Final Table ICM, Seat by Seat: Opening, 3-Betting and Jamming by Stack

Every preflop chart you have ever drilled makes one quiet assumption: that everyone at the table has the same stack you do. Open the 25bb chart, memorize the borders, done. Then you sit at a real final table, or the bubble that feeds it, and the stacks look nothing like that. A 35bb chip leader on your left, a 10bb short two seats over, blinds down to 12bb. We pulled a real solved eight-handed table from GTO Ranges+ where the stacks run from 10bb to 35bb, all under final-table ICM (Independent Chip Model) pressure, and read every seat against the flat 25bb chart. The gaps are not small, and they run through every decision: opening, three-betting, rejamming, and open-jamming.

Three archetypes matter. The chip leader (CO, 35bb) covers everyone and risks only chips. The covered mid-stacks (UTG through HJ, 19-33bb) have chips but can be busted by the leader, so they carry the most to lose. The short stacks (BTN 10bb, blinds 12-16bb) are playing a jam-or-fold game where every chip is a survival decision. Watch how each one plays the same four decisions.

Opening: the leader widens, the shorts switch weapons

On the flat 25bb chart, opening width is a clean function of position: tighter early, wider late, and nobody jams. At the real table, stack size overrides that.

The chip leader opens 38.2% from the cutoff, well past the 27.1% the flat 25bb chart assigns that seat. It can, because busting is off the table for the one stack that covers the room. Meanwhile the short button does not have a raise-or-fold decision at all. Its menu is min-raise or ship, and it takes the ship far more often than it min-raises. Here is the whole table's opening plan next to the flat baseline.

Two things jump out. The covered mid-stacks (HJ, MP) actually open a touch wider than their flat-chart twins, because they have real fold equity over the shorter stacks behind them. And the short button trades a wide raising range for a jamming range: 26.7% of hands go all-in, because a 10bb open-raise commits you anyway, so you may as well deny the reraise and take the fold equity now.

Open-jamming: a weapon that only exists when you are short

Notice the open-jam column is zero for every stack above 16bb and lights up only for the short button. That is the point. Open-jamming is not a style, it is a stack-depth tool. Somewhere around 12-15bb the min-raise stops making sense because it commits too much of your stack to fold, and the shove takes over. On the flat 25bb chart the jam never appears, which is exactly why players who only study those charts freeze when they get short: they have no memorized answer for the one move that matters most at 10bb. If you want the exact ship-or-fold borders by depth, the Final Table Push-Fold Charts for Every Stack Depth map them stack by stack.

Three-betting: pressure comes from the covered stacks, not the shorts

Now UTG opens and the table responds. On the flat chart, everyone mixes a few small three-bets and the occasional jam at similar frequencies. At the real table the response splits sharply along stack lines.

The covered deep stacks (MP, HJ, and the CO leader) three-bet around 8%, roughly four times their flat-chart rate. They can apply that pressure cheaply because a small three-bet risks chips they can reload from, not their tournament life. The short stacks do almost the opposite: their small three-bet collapses toward zero and their aggression moves entirely into the rejam column. This is the single most useful adjustment to internalize. When you are deep and covered, a light three-bet is a pressure tool. When you are short, the three-bet is a trap that bloats a pot out of position, so you ship or you fold.

Rejamming: the short stack's real continuing range

The short stacks rejam over UTG's open two to three times as often as the flat 25bb chart suggests: the button at 9.8%, the big blind at 8.3%, the small blind at 6.4%. The big blind also flats 19% of hands, defending wide because it is closing the action and getting a price. But its raising range is not a three-bet, it is a shove. Look at what the button actually does with a continue.

Same seat, same open in front of it, and the button's entire toolkit changes with its stack. At 25bb it keeps small three-bets in the mix. At 10bb those are gone, replaced by a wider all-in range that reclaims fold equity a flat call would surrender. Studying the flat chart, you would three-bet 8bb of your stack and fold to a jam. Studying the real spot, you rip it in and make the covered opener decide for their tournament. This is the same force that makes the Surviving the Money Bubble: ICM Pressure and Optimal Play so punishing, and it is why Defending the Big Blind at 15bb: The Most Critical Tournament Spot You Are Misplaying looks nothing like the deep-stacked version.

The takeaway: your range is a function of everyone's stack

The flat 25bb chart answers one question: what do I open when everyone is 25bb deep. A real table asks a harder one: what do I do when I am 10bb, the leader who covers me is 35bb, and busting before the two shorter stacks costs me a pay jump. The answers are not tweaks to the chart, they are different decisions. Open-jamming exists at 10bb and vanishes at 25bb. Three-betting is a big-stack pressure tool, not something the short stacks should copy. Rejamming replaces three-betting as you get short. And the chip leader plays the widest, most aggressive game at the table precisely because it is the one stack ICM cannot punish. To drill these by your exact stack, your opponents' stacks, and the payout pressure in front of you, GTO Ranges+ carries the solved ICM ranges for uneven tables at every seat and depth, with a trainer that tracks streaks and flags the spots that keep costing you accuracy in its mistake-breakdown reports. The ICM Deal Calculator shows the payout math that drives all of it. Learn the real table, not the flat chart, and the sharp end of a tournament stops being a guess.