Satellite Bubble Calculator

A satellite awards entries into a bigger event rather than a cash ladder. Finish in the top four of a five-handed satellite that awards four seats and you win a seat. Finish fifth and you win nothing. Crucially, the seat you win for finishing first is identical to the seat you win for finishing fourth.

In a normal tournament, chips have a sliding value: more chips means a better shot at the top of the payout ladder, so accumulating always has some upside. In a satellite, that upside gets capped hard. Once you have enough chips to be safe, extra chips buy you almost nothing, because there is nothing above a seat to buy. Meanwhile the downside is unchanged: losing your chips still costs you the entire seat.

That asymmetry is the whole game. Every risk you take is priced against a prize you cannot improve on. The result is a style of play that looks absurdly passive to anyone used to regular tournaments and is in fact ruthlessly correct.

If you are new to tournament structures generally, start with what MTT poker is and come back. The strategy here assumes you already know how a normal bubble works.

Here is the situation the calculator above is built to explain. Five players remain. Four seats are awarded, and all four are worth exactly the same. The stacks are:

• You: 15,000
• The shover: 10,000
• Three short stacks: 4,000, 3,000 and 2,000

The 10,000 stack moves all in. You are next to act with 15,000 behind. The effective stack, meaning the smaller of the two stacks involved and therefore the most you can actually lose in the hand, is 10,000.

You look down at pocket aces. Against a single random hand, all in preflop, aces have 85.25% equity, counting split pots as half a win. That figure comes from this site's own Monte Carlo simulation over 400,000 trials with a fixed seed. The published textbook number is about 85.2%, so the two agree.

You are being offered the best starting hand in poker, face up, against a hand that is worse. You should fold.

Take it slowly, because the reasoning is what matters, not the conclusion.

Step one: what you are worth right now. With 15,000 chips at a five-handed table paying four seats, your seat equity is 98.70%. You are not certain to win a seat, but you are close. Four of the five players get one, you have the biggest stack, and three opponents are sitting on 4,000, 3,000 and 2,000. All you have to do is not be the one who goes broke.

Step two: what you gain by winning. If you call and win, your seat equity is 100.00%. That is a gain of 1.30 points. Notice what happened: by winning, you eliminated the shover, which leaves four players competing for four seats. Everyone still standing has a seat. You did not just lock up your own seat, you handed one to each of the three short stacks as well. All those chips you just won are worthless to you, because there was nothing left to buy.

Step three: what you lose by losing. If you call and lose, you are down to 5,000 chips and your seat equity falls to 88.17%, which the engine prices as a loss of 10.52 points. You are still alive, and still likely to make it, but you have gone from near lock to genuinely at risk.

Step four: the ratio. You are risking 10.52 points to win 1.30. That ratio is your bubble factor, and the engine puts it at 8.089. Losing this pot hurts you roughly eight times as much as winning it helps.

One note before the last step. The equities on this page are rounded to two decimals so they are readable, so subtracting or dividing the rounded figures can land a hundredth away from the engine's own answer. The engine carries the full precision and the figures quoted here are its output, not the result of the arithmetic you can do on screen.

Step five: the break-even number. Folding leaves you at 98.70%, so calling has to beat that. Calling gains 1.30 points some of the time and costs 10.52 points the rest of the time, and the two balance when 1.30 times your win rate equals 10.52 times your loss rate. That happens at 89.00% equity. Aces have 85.25%. You are 3.75 points short, so calling is a losing play with the best hand in poker. Fold.

That is the argument, and it is worth seeing what it rests on. It assumes folding leaves you exactly where you are, at 98.70%. It assumes nothing about the shover beyond a completely random hand. And it leans on the stacks as much as on the payouts: three players small enough to bust ahead of you are what make your seat so close to locked. One thing that cannot rescue the call here is dead money, because winning already takes you to 100.00% and there is nothing above that to collect. Change the stacks or the seat count and the number changes. The direction does not.

To see how much of this is the satellite and how much is the stack configuration, hold the stacks and the shove fixed and change only how the prizes are paid.

