Rake: How the House Cut Changes Your Breakeven
The house takes a cut of most pots, and that cut quietly raises your required equity, shifts your bluff breakevens, and breaks marginal preflop calls — most of all in small pots at low stakes.
Assumptions: This lesson uses a 100bb 6-max online cash game with a rake of 5% of the pot capped at 3bb, taken only from pots that reach a flop, and reworks the track's core formulas with the final pot shrunk by that cut.
Every formula in this track so far has assumed a clean pot: whatever goes in, the winner takes out. That is a lie of convenience. In a real cash game the house takes a slice of nearly every pot you win, and that slice has a name — rake — and a precise effect on your math. It does not change the shape of the formulas. It changes the numbers you plug in, always against you.
For this lesson, fix the structure explicitly: rake is 5% of the pot, capped at 3bb, and it's only taken from pots that go to a flop (the common "no flop, no drop" rule). Memorize that one sentence, because every number below comes from applying it.
The core idea: the reward side shrinks
When you call a bet hoping to win, you risk your call to win the pot. Rake attacks the reward side of that trade — the pot you collect is smaller than the pot on the table, because the house skims it before you stack it. Your risk, the chips you put in to call, is not refunded if you lose, so the risk side is untouched. Reward down, risk flat: every breakeven moves against you.
How much it moves depends entirely on whether you're near the cap. In small pots, 5% bites the full 5% and the cap never comes into play. In big pots, the 3bb cap freezes the cut at a tiny fraction of the pot, and rake becomes almost irrelevant. So rake is a small-pot tax. Hold that intuition; the arithmetic below confirms it.
Reworking required equity
Take a clean spot from earlier in the track: a 6bb bet into a 9bb pot.
Unraked. You call 6 to win the 9bb pot plus the 6bb bet. The final pot once you call is 9 + 6 + 6 = 21bb. Required equity = call ÷ final pot = 6 ÷ 21 = 28.6%. That's the number you learned.
Raked. Now the pot you actually collect is raked. The final pot is 21bb; 5% of 21 is 1.05bb, which is under the 3bb cap, so the house takes the full 1.05bb. When you win you no longer drag 15bb of profit (the 9bb pot plus the 6bb bet) — you drag 15 minus your share of the rake. Folding the breakeven the honest way: you risk 6 to win (15 − 1.05) = 13.95bb. Required equity = 6 ÷ (6 + 13.95) = 6 ÷ 19.95 = 30.1%.
So rake pushed your required equity from 28.6% up to roughly 30% — about a point and a half. That doesn't sound like much, and in any single hand it isn't. But it is a permanent, structural shift applied to every marginal decision you make, thousands of times a month. Calls that were break-even at 28.6% are now losers at 30%. The thin edges that pay your win-rate are exactly the ones rake eats first.
Reworking the bluff breakeven
The bluff side moves too, and it's the same logic: rake shrinks the reward when you succeed. A pure bluff wins the current pot when the opponent folds. If that pot gets raked, your reward is smaller, so you need folds slightly more often to break even.
Take a pot-sized bet: bet 10bb into a 10bb pot.
Unraked. Break-even fold% (alpha) = bet ÷ (bet + pot) = 10 ÷ 20 = 50%. The opponent must fold half the time or your bluff loses.
Raked. When he folds, you win the 10bb pot, and that pot is raked: 5% of 10 is 0.5bb, under the cap, so the house takes 0.5bb. Your reward when he folds is now 9.5bb, not 10. Risk is still your 10bb bet. Break-even fold% = bet ÷ (bet + pot − rake) = 10 ÷ (10 + 10 − 0.5) = 10 ÷ 19.5 = 51.3%.
So the bluff now needs the opponent to fold 51.3% of the time instead of 50% — about 1.3 points more. Again: tiny per hand, structural over a career. Note the asymmetry the scope flags: the reward side is raked (the pot you'd win) while the risk side — your bet — is yours regardless. That one-sidedness is why rake always pushes the breakeven the wrong way and never the right way.
