Push/Fold Chart: When to Shove or Fold a Short Stack

Shove-or-fold is not how you play a whole tournament. It is how you play a short stack, and outside of that range it is wrong. The framework is correct from roughly 1 big blind up to about 15, with a little extra room from the very last positions where almost no players are left to act behind you. Inside that band, moving all-in or folding is genuinely the highest-EV way to play a hand when the action folds to you.

Picture why. You have 12 big blinds and you make a standard raise to 2.5. Now you have 9.5 behind and a small pot in front of you. If anyone moves all-in over the top, you are trapped: the price is too good to fold comfortably, but calling puts your tournament on the line with a hand you only opened. The raise folded nobody, protected nothing, and cost you the freedom to get away. Below about 15 big blinds, small raises leak value like this over and over, so you strip the decision down to its profitable core - commit everything at once, or commit nothing.

Above that band, you should be open-raising - putting in a small raise and keeping chips behind to play after the flop - not jamming your whole stack in. With 25 big blinds, a 2 big-blind raise risks a fraction of your stack to win the pot, while an all-in risks everything to win the same blinds and antes. The math that makes shoving profitable when you are short falls apart when you are deep. The chart above will still draw shove ranges up to 25 big blinds, since the model runs that far, but treat the deep-stack cells as theory rather than advice. In practice, from about 15 big blinds up, you are raising, not shipping it. Our lesson on M-ratio and the push/fold zones covers how to tell when you have actually arrived in shove-or-fold territory.

The grid is the standard hand matrix: 13 ranks across and 13 down, every possible starting hand shown exactly once. The diagonal from the top-left corner to the bottom-right is the pocket pairs. Everything above that diagonal is the suited hands (both cards the same suit); everything below it is the offsuit hands. Aces sit in the top-left corner, so the strongest hands cluster there and the range blooms outward from that corner as it widens.

Set two things, read one answer. Choose your position and drag the slider to your effective stack in big blinds. Every hand that is a profitable all-in lights up green; every hand you should fold stays dark. The number printed in each green cell is the deepest stack, in big blinds, at which that hand is still a shove - so a cell reading 12 means the hand is a shove at 12 big blinds or fewer and a fold above that. Watch the green spread as you drag the stack down. That spreading green is your range widening as your stack shrinks, and it is the single most important pattern in short-stack play.

The stack figure on the chart is your effective stack, and in an all-in it is the only stack that counts. Effective stack is the smaller of the two stacks in a confrontation, because that is the most anyone can actually win or lose. If you have 30 big blinds and shove into a big blind who has 8, only 8 are ever at risk - you are playing an 8 big-blind pot, so you read the 8 big-blind cell, not the 30. When you are the short stack, your own stack is the effective one. When someone covers you, use yours. When you cover a player you might get all-in against, use theirs.

One more condition. The shoving side of this chart is a first-in guide: it answers "everyone has folded to me, do I move all-in?" The moment a player has entered the pot ahead of you, you are no longer first in, and a tighter decision applies, because now you can be called by a hand that has already shown strength by limping or raising. So the routine at the table is: find your seat, convert your stack to big blinds, confirm the pot is still unopened, and read the cell.

The first time players see a correct short-stack range, they assume it is a misprint. At 10 big blinds you are shoving hands you were taught to fold. There are two reasons, and understanding them is worth more than memorizing any single cell.

The first is fold equity: the profit you make on all the times everyone folds and you collect the blinds and antes without a fight. When you move all-in with a short stack, you have two separate ways to win - everyone folds right now, or someone calls and your hand holds up. Most beginners count only the second one, so they only shove hands that win showdowns. That is the core mistake. King-seven offsuit on the button at 10 big blinds is a clear shove not because it is strong, but because it wins the blinds and antes uncontested often enough to profit on its own. You are not betting that your hand is good. You are betting that nobody still to act behind you has a hand good enough to call.

The second is the ante. This chart models the modern big blind ante: a full 1 big-blind ante posted before every hand, on top of the 0.5 and 1 big-blind blinds. Add it up and there are 2.5 big blinds of dead money sitting in the middle before a single card is dealt. Dead money is money nobody has a claim on yet, and fold equity is how you collect it. When you shove 8 big blinds to pick up 2.5 that are just lying there, you are risking your stack for a reward worth almost a third of it every time everyone folds - and that reward is fixed regardless of your cards, so it pads the value of every hand you jam, the weakest ones most. Antes are the single biggest reason short-stack ranges are as wide as they are. Take the ante away and the green shrinks dramatically. It is not recklessness; it is arithmetic.

Drag the position selector from early to late and the green explodes. That is not a quirk of the model; it is the whole point of position in a push/fold world. This chart is indexed by the number of players left to act behind you, and that number is everything, because every player still to act is another chance for someone to wake up with a strong hand and call.

