Pot odds, solved.
Learn how to price a call with worked examples, interactive calculators, a reference chart and six practice questions.
How this page works.
Three minutes here will save you twenty later. Every section answers one question and ships with a working tool — read top to bottom, or jump straight to the calculator.
One fraction. Required equity = cost / final pot. Everything below builds on this.
Drop in real numbers. Calculators show what the answer looks like in practice.
Six worked spots at the end. Get five right and you're ahead of the room.
Every call is a tiny bet.
When villain bets $50 into a $100 pot, you're not deciding "is my hand good?" — you're being offered a tiny side bet at 3-to-1 odds. Your hand only has to win 25% of the time to break even. Pot odds is the framework for evaluating that side bet against your hand's equity.
If there is no further betting and no rake, a call has positive expected value when your equity exceeds the call price divided by the final pot. Other spots also require future-cost and payoff estimates.
Pot odds give the break-even price. Compare that with equity against the betting range, while accounting for the assumptions behind both numbers.
Plug in real numbers.
Pre-set the most common bet sizes — 1/3 pot, 1/2 pot, 2/3 pot, pot, overbet — and see the required equity emerge live.
Counting outs.
Outs estimate your chance of improving. Whether those cards win depends on the board and the opposing hand or range.
Implied odds.
Extra money won after improving can make a call profitable. The amount depends on effective stacks, the opponent and whether your improved hand wins.
Minimum Defense Frequency.
MDF is a range-level benchmark from a simplified heads-up river model. It is different from the equity needed to call.
Bluff:value ratios.
In a simple polarized river model, the balanced bluff share makes a pure bluff-catcher indifferent to calling. Larger bets allow a larger bluff share.
River bets against real players.
The prices above, set against what players actually did with 1.5 million river bets in real-money games of 2009: how often the other player folded, and how often a call won.
folded to a bet of about half the pot
A bet that never wins when called needed folds 34% of the time to break even. Players folded more than that at every bet size.
of calls against it won
The price of those calls averaged 25%. Those who called won often enough to clear it.
back for each $1 called, at worst
Counted in money, callers came out ahead at every size, before rake: from $1.16 per $1 called against overbets to $2.94 against the smallest bets.
| Bet, share of the pot | Folded | Bluff needs | Callers won | Call needs | Back per $1 called |
|---|---|---|---|---|---|
| Up to 30%322,878 bets | 27% | 16% | 40% | 13% | $2.94 |
| 30% to 45%264,138 bets | 46% | 27% | 39% | 21% | $1.89 |
| 45% to 60%327,824 bets | 53% | 34% | 40% | 25% | $1.66 |
| 60% to 80%280,747 bets | 58% | 41% | 38% | 29% | $1.36 |
| 80% to 120%252,277 bets | 67% | 49% | 36% | 33% | $1.19 |
| Over 120%99,717 bets | 80% | 65% | 36% | 38% | $1.16 |
“Bluff needs” is the fold share a pure bluff needs to break even; “call needs” is the share of showdowns a call must win, averaged over the called bets; “back per $1 called” is what the callers’ wins returned. All before rake. Where the win rate sits under the average price yet the money comes out ahead, the calls won were in bigger pots. The rest of each band raised: from 3% against overbets to 19% against the smallest bets.
How the bets were counted
The hands. 14.0 million real-money no-limit hold’em, 10¢/25¢ to $5/$10 hands from PokerStars, PartyPoker, Full Tilt and Absolute Poker, July 2009: the HandHQ histories published by the University of Toronto’s Computer Poker Research Group (the dataset, CC BY 4.0).
The bets. Every river where exactly two players were left and neither was all in: the first bet, sized against the pot before it (folded players’ money included, the bet capped at what the other player could call), and the other player’s next move. A raise counts as defending, not as a call. 1.5 million bets; a size band is shown only with 1,000 or more.
The prices. Each band is set against its own bets’ prices, worked out bet by bet: a bet that never wins when called breaks even if the other player folds bet ÷ (pot + bet) of the time, averaged over all the band’s bets; a call breaks even with bet ÷ (pot + 2 × bet), averaged over the bets that were called. “Back per $1 called” is the callers’ money: what their wins returned (pot + 2 × bet, half on a split) for every dollar they called. Rake is left out.
Who won a called bet. The showdown decides: the winning hand is always shown, so a hand not shown lost, and two shown hands are ranked. Payouts could check this only at Full Tilt and PartyPoker (PokerStars and Absolute record none for most showdowns): they agree in 115,022 of 117,791 hands. Of PartyPoker’s 2,769 disagreements, 2,166 are pots the shown cards split but the payouts record unevenly, 426 a short caller’s all-in where the payouts count the bettor’s returned excess, and 177 other. The callers’ win rates by band are close at all four sites.
What it cannot say. The fold share says what a bet that never wins when called would have earned, not that bluffs paid: the bets were a mix of strong hands and bluffs. Callers include strong hands too, so callers coming out ahead says nothing about calling with a hand that only beats a bluff. A few regulars play many of these hands and the error by player was not worked out, so only whole percents are shown. These are players of 2009.
Checked. The definitions were written before any code (with two changes after the first runs, both written down: where payouts are missing the showdown decides, and then the showdown decides wherever it is known). A second program written separately recounted all four sites: all 304 figures agree.
The numbers, on one page.
Compare matched heads-up bets with the equity required to call and the pot odds offered. All pot sizes are before the bet; rake and future betting are excluded.
The eight common leaks.
Knowing the math is half the battle. Here's the half nobody talks about — the ways smart players still bleed money even after they understand pot odds.
Counting outs that are not clean
A card that improves your hand can also give an opponent a better hand. Count only winning outcomes; there is no universal 50% discount.
Assuming a future payoff
Implied odds depend on extra money an opponent will actually pay when you hit. Deep stacks make that possible, not guaranteed.
Skipping the river price
Compare the call price with your equity against the betting range. If your hand only beats bluffs, estimate the bluff share rather than assuming every large bet is value.
Treating MDF as a mandatory call rate
MDF is a simplified range-level benchmark. Against a player who bluffs too little, folding more can be correct.
Overbluffing opponents who call too often
Reduce bluffs against opponents who rarely fold. The adjustment depends on their actual calling range and the price of the bluff.
Ignoring effective stacks
You cannot win unmatched chips. Cap the call and any projected payoff at the chips the players can actually contest.
Ignoring reverse implied odds
Improving to a second-best hand can cost additional bets. A draw is less valuable when hitting it can still leave you behind.
Treating tournament chips like cash
Chip-EV pot odds do not include payout effects. ICM can change the equity needed for a tournament call, depending on stacks and payouts.
Six spots. How many can you solve?
Six practice questions covering call prices, clean outs, defense frequencies and effective stacks.
-
1. On the turn, villain bets $50 into $100. You have nine clean flush outs, no other way to win and no future payoff. Is calling profitable on direct pot odds?
Answer No — about 19.6% is below the required 25%A call needs 50 / (100 + 50 + 50) = 25% equity. Nine clean outs among 46 unseen cards give 9 / 46 ≈ 19.6%. With no additional ways to win or later payoff, the call has negative expected value.
-
2. Villain bets ⅔ pot on the river. What is the minimum equity you need to call, before rake?
Answer 28.6%For a $60 bet into $90, the final pot after your $60 call is $210. Required equity is 60 / 210 ≈ 28.6%. MDF is a different number: 90 / (90 + 60) = 60% of the defending range.
-
3. On the flop, villain makes a pot-sized all-in bet. You have a bare gutshot with four clean outs and no other way to win. Does calling meet the direct pot-odds threshold?
Answer No — about 16.5% is below the required 33.3%A pot-sized bet needs 1 / (1 + 1 + 1) = 33.3% equity. With four fixed clean outs and both cards guaranteed, the chance of hitting is 1 − (43/47 × 42/46) ≈ 16.5%. The all-in condition removes any later payoff.
-
4. In the simple heads-up river model, what MDF corresponds to a half-pot bet?
Answer 66.7%MDF = pot / (pot + bet) = 1 / 1.5 ≈ 66.7%. Defending this often makes a zero-equity bluff break even. It is a range-level benchmark, not a requirement to call a particular hand or an underbluffing opponent.
-
5. In the simple polarized river model, what fraction of a pot-sized betting range consists of bluffs at equilibrium?
Answer 33.3%The bluff share is bet / (pot + 2 × bet) = 1 / 3 ≈ 33.3%. That is two value hands per bluff and makes a pure bluff-catcher indifferent to calling.
-
6. Heads-up on the river, the pot is $400 before villain shoves $400. You have only $80 left. What equity do you need to call?
Answer 14.3%You can call only $80. Of the $400 shove, $80 is matched and the other $320 returns to villain. The pot you can win after calling is $400 + $80 + $80 = $560. Required equity is $80 / $560 ≈ 14.3%, before rake.
Three numbers. That’s it.
Pot odds is one fraction with three inputs. Once you have it, every interactive on this page is just an extension of the same idea.
Required equity
Call / (pot before bet + matched bet + call). Include only money you can win; this direct-price calculation excludes rake.
Your equity
Nine fixed clean outs hit about 35.0% from flop to river if you see both cards, or 19.6% from turn to river. Actual equity also depends on the opposing range.
Compare
Compare equity with the direct-price threshold, then account for later costs, realistic payoffs and tournament payout effects where relevant.