Where rake bites hardest: set-mining
Set-mining is the cleanest illustration, because it lives entirely in the small-pot, full-rake zone before the pot is built. The classic threshold: to call a preflop raise hoping to flop a set, you want roughly 15-to-1 in implied odds, because you flop a set about one time in 8.5 and you only stack opponents a fraction of those times. The 15-to-1 rule already bakes in that you won't get paid in full every time.
Rake makes the rule stricter. When you do flop your set and win a big pot, the pot gets raked — but in a deep set-mining pot the 3bb cap saves you, because 3bb is trivial against a 100bb+ stack. The real damage is upstream: rake shrinks the medium pots you win when your set gets a little action but not a stack-off, which are a big share of set-mining's profit. The practical adjustment is to demand a touch more than 15-to-1 — call it 16-to-1 in spots where you expect medium pots rather than full stack-offs — and to fold the most speculative set-mines (tiny pairs, multiway with weak stacks behind) that the unraked rule would barely permit.
The marginal preflop defense is the other casualty. A hand like J♥6♥ defending the big blind against a raise is barely break-even unraked — it flops thinly and realizes its equity poorly. Shave the pots it wins by 5% and it tips from a thin call to a clear fold. The general rule: every preflop call that was "barely worth it" before rake is "not worth it" after rake. Tighten your weakest defends.
Where rake barely matters: the cap
Now the other end. Suppose two 100bb stacks get it all in on the flop — a 200bb pot. Rake is 5% of 200 = 10bb, except the cap freezes it at 3bb. Three big blinds out of a 200bb pot is 1.5% of the pot. Your required equity in a stacks-in spot barely moves — a coin flip that was 50% to break even is now about 50.4%. For deep all-in math, you can almost ignore rake.
So the structure tells you exactly where to apply effort:
- Small pots, low stakes: rake takes the full 5% and the cap never triggers. This is where required equity climbs, bluff breakevens climb, set-mines tighten, and marginal defends fold. Pay attention here.
- Big pots: the 3bb cap makes rake a rounding error. Stack-off math is nearly unaffected.
Ranking which spots rake damages most
Given three spots, you can rank rake's bite by where each sits relative to the cap and how thin the edge is:
- A small multiway flop pot with a marginal made hand — full 5% taken, edge already razor-thin. Rake damages this most; it routinely flips marginal calls to folds.
- A medium single-raised pot with a flush draw — 5% still applies (pot likely under the cap), and the draw's edge is moderate, so required equity climbs by a point or two. Real but survivable.
- A deep flop stack-off — the 3bb cap freezes rake at ~1.5% of the pot. Barely damaged.
A practical adjustment menu
You don't need to recompute rake on every hand — that would violate the speed rules from the mental-math lesson. Instead, internalize a few standing adjustments that bake the cut in automatically:
- Add ~1.5 points to your required-equity number in small single-raised pots. If your shortcut says you need 25%, treat the real bar as ~26-27% before you call thin. The marginal calls you were making at exactly the bar are the ones rake deletes.
- Demand a touch more than 15-to-1 to set-mine in spots where you expect medium pots rather than full stack-offs, and fold the most speculative small-pair calls multiway.
- Trim your weakest preflop defends. The bottom of your big-blind defending range — offsuit junk like J6o that barely realized its equity unraked — is now a clear fold.
- Ignore rake entirely in deep stack-off spots. Past the 3bb cap, the cut is a rounding error; don't let it talk you out of a coin flip you're getting in for stacks.
These four habits convert the arithmetic into reflexes. You compute the exact raked number once, here, to understand why the adjustments exist; at the table you just apply them. The understanding is the insurance against applying them blindly — when a spot is unusual (a huge pot, a heads-up pot, a spot well past the cap), you'll know to drop the small-pot adjustment because you know where it came from.
The lesson in one line: rake is a tax on small edges in small pots. It won't show up when you stack someone; it will quietly delete the marginal calls and thin bluffs that, unraked, looked like free money. Account for it by tightening your weakest defends, demanding a little more from set-mines, and treating every "barely break-even" number as a fold once the house has taken its cut.