From early position, with seven or eight players yet to act, you need a genuinely strong hand, because you have to get through all of them. On the button, with only the two blinds behind you, you can shove a huge range, because there are just two hands left that could call and both will usually miss. This is also why the chart is correct for six-handed tables, not only full ring: a cutoff at a 6-max table and a cutoff at a 9-handed table both have three players left to act, so they face the same decision and read the same cell. Count the players behind you, not the seats at the table. If short-handed play is where you keep landing, Part Eleven on short-handed tables drills it directly, and Part Twelve takes it down to heads-up, where the button acts first and almost every hand is a short-stack shove-or-fold decision.

The chart also includes a big blind calling grid: when someone else shoves and you are sitting in the big blind with a short stack, which hands do you call with? That grid is tighter than the shoving grid, and the reason is a clean piece of logic worth holding onto - the player who shoves can win two ways, but the player who calls can only win one.

A caller needs less raw equity than you might expect. Because there is already so much in the pot - the blinds and antes, plus the chips the shover just committed - you are getting pot odds: you risk fewer chips than you stand to win, so your break-even point falls below 50% equity. Against a wide late-position shove the big blind can correctly call surprisingly light, because it is being paid a price. But a caller has no fold equity. When you call, you can only win by having the best hand at showdown; you have given up the "everyone folds" way of winning entirely. The shover wins two ways, the caller wins one. That missing second way is why calling ranges come in tighter than shoving ranges even though the caller is getting the better price on the chips. When you shove you are selling a hand that might be behind; when you call you are buying a price. In doubt, it is almost always better to be the player shoving than the player calling a shove.

Drag the slider deeper and watch the green retreat. It is the same fixed reward working in reverse: the dead money you are shoving to win stays at 2.5 big blinds no matter how deep you are, but the amount you risk to win it climbs with your stack. At 5 big blinds you risk little to collect that reward; at 20 you risk a tournament-defining pile for the same 2.5, so only genuinely strong hands stay profitable. Deeper stack, tighter shove. The uncontested blinds and antes are the free money that makes loose shoves pay, and the deeper you are, the smaller that free money looms against everything you are putting at risk.

Notice that shoving too early feels reckless, but shoving too late is the more common and far more expensive error. At 12 big blinds your all-in genuinely threatens people. At 3 it does not, because the pot is laying callers a price they cannot refuse, so they call and you are back to needing the best hand. Between those extremes is the band where fold equity does the most work, and it is exactly the band this chart is built for.

A chart you cannot see inside is a chart you should not trust, so here is exactly what this one is built on. It is a free tool, with nothing to sign up for and nothing to install.

• It is a Nash equilibrium. The ranges are game-theory-optimal (GTO): the mathematically unexploitable shove-and-call solution, the point where neither side can improve by deviating. It assumes your opponents also play correctly. Against players who fold too much or call too much you can profitably deviate from it, but the equilibrium is the correct baseline to learn from.
• Blinds of 0.5 / 1, plus a 1 big-blind ante every hand - the modern big blind ante structure, which puts 2.5 big blinds of dead money in every pot before the cards are dealt.
• Indexed by players left to act behind you on a 9-handed structure, which is what makes it correct for 6-max as well. Read it by the number of players behind, not by the size of the table.
• Chip-EV. It assumes every chip is worth exactly the same amount of money. The hand-versus-hand equities come from Monte Carlo simulation - dealing out many random boards for every matchup across the full 169x169 preflop all-in grid and counting who wins - so the equity behind each cell is measured, not eyeballed.
• Stacks from 1 to 25 big blinds. Useful across the entire short-stack range, with the caveat above that the deepest cells are theory rather than table advice.

There is no claim here beyond what a Nash shove/fold solve actually is.

Here is the caveat that matters more than all the others, and the one most push/fold charts online quietly skip. This chart is chip-EV: it treats every chip as worth the same fixed amount of money. In a tournament that is not quite true, and near the money it is badly untrue.

Because tournament payouts are top-heavy and busting pays you nothing, survival has a value that raw chips do not capture. The chips you lose cost you more than the chips you win gain you. That idea has a name - ICM, the Independent Chip Model - and its effect on push/fold is direct. Near a bubble (the point where one more elimination puts everyone left in the money) or at any pay jump, you must shove tighter than this chart says, and call all-ins much tighter, because stacking off and busting is far more expensive than the chip-EV math admits. Calling is where players bleed the most tournament equity, so that is the range to tighten hardest.

