Sunday Million ICM, a worked walkthrough
Eight-handed final table ICM, with the calculator outputs. The chip leader is worth $25,800 less than their chip share. Here's why.
The Sunday Million final table is one of online poker's classic ICM spots: eight runners, top-heavy prize pool, dramatic pay jumps between every position. It is also one of the cleanest worked examples of why the Independent Chip Model exists, and why every serious tournament player should be running these numbers in their head before they snap-call a marginal jam.
Here is a worked example using the ICM deal calculator with realistic numbers from a recent final table. The exact stacks and pay levels matter less than the structure of the answer; the lessons generalise to every ICM spot a tournament player encounters.
Stacks and payouts
Stacks at the start of the final table, chip leader to short stack: 28 million, 24 million, 18 million, 15 million, 12 million, 9 million, 6 million, 3 million. Total chips in play: 115 million.
Payouts, first place down: $187,000 / $137,000 / $99,000 / $73,000 / $54,000 / $40,000 / $30,000 / $22,500. Total prize money remaining: $642,500. Pay jumps are uneven: the jump from 8th to 7th is $7,500, the jump from 2nd to 1st is $50,000. That asymmetry is the entire reason ICM exists.
The chip leader is worth less than their chip share
Run those numbers through the ICM deal calculator and the chip leader's dollar equity is $112,400. Their chip-share equity, which is what you would get if all eight players agreed to chop the remaining prize pool proportional to chips, would be $138,200. The chip leader is worth $25,800 less in ICM-equity terms than in chip terms.
The reason is structural. The chip leader cannot win first place twice. Every chip they accumulate beyond the next pay jump returns a smaller and smaller marginal payout, because the only payout left to win is the gap to first. Meanwhile, the short stack is worth $22,500 in locked equity (eighth-place money) regardless of how few chips they hold. Their dollar equity is $34,200, well above their chip share of $14,400. The model rewards survival at the bottom and penalises hoarding at the top.
What this changes about play
The strategic implications cascade through every decision at the table. The 9 million and 12 million stacks, the medium stacks, are in the worst spot in the room. They have real ICM downside (busting in the next two spots costs them roughly $20,000 in equity each) and limited ICM upside (doubling up only moves them toward second-place money, not first). The correct response is to fold marginal hands the chip leader is jamming, even when those hands are genuinely ahead of the jamming range in pure chip-EV terms.
The 3 million stack, the short stack, should be widening shove ranges aggressively. They have minimal ICM downside (busting only costs $11,700 from their already-locked $22,500) and large ICM upside on a successful double. Any hand that is a coinflip or better in chip-EV is a clear shove for the short stack.
The chip leader, counterintuitively, should be tightening up some of their open ranges and loosening up almost all of their jamming-over ranges. They have the most ICM downside (every chip lost is worth more than every chip won) but they also have a uniquely powerful jamming weapon because the medium stacks cannot call without burning equity.
How to use this at the table
The full Malmuth-Harville recursive formulation is too slow to run in your head at the table. The shortcut every serious MTT player uses is to memorise three things: roughly how much ICM equity the chip leader gives up versus chip share at common final-table stack distributions, roughly how much the short stack picks up, and roughly the inflection point where a medium stack should fold marginal hands the short stack is shoving. Drill those three numbers and the ICM model becomes a usable real-time tool, not just a post-session study aid. The ICM deal calculator on this site uses the Malmuth-Harville recursive formula and handles final-table chops to the cent.