one overbet, or scale up turn and river. Plan three streets ahead. Before firing the flop, ask: "If I bet X here and Y on turn, what does my river bet have to be?" If the answer is "300% pot," you started too small.
Principle: The flop bet you choose constrains every future street. Plan the SPR you want at the river before you press a button on the flop.
Sizing as a Function of Opponent Type
GTO sizing is the baseline; exploitative sizing is where the money is.
- vs Calling station: Bet larger for value — they pay off pot-sized bets just as often as half-pot bets. Bluff less, and when you do, bluff small.
- vs Nit: Smaller value bets — they fold to anything substantial. Never bluff — they have a 90%+ fold-to-cbet rate, but when they call or raise, they have it.
- vs Aggressive reg: Balance. Use a polar-and-condensed split; mix in overbets so you're unexploitable. Be willing to check-raise the turn.
- vs Whale: Value-bet everything, in any size that gets called. Don't bluff — they don't fold marginal pairs.
Sizing Tells: How Bet Size Reveals Leaks
Most players' sizing distributions are wildly unbalanced. Once you see the pattern you can print money.
The "always 50% pot" reg. He has no overbets in his range. When he bets big (75%+), he is always polar — easy fold unless you have a strong bluff-catcher. When you flop a monster, don't slow-play — he won't bet your stack off in his small sizings.
The "huge bet = bluff" leak. Many recreational players overbet only when bluffing. Call wider against their overbets.
The "tiny bet" tell. Some players block-bet their nuts hoping to induce a raise; others block-bet only thin showdown value. Once you've identified which, you have a binary read on every block bet they make.
Sizing that doesn't match texture. When villain bets 25% on a wet J♥T♥8♣ board, he's almost never strong — strong hands want to charge draws. Float wide, barrel turns.
Seven Worked Examples
1. Block bet on a paired turn
BTN opens, BB calls. Flop K♣8♦4♥, BTN c-bets 33%, BB calls. Turn 4♣. BTN has K♥J♥. Bet 25%. The pair doesn't change ranges — BB rarely has a 4. Top pair stays good; block bet keeps weak Kx in and denies bluff-raise leverage.
2. Overbet bluff with the nut blocker
CO opens, BTN 3-bets, CO calls. Flop A♠Q♠5♣ bet/call. Turn J♣ checks through. River 2♦, CO checks. BTN has K♠T♠. Overbet 130%. You hold the K♠ blocker; CO's range is capped at AQ/AJ/KQ/JJ (sets/two pair would check-raise turn). Your overbet reps AK, sets, turned straights. Pot odds offered: 36%; bluff-catchers fold above that.
3. Thin value on a connected board
MP opens, BTN calls. Flop 9♣8♣6♦. MP c-bets 75%, BTN calls. Turn 2♥, MP checks, BTN bets 50%, MP calls. River K♠. MP checks. BTN has T♦T♠. Bet 40%. MP capped at one-pair and busted draws. TT beats QJ/AQ/JJ — small enough they call, not so big you isolate yourself against KK/sets.
4. Small c-bet on dry Axx
BTN opens, BB calls. Flop A♥7♣2♦. BTN has K♥Q♥. Range bet 25%. BTN has all the Ax and overpairs; BB has very few (he'd 3-bet AK/AQ). Range advantage is enormous; villain folds 50%+ to a 25% bet. Fire turn on a Q or J.
5. Large c-bet on J♥T♥8♣
CO opens, BTN calls. Flop J♥T♥8♣. CO has J♣J♦ (top set). Bet 75%. Every hand has equity; JJ wants to charge every draw. At 50%, pot lays draws the right price. At 75%, villain's pot odds are 30% while a bare flush draw has 35% equity but no implied odds. Geometric plan: 75% flop, 80% turn, jam river.
6. Multi-street sizing plan
100bb, BTN opens 2.5bb, BB calls. Pot 5.5bb. Flop K♠9♥4♦. BTN has A♠K♥.
Flop (5.5): bet 50% = 2.75bb, call → pot 11, stack 94.75
Turn (11): bet 75% = 8.25bb, call → pot 27.5, stack 86.5
River (27.5): SPR ≈ 3.1
On the river, shove ~315% overbet vs a station (he calls KQ/KJ/KT), or bet 75% (≈20.6bb) vs a thinking reg (KQ/KJ call, KT folds). The flop/turn sizing made river action meaningful.
7. Reading unusual sizings
BB vs CO open. Flop K♣7♣2♦. You check, villain bets 18%. Below standard. Either he's a weak player block-betting Kx, or a reg using a tiny stab to take the pot. Either way: range is capped — he almost never has KK/77/22/AK at this sizing. Float wide with backdoor equity; barrel any club, jack, ten, or ace. Or check-raise to ~72% with flush draws and gutshots — at 18%, a check-raise is a fold-equity factory.
Putting It All Together
The framework is recursive:
- Identify range and nut advantage for both players on this board.
- Pick the sizing bucket matching the polarity of the betting range you'd construct.
- Verify the math — does the breakeven fold rate make sense vs villain's calling range?
- Plan three streets ahead — does your flop sizing leave a workable SPR?
- Adjust for opponent type.
If you're still using one sizing for everything, you're leaving 3–5bb/100 on the table at NL200+. The fix isn't memorization — it's training your eye to see what the board, the ranges, and position are telling you.
Principle: Bet sizing is the most concentrated form of exploitation in poker. The same hand played for 33% vs 75% can have a 10bb EV swing — and most opponents pick the wrong one.
🎯 Get sizing recommendations on your hands → AI Coach
DEEPFOLD's coaching engine ingests hand histories from PokerStars, GGPoker, and any tracker export, then runs each spot through a sizing-aware EV calculation. Instead of "you should have bet bigger," you get the actual EV difference between your chosen sizing and the optimal one — quantified in big blinds, separated by street, benchmarked against how a solver would have sized. Drop a session in and find out which sizing buckets you're leaving money in.