Poker Tournament Strategy: The Complete Guide

Every tournament has three distinct phases, and each one demands a different strategic mindset.

Phase 1: Deep-Stack (100bb+, levels 1-6)

Early levels look like cash games. You have 100-300 big blinds, the blinds are tiny relative to your stack, and there is no immediate ICM pressure. The temptation is to play loose and "build a stack" — this is the most expensive mistake in tournament poker. Most early-stage chips you accumulate are converted at chip-EV, not money-EV. Risking your tournament life with marginal hands costs you tournament equity even when you win the pot.

What works:

• Tight-aggressive opens. Open about 17-22% of hands from late position, drop to 10-13% from early position. Open 2.0-2.5x the big blind.
• Three-bet polarized: premium hands (QQ+, AK) and well-chosen bluffs (suited connectors, suited wheel aces).
• Avoid bloating pots out of position with marginal holdings. AJo from UTG against a late-position 3-bet is a fold, not a 4-bet.
• Set mining is profitable when implied odds are massive and stacks are 200bb+.

What doesn't:

• Hero calls with one pair on rivers in 4-bet pots. Tournament players underbluff the river.
• "Speculative" calls of 3-bets out of position with hands like 89s or 76s. Postflop variance OOP destroys these hands' EV.

Phase 2: Mid-Stack (40-80bb, levels 7-15)

This is where most tournaments are won and lost. The blinds are big enough to matter, antes appear, the field is thinning, and you need to start stealing aggressively. Stack depth dictates your strategy — your 60bb stack plays differently than your opponent's 30bb stack across the table.

Open ranges by position (40-50bb):

• UTG: 12-14% (88+, AJ+, KQs, suited broadways)
• Hijack/Cutoff: 22-26% (66+, A9+, suited connectors down to 65s, broadways)
• Button: 35-45% (any pair, any suited ace, suited connectors, any suited broadway, KTo+)
• Small blind: 35-40% (limp/raise mix, drop trash)

Three-bet ranges expand:

• Versus button open: 3-bet QQ+, AK for value, Ax suited wheel and broadways as bluffs. ~9-11% total.
• Versus EP open: tighter — JJ+, AKs for value, mostly call AQ/AK and pairs.

Resteal frequencies matter more than ranges. When the cutoff opens for 2.2bb and you have 40bb in the small blind, a 9bb 3-bet jam needs only ~22% fold equity to be profitable with hands like A5s, KJs, 99. The math isn't the hand strength — it's the fold equity at this stack depth.

Phase 3: Short-Stack (20bb and below, late stages)

Below 20bb, postflop play essentially disappears. Almost every chip moves preflop. This is the push/fold and shove/restealing zone, where solver-derived ranges dictate every decision. Check any close shove or call against the likely range in the equity calculator, and against the payout structure in the ICM calculator.

Approximate push ranges from the small blind versus a folded big blind:

• 20bb: 35-40% of hands
• 15bb: 50-55% of hands
• 10bb: 65-70% of hands
• 8bb: 70-78% of hands
• 5bb: 90-95% of hands (any two pretty much)

But these ranges shift dramatically when ICM pressure is on. The same 15bb stack at 20% ITM (in the money) plays radically different from a 15bb stack three players from the bubble.

The Independent Chip Model (ICM) is the math that converts your chip stack into a real-money equity percentage of the remaining prize pool. It is the reason a 20bb stack on the bubble plays nothing like a 20bb stack in a winner-take-all sit-and-go.

The intuition: if a 9-handed final table pays 1st = $10,000, 2nd = $7,000, 3rd = $5,000, ... 9th = $500, then your 100,000 chips have a different dollar value depending on the chip distribution of every other player. Doubling your stack doesn't double your money — it adds maybe 60-70% of your current equity, because the higher prizes are still distant.

This non-linearity creates risk premium: in spots where you'd be a slight chip-EV winner, you might be a money-EV loser. The classic case is the bubble — folding your way into the money pays better than playing pots that are coin-flips at chip-EV.

