Standard push/fold charts come with an invisible asterisk: they assume every chip is worth exactly the same amount of money. That assumption is fine early, harmless in a cash game, and dangerously wrong at a final table. Understanding the gap between a Nash (chip-EV) chart and an ICM-adjusted one is the difference between laddering up the payouts and busting in a spot the chart called "correct."

Our push/fold chart shows chip-EV (Nash) ranges — the right baseline. This article is about when to deviate from it.

What a Nash push/fold chart actually solves

A Nash equilibrium push/fold chart answers a clean game-theory question: if both players know it's a shove-or-fold game, what shoving and calling ranges make each player unexploitable in terms of chips won? It's a two-player solve, repeated for every stack depth, and it's an excellent default. When chips equal money — early tournament, deep field, cash-game-like spots — the Nash chart is simply correct.

The key phrase is "in terms of chips won." Nash maximizes chip EV. It has no idea a pay jump exists.

Why chips stop equaling dollars

The Independent Chip Model (ICM) converts your chip stack into a share of the prize pool based on the payout structure. Its central, counterintuitive result: the chips you can lose are worth more than the chips you can win. Doubling from 10bb to 20bb does not double your money equity — but busting takes you from a guaranteed payout to zero. Your stack's dollar value is a curve that flattens as you climb, not a straight line.

That asymmetry is the whole story. Because losing hurts more than winning helps, any decision that risks your tournament life needs more edge than chip-EV alone requires. That "extra edge required" has a name.

The risk premium

The risk premium is the additional equity, beyond the chip-EV break-even, that you need before calling off your stack under ICM pressure. If a call is break-even in chips but you're on a pay ladder, ICM turns it into a clear fold — you needed, say, 38% equity instead of the 33% the pot was laying.

The shape of the adjustment

Chip-EV (Nash) call threshold:  ~33% equity
+ Risk premium near a pay jump:  +5% to +12%
= ICM call threshold:  38%–45% equity

Bigger pay jumps & more laddering → larger risk premium.

The risk premium isn't one number — it scales with how much money is on the line and who's at risk. A satellite on the bubble (where every survivor wins the same seat) produces a gigantic risk premium; a winner-take-all sit-and-go produces almost none. Most real final tables sit in between.

How ICM reshapes your ranges

ICM pushes your ranges in two directions at once, and confusing them is the most common mistake:

  • Calling ranges tighten — a lot. Calling a shove risks your stack with no fold equity to compensate. This is where the risk premium bites hardest. Hands that snap-call in a Nash chart become folds near a pay jump.
  • Shoving ranges tighten less — and sometimes widen. When you shove, you still get fold equity, and forcing opponents into ICM-pressured decisions is profitable. Against players who can't call (because they're protecting a ladder), your profitable shoving range can actually get wider.

So the final-table adjustment isn't "play tighter." It's "call much tighter, and shove selectively wider against the players who can least afford to call." That's the heart of bubble and final-table strategy.

A concrete example

Same hand, two different answers

You're in the big blind with A9s, 12bb effective. The button shoves. By the Nash calling chart, A9s is a comfortable call against a button shoving range — you have well over the ~33% chip-EV threshold.

Early in the tournament: snap-call. Chips are money; the chart is right.

Three off the final table, you're a medium stack, two players are critically short: fold. The risk premium pushes your required equity toward 40%+, A9s no longer clears the bar, and there are two stacks more likely to bust before you. Folding into a pay jump is worth more than a thin chip-EV edge.

Same hand, same stack, same shove. Opposite correct plays — because the money changed, not the cards.

When to trust Nash, when to override it

You don't need an ICM solver running in your head at every table. You need to know which situation you're in:

  • Trust the Nash chart early, with a deep field, far from the money, and in flat or winner-take-all structures. This is most of the hands you'll play.
  • Apply a risk premium on the bubble, on pay-jump ladders, at the final table, and especially in satellites. Tighten calls first and hardest.
  • Attack with the chip lead. If you cover the table, ICM pressure is your weapon — open-shove wider into players who can't call without risking a ladder.

Start from the right baseline

The chip-EV chart is your foundation — learn it cold, then layer ICM on top.

♠ Open the Push/Fold Chart

Build the chip-EV base first: read push/fold charts by stack depth, then drill the spots in the preflop range trainer. Once the Nash ranges are automatic, the ICM overlay becomes a small, fast adjustment instead of a guess. For the bubble specifically, see ICM bubble strategy.