• Winner take all: bubble factor 1.000, you need 50.00% to call, and aces are an easy call.
• Normal ladder paying 50/30/20: bubble factor 1.568, you need 61.05% to call, and aces are still a comfortable call.
• Satellite awarding 4 seats: bubble factor 8.089, you need 89.00% to call, and aces are a fold.

Identical chips. Identical hand. Identical opponent. The only thing that moved was the prize structure, and it moved the calling requirement from 50.00% to 89.00%. That is what ICM pressure is. It is not a vague instruction to play tight. Within the model it has a specific size you can compute before you act, and that size is what tells you how tight to be.

To see the middle case for yourself, switch the calculator above to the normal ladder preset. The bubble factor calculator covers ordinary tournament bubbles in more depth.

Hold the other four stacks fixed and vary only your own, so you can see what your stack alone is worth. The calculator shows these three figures for whatever stack you enter. They are not all states this table can reach, since the chips you win have to come out of somebody else's stack, but the shape of the curve is the point:

• 15,000 chips: 98.70% seat equity
• 30,000 chips, double: 99.80%
• 60,000 chips, quadruple: 99.98%

Doubling your stack to 30,000 buys you 1.10 points of seat equity. Quadrupling it to 60,000 buys 1.28 points, so the second doubling adds barely a fifth of a point. You cannot buy more than the 1.30 points that separate you from a locked seat, because a seat is the ceiling. Every chip you win past the point of safety is a chip you are risking real equity to acquire for almost no return.

Compare that to the downside. Busting out takes you from 98.70% to zero. The curve is steep on the way down and almost flat on the way up. When a payoff structure looks like that, the correct response is to stop taking bets, and that is precisely what good satellite players do.

This is also why the standard tournament instinct to build a stack for the final table needs rewiring here. Chips are still worth having, because they buy fold equity, meaning the chance that opponents fold to your bet rather than call it, and because they outlast the blinds. What they are not worth is gambling for. There is no final table prize to build toward, only a seat, and you either get one or you do not.

The full picture for five players and four seats:

• 15,000 chips: 98.70%
• 10,000 chips: 96.52%
• 4,000 chips: 80.80%
• 3,000 chips: 71.11%
• 2,000 chips: 52.87%

Those five percentages sum to 400.00%, that is 4.000 seats, which is exactly the four seats being handed out. That is a useful sanity check on any satellite ICM output: the seat equities must add up to the number of seats, because every seat goes to somebody.

Read down that list and the strategic map draws itself. The 2,000 stack is close to a coin flip to survive and has to act. The 3,000 and 4,000 stacks are in real danger but not desperate. The 10,000 and 15,000 stacks are nearly home and have almost nothing to gain from a confrontation. Nobody at this table wants to play a big pot except the players who have no choice, and the players who have no choice are the ones you least want to be.

Turn all of the above into table behaviour.

If you are comfortably above the threshold, refuse every big pot. Do not call all ins without an enormous edge, and remember that with aces in the flagship spot you did not have an enormous enough edge. That is not the same as folding every hand. When you are first into the pot you should be attacking, because the players behind you can almost never call. Passive when facing action, aggressive when opening, is the whole big stack strategy.

Attack the player closest to locking a seat that you can still cover. The more nearly locked up an opponent is, the more a lost pot costs them and the less a won pot gains them, so their calling requirement is the highest at the table. In the flagship table that is the 10,000 stack at 96.52%, not the shorter stacks at 80.80% and 71.11%. Switch the "You are" picker to their seat and read their row: you will see how little they can call with. This is the one aggressive move that survives satellite math.

Avoid the players who must gamble. The shortest stack has to get their chips in soon, but do not mistake that for having nothing to lose. At 52.87% they still hold half a seat and they are picking their spots. What is true is that you gain very little by being the one who knocks them out, since the blinds will often do it for you. There is no need to tangle with them.

If you are the short stack, act before you are forced to. The 2,000 stack at 52.87% still has real seat equity, and it still has fold equity, meaning the chance that opponents fold to a shove rather than call it. Shoving while opponents can still fold is worth far more than shoving after the blinds have eaten you alive. A push fold chart gives you the baseline ranges, and the M ratio calculator tells you how many orbits you have left before the decision is made for you.