Treat this chart as your baseline for a chip-EV world: deep-field early play, or any spot with no pay jump in sight. Then tighten from it as survival starts to matter. It will not tell you how much to tighten - that judgement is the skill. Part Ten covers adjusting to stack sizes and ICM near pay jumps, and it is the piece that turns a chart into good tournament decisions.

The chart is a reference, but you will not have it open at the table, and you would not have time to read it hand by hand if you did. The real goal is to internalize the shapes: strong and narrow from early position, wide open from the button, wider as the stack shrinks, tighter as the money approaches. Once those patterns are automatic, you can make the decision in the two seconds you actually get.

That comes from reps, not reading. Part Nine on M-ratio and the push/fold zones explains when you have entered shove-or-fold territory in the first place, with M-ratio - your stack divided by the cost of one full orbit - being the cleanest measure of how short you really are. From there, the full trainer puts you in the seat: 136 lesson hands played street by street, where you commit to a decision before you see the reasoning, including the short-stack spots this chart is built for. If you are new to tournaments entirely, start with what an MTT is and work forward.

A push/fold chart is a short-stack guide that tells you, for a given position and stack size, exactly which hands to move all-in with and which to fold. When you have only a few big blinds left, raising small stops working, so your realistic options collapse to shoving all-in (push) or folding. The chart maps that decision hand by hand. The interactive one above lets you set your position and effective stack and lights up every profitable shove in green.

Use pure push/fold when you are short, roughly 1 to 15 big blinds, with a little extra room from the very latest positions. Inside that band a small raise leaves you too committed to fold yet too shallow to play well after the flop, so jamming all-in captures more value when the action folds to you. Above about 15 big blinds you should be open-raising and keeping chips behind, not jamming, so the deep-stack cells on the chart are theoretical rather than practical advice. See our lesson on adjusting to different stacks.

Far more than most beginners expect, and it depends heavily on your position. Set the chart above to 10 big blinds and your seat to see the exact range. From late position at 10 big blinds the shove range includes many hands players are taught to fold, because at that depth you win most of the profit from fold equity - everyone folding and handing you the blinds and antes - rather than from winning a showdown. From early position, with more players still to act, the range tightens sharply.

Two reasons. First, fold equity: when you shove short, everyone often folds and you collect the blinds and antes uncontested, which is profitable even with a mediocre hand. Second, the ante. A modern big blind ante drops extra dead money into the pot before anyone acts - 2.5 big blinds in total with the blinds - and the more dead money there is to win, the more hands become a profitable shove. Together they widen correct ranges well past what instinct suggests.

Yes, dramatically. Antes are the single biggest reason short-stack shoving ranges are as wide as they are. An ante adds dead money to the middle before anyone acts, and fold equity is how you collect it, so every extra chip of dead money makes more hands a profitable all-in. This chart models a modern 1 big-blind ante, which is why the ranges look wide. Strip the ante out and the profitable shoving ranges tighten considerably.

Yes. The chart is indexed by the number of players left to act behind you, not by the number of seats at the table. A cutoff at a 6-max table and a cutoff at a 9-handed table both have three players behind them, so they face the same decision and read the same cell. Count the players yet to act behind you and use that, and the chart is correct at any table size.

It is a Nash equilibrium shove-and-call solution, which is the game-theory-optimal, unexploitable baseline for first-in shoving with a short stack. It assumes your opponents also play correctly; against players who fold or call too much, you can profitably deviate. What it is not is ICM-adjusted: it values every chip equally, so it is GTO in the chip-EV sense, a baseline to tighten from near a pay jump rather than a complete answer.

Use it as a starting point, then tighten. This chart is chip-EV, meaning it treats every chip as worth the same fixed amount. Near the bubble or a pay jump, survival is worth extra because busting pays nothing, so you should shove tighter than the chart says and call all-ins much tighter. Calling is where the most tournament equity is lost, so tighten that range the hardest. Part Ten covers exactly how to adjust for ICM near pay jumps.

Effective stack is the smaller of the two stacks in a confrontation, measured in big blinds. It matters because that is the most anyone can actually win or lose in the hand. If you have 30 big blinds and get all-in against someone with 8, only 8 are at risk, so you read the 8 big-blind ranges. When you are the short stack, your own stack is the effective one; when you are covered, use yours; when you cover your opponent, use theirs.

Because calling gives up fold equity. When you shove, you can win two ways: everyone folds, or you get called and win the hand. When you call a shove, the folding way is gone and you can only win at showdown. A caller actually needs less raw equity thanks to the good price the dead money offers, but with no fold equity to add on top, the calling range still comes in tighter than the shoving range. When unsure, it is usually better to be the one shoving.

Yes. The chart above is a free tool, and so is the rest of the site: 136 lesson hands played street by street plus a graded exam. There is nothing to buy and nothing to install.