In a cash game a chip is a dollar, and doubling your stack doubles your money. Tournaments do not work that way — doubling your chips does not double your prize expectation — and that gap changes how you should play. ICM is the model that describes it.
- ✓Tournament chips cannot be cashed. What they represent is your probability of finishing in each paying position.
- ✓The more chips you hold, the less each additional one is worth.
- ✓That means losing chips hurts more than winning the same number helps.
- ✓The bubble and the final table are where this matters most, and folding becomes correct more often than it would in a cash game.
- ✓Short stacks carry the heaviest ICM pressure; big stacks get to apply it.
Why chips are not money
Take a hundred-runner tournament, a thousand per entry, paying the top ten.
Your chips cannot be exchanged for cash. Their only function is to keep you alive long enough to reach a paying position. So what a chip is actually worth is how much it raises your chance of finishing in the money.
Here is the catch: the prize pool is fixed. When your stack goes from ten thousand to twenty thousand, your chance of winning does improve — but by less than double. You already had some equity before, and first prize does not grow because you hold more chips.
That is ICM in one line: the marginal value of a chip decreases as your stack grows.
The consequence: losses hurt more than wins help
If winning ten thousand chips adds less prize equity than losing ten thousand takes away, then a coin-flip all-in is a losing proposition in a tournament.
This is the single biggest strategic difference between formats:
- Cash game: any edge above 50% makes the all-in worth taking.
- Tournament: a 50% shot is often a fold, because the downside outweighs the upside.
Exactly how much equity you need depends on the payout structure and how the other stacks are distributed. The more top-heavy the payouts and the closer you are to the money, the higher the bar.
The bubble
The bubble is the stretch just before the money starts. During it:
- Every short-stack decision is expensive. Busting pays nothing, and surviving a few more hands pays something.
- Medium stacks are the most constrained — unwilling to bust, unwilling to sit still.
- Big stacks can apply pressure, because their own risk of elimination is lowest.
Practical adjustments:
- Tighten your calling range against potential all-ins.
- Widen your opening range against players who cannot afford to call.
- Remember that a shorter stack than yours at the table makes your own position safer than it looks.
The final table
Each elimination moves everyone up a payout step, and the steps get larger near the top. That makes ICM even sharper here than on the bubble.
- Three-handed, the gap between third and first can be several times the buy-in. Clashing with the other big stack is costly.
- When someone is about to be blinded out, the others have an incentive to wait — that elimination is a free pay jump.
- The short stack, conversely, should act. Waiting only shrinks the stack further.
Applying it at the table
Nobody calculates ICM in real time. In practice it comes down to a few questions:
- How expensive is busting right now? Closer to the money and steeper payout jumps both raise the cost.
- Is there a shorter stack than me? If so, let them absorb the pressure first.
- Could this decision eliminate me? If not, ICM barely applies.
- Is my opponent under the same pressure? Big stacks pressure short stacks precisely because it is asymmetric.
Review hands afterwards with an ICM calculator. That is far more useful than trying to do the arithmetic mid-hand.