Count the seats out loud. The single most valuable habit on a satellite bubble is knowing how many players need to bust before the tournament ends. When one bustout ends it, folding every hand is often literally the highest equity play available to a big stack.

Enter the number of seats being awarded and each player's chip stack, then use the "You are" picker to select your seat and the "All in against" picker to select the player who has shoved. Payout structure presets let you switch between the satellite, a normal ladder and winner take all so you can reproduce the comparison above yourself.

The output shows the effective stack, your seat equity now, your seat equity if you win, your seat equity if you lose, your bubble factor, the required equity to call, and a verdict comparing that requirement against the 85.25% that pocket aces have against a random hand. Below that it lists your bubble factor against every other player at the table, one row per opponent.

Two things to keep in mind about what you see on screen. Bubble factor displays to two decimal places, so the flagship 8.089 appears as 8.09, and required equity displays to one decimal place, so 89.00% appears as 89.0%. The tool shows one player's set of bubble factors at a time rather than the full matrix, so switch the "You are" picker to see the table from somebody else's chair. That is often more instructive than looking at your own row, because it tells you which opponents can afford to call you.

For non satellite structures, the general purpose ICM calculator handles cash ladders.

Every number on this page is exact for the model it comes from, and the model is a simplification. Be clear about where it stops.

It ignores blinds, antes and position. ICM treats your stack as a static claim on the prizes. In reality you are about to post a big blind, the shortest stack may be one orbit from elimination, and being on the button matters. A stack that looks safe on a spreadsheet can be under real pressure two hands later.

It settles finishing order by repeated chip share. The Malmuth Harville method used here gives each player a chance of finishing first equal to their share of the chips in play, then removes that player and applies the same rule to whoever is left to settle second, and so on down to last. That is a reasonable approximation, not a description of how a real table plays, and it slightly overstates the equity of short stacks in some spots.

It ignores skill. If you are markedly better than the table, survival is worth more to you than the model says. If you are the weakest player, gambling costs you less than the model says.

The random hand assumption is the clean case, not the real one. Aces have 85.25% equity against a random hand, and that is the figure the verdict on this page compares against. A real shover has a range rather than a random hand, and a range moves the equity in either direction: aces gain against the weak aces and offsuit junk that a wide range contains, and lose ground against the other big pairs that a tight range is full of. What does not move is the 89.00%, because the stacks and the payout structure set that requirement no matter what the shover holds. Treat the flagship fold as the textbook version of the spot, and work out the actual matchup before you carry it to a live table.

Real satellites are messier. Some award unequal prizes, some pay cash to the bubble, some have a last longer or a partial refund. Any of those blunt the effect. Read the structure sheet before assuming the arithmetic on this page applies.

None of that undermines the core point. The direction and rough magnitude of ICM pressure in a satellite are real and enormous, and a player who folds too much on a satellite bubble is making a much smaller error than one who calls too much.

Satellites are the most extreme case of a pattern that shows up everywhere in tournaments: the value of a chip depends on the prizes still on the table. Once you can see it in a satellite, where the effect is a bubble factor of 8.089, you start noticing the milder version of it on every normal bubble and every pay jump.

The lesson series builds that intuition from the ground up, and part nine covers ICM and bubble play in depth. Work through it alongside this calculator and the arithmetic starts becoming something you feel at the table rather than something you look up afterwards.

Yes, in a satellite. In the flagship spot on this page, five players remain, four equal seats are awarded, you have 15,000 and a 10,000 stack shoves. You need 89.00% equity to call and aces have 85.25% against a random hand, so folding is correct by 3.75 points. Outside satellites, folding aces preflop all in is essentially never right: in a normal 50/30/20 ladder with those same stacks you need only 61.05%, and in a winner take all format only 50.00%.

Bubble factor is the ratio between what losing an all in costs you and what winning it gains you, measured in prize equity rather than chips. You do not have to be eliminated for it to bite: in the flagship spot on this page, losing leaves you with 5,000 chips and still costs 10.52 points against a gain of only 1.30, a bubble factor of 8.089. It converts directly into a break even requirement: the higher the bubble factor, the more raw hand equity you need before calling is profitable. You can compute it for normal tournaments in the bubble factor calculator.