Concrete example. Final table of a $100 buy-in MTT, 9 players, $20,000 prize pool with the top-heavy structure above. You hold AKo on the button with 30bb. The cutoff (35bb) min-raises, the small blind (8bb) jams. The big blind has 12bb and folds.

• Pot odds: you need to call 28bb to win ~46bb. ~38% pot equity required at chip-EV.
• AKo vs the cutoff's calling range and the 8bb's all-in range: ~46% equity at chip-EV. Call.
• ICM-adjusted: the cutoff's range tightens (he doesn't want to bust into 9th), and your stack falling to 2bb is catastrophically bad — you'd lose ~$1,600 of money equity if you bust, vs gaining ~$1,200 if you win. ICM-adjusted equity required: ~52%. AKo becomes a fold.

This is the kind of decision the ICM calculator makes routine. Plug in stacks, plug in payouts, see exactly how much equity you need to call profitably.

Read the full ICM guide →

The bubble — the few hands before the money — is where the largest equity transfers happen. Big stacks bully medium stacks. Medium stacks abuse short stacks. Short stacks fold their way to a min-cash.

If you are the big stack:

• Open 50%+ of hands from late position. Most opponents fold too tight.
• 3-bet medium stacks aggressively — they cannot risk their tournament life, so they fold 70%+ of opens.
• Avoid risking your stack against another big stack. Two big stacks colliding gives all the value to the medium stacks.

If you are a medium stack:

• Tighten up. Your fold equity is intact — use it for resteals against late-position openers, not for marginal opens of your own.
• Avoid coin flips against big stacks. Look for spots where your fold equity does the work.
• Steal from short stacks who can't 3-bet, and avoid the big-stack bullies.

If you are a short stack:

• The "minimum-cash" mindset is wrong unless the next pay jump is huge. A min-cash is often only 1.5x your current ICM equity — not worth folding KK to preserve.
• Look for spots where you have 50%+ fold equity (3-bet jams over loose openers, restealing the small blind from the big blind).
• Avoid the medium stacks who will call you wide. Pick on big stacks who have nothing to gain by calling.

Specific number: at the bubble of a typical MTT (100/200 blinds, 25 ante, 12bb stack, you're in the small blind, big blind has 18bb), your push range is roughly 15-20% — mostly any pair, any A, suited Kx and Qx, suited broadways. The big blind's call range against you is about 7-10% — pairs 66+, AJ+. You have 90%+ fold equity. The math works.

Once you reach the final table, the equity-per-decision goes through the roof. A typical 9-handed MTT final table has pay jumps like:

• 9th: $1,000
• 7th: $1,800
• 5th: $3,200
• 3rd: $6,500
• 1st: $15,000

Each elimination adds thousands to your money equity. ICM pressure intensifies as the pay jumps grow steeper toward the top. The strategic implications:

Don't be the next out. Even hands like JJ are folds in some 8-handed spots when calling an early-position jam from a covering stack with two short stacks in danger. The dollar cost of busting in 8th is sometimes higher than the dollar gain of doubling.

Steal relentlessly when ICM-protected. If you're chip leader with two short stacks in the blinds, you can open ATC (any two cards) from the cutoff or button. Their resteal ranges are ICM-tight, your folding losses are minimal, and you bleed chips off them every orbit.

Deals matter. Most final tables get to a deal. The chip leader wants chip-chop; the short stacks want ICM. The right deal for you depends on your skill edge and risk tolerance. Use the ICM calculator to know your exact equity number before negotiating.

Read the full ICM and final table deal guide →

Push/fold is the GTO solution to the simplest tournament situation: you have 15bb or less, the action folds to you, and you must decide to shove or fold. The Nash equilibrium has been solved exactly for this game tree.

Approximate Nash push ranges from the small blind, heads-up vs big blind:

These are simplified ranges. Real solver play accounts for ICM, antes, and the big blind's call range. But the table above is correct within ~5% for most chip-EV situations.

The most common push/fold mistake is folding too tight at 8-12bb. A 10bb stack folding ATo from the small blind is leaving 0.4bb/orbit on the table — over 50 orbits of a deep tournament, that's a significant chunk of equity.

Read the GTO poker strategy guide →

Tournament variