Turn tournament chip stacks into real money. Enter the stacks and the payouts, and see what each seat is actually worth.
In a tournament, the chips in front of you are not money — they are a claim on a share of the prize pool, and that relationship is not one-to-one. The Independent Chip Model (ICM) is the standard way to work out how much each stack is really worth in dollars, given the remaining payouts. Use it to settle a final-table deal, to understand why the chip leader isn't worth their chip count, and to see why folding is often right on the bubble even when the raw pot odds say call. Enter the stacks and prizes below; the equities update instantly.
| Player | Stack | Chips % | ICM $ | Pool % | vs Chip-Chop |
|---|
The vs Chip-Chop column is the heart of it: it compares each ICM equity to a naive split by chip count. The chip leader's figure is negative and the short stacks' are positive, because ICM moves money from big stacks to small ones. That gap is the "ICM tax", and it is why a big stack should not treat every chip as a dollar.
The Independent Chip Model answers one question: if this tournament finished from here, how much is each player's stack worth in cash on average? It does that with a simple, elegant assumption — that a player's chance of finishing in any given place is proportional to their share of the chips still in play. The chip leader is most likely to come first, the short stack most likely to bust next, and everyone else falls in between. Run that logic across every possible finishing order, weight each by its probability, and multiply by the prizes, and you get each stack's dollar equity. That is exactly what the calculator above does.
The key fact ICM captures is that the chips you can win are worth less than the chips you can lose. Doubling your stack does not double your equity, because a big stack cannot cash more than first place, but any stack can bust to zero. The upside is capped and the downside is total, so each additional chip is worth a little less than the last.
Take the default example above: three players left with 50,000, 30,000 and 20,000 chips and prizes of $500, $300 and $200. By raw chip count the leader owns half the chips, so a chip-chop would hand them $500. ICM says their stack is worth only about $384. The other $116 is spread to the shorter stacks, who are more likely to ladder up into a paying spot than their chip counts suggest. Nobody has been eliminated, yet the money has already shifted.
ICM is a model, and it makes simplifying assumptions worth knowing. It ignores skill — it assumes every chip is equally likely to end up anywhere regardless of who is playing — and it ignores position, blind levels and future play, treating the tournament as if it froze right now. Better "future game simulation" models exist for close spots, but they are far more complex, and for deals and everyday bubble decisions ICM is the accepted standard and gets you almost all of the way there.
ICM, the Independent Chip Model, is a formula that converts a tournament chip stack into its real-money value based on the remaining prizes. It assumes your chance of finishing in each place is proportional to your share of the chips in play, which lets you price every stack in dollars rather than chips.
Your chance of finishing first is your share of the total chips. Given who finishes first, your chance of finishing next is your share of the chips that remain, and so on down through every paid place. Each finishing order has a probability; multiply those probabilities by the prizes for each place and sum them, and you get each stack's expected payout. This calculator runs that recursion (the Malmuth-Harville method) for you.
When the players still in a tournament agree to divide the remaining prize money rather than play it out, that is a deal or chop. An ICM deal pays each player the current ICM value of their stack. Because a pure ICM chop is toughest on the chip leader, deals are often ICM numbers with a little extra set aside for the biggest stack, or a small amount left to play for.
Because you can only win first place once, but any stack can bust. The upside of a big stack is capped at the top prize while its downside is total, so each extra chip is worth a little less than the one before it. ICM shifts that "excess" value from the big stacks to the shorter ones, who are more likely to ladder into the money.
No. ICM assumes every chip is equally likely to end up with any player, ignoring skill edges, position and future blind levels. That makes it a model rather than the whole truth, but it is accurate enough to be the standard for deals and bubble decisions.
Knowing the number is one thing; recognizing the bubble and final-table spots where ICM should change your play, in real time, is another. That comes from reviewing your own tournament hands and seeing where the money actually leaked. Poker Spot tracks every tournament you play, saves the pivotal hands, and lets you replay them with the context attached, so the ICM lessons stick. Pair this with the preflop charts for the tighter tournament ranges and the odds calculator when you need the raw equity behind a call.