ICM stands for Independent Chip Model. It converts your chip stack into a share of the prize pool by estimating how often you finish in each position, based on stack sizes. In a normal tournament that share is money. In a satellite, where every seat is worth the same, it becomes seat equity: your probability of walking away with a seat. The ICM calculator handles cash ladders.

Seat equity is your chance of finishing in a paying position and winning a seat, expressed as a percentage. In the flagship table, 15,000 chips is worth 98.70%, 10,000 is 96.52%, 4,000 is 80.80%, 3,000 is 71.11% and 2,000 is 52.87%. Those figures sum to 400.00%, that is 4.000 seats, matching the four seats being awarded, which is a useful check that the numbers are consistent.

Because a seat is the ceiling and you cannot win more than one. In the flagship spot, 15,000 chips is worth 98.70%. Doubling to 30,000 gets you 99.80%, and quadrupling to 60,000 gets 99.98%. Going from 15,000 to 60,000, four times the chips, buys 1.28 points in total. Busting, by contrast, costs you everything, so the risk to reward on accumulation is dreadful once you are safe.

Refuse every large pot, and open relentlessly against the players who cannot afford to call. Your seat is nearly locked, so any big pot risks far more than it can win. When you are first in, though, aggression is close to free, because the opponents nearest to locking a seat of their own have the highest calling requirements at the table. Stay out of the way of the shortest stack, who has to gamble anyway. If one more bustout ends the tournament, folding every hand is often your highest equity option.

Move first, while your opponents can still fold. The 2,000 stack in the flagship table still has 52.87% seat equity, but that decays every orbit as the blinds take chips you cannot replace. Shoving into opponents who are risking a nearly locked seat to call you is a genuinely strong position, because much of the time they simply cannot call. Do not carry the 89.00% figure over to your own shove, though: that is what the 15,000 stack needs against a 10,000 all in, and when you move in for 2,000 the effective stack is only 2,000, so your opponents are risking far less and their requirement drops a long way. Put your own stacks into the calculator above to see what each of them actually needs. A push fold chart gives you the ranges and the M ratio calculator tells you how long you have.

It depends entirely on the stacks and the number of seats, which is what the calculator above is for. In the flagship spot the answer is 89.00%, which no starting hand clears: aces against a random hand is the best any holding can do, and that is 85.25%. Change the payout structure and the same shove requires only 61.05% on a normal 50/30/20 ladder or 50.00% winner take all. Always compute it for your actual table rather than relying on a remembered number.

The effective stack is the smaller of the two stacks involved in a hand, because that is the most either player can actually win or lose. In the flagship spot you have 15,000 and the shover has 10,000, so the effective stack is 10,000. You are risking 10,000 of your chips, not all 15,000, and losing leaves you with 5,000 rather than nothing. It matters for bubble factor too: the less you have at risk in a pot, the smaller the gap between what losing costs and what winning gains.

No. It requires a specific and fairly extreme configuration: enough seats relative to players that one elimination ends the tournament, a comfortable stack for you, and short stacks behind you who are likely to bust on their own. The flagship spot on this page is the classic version of it. Change the stacks, the seat count or the payout structure and the requirement drops fast. Run your own table through the calculator above before assuming it applies.

It is exact arithmetic applied to a simplified picture of the tournament. The numbers on this page were computed by the site's own exact ICM engine and cross checked against brute force enumeration of every finish order, so there is no approximation error in the calculation itself. The approximation is in the model: it ignores blinds, position and skill, and it settles the finishing order by giving each player a chance of finishing first equal to their share of the chips, then removing that player and repeating among the rest. That slightly overstates short stack equity in some spots. Use it for direction and magnitude, not as a precise forecast.

The tool rounds for display. Bubble factor renders to two decimal places, so the exact value 8.089 appears on screen as 8.09, and required equity renders to one decimal place, so 89.00% appears as 89.0%. The underlying computation carries more precision than the display shows. This page quotes the fuller figures so the arithmetic in the walkthrough is easy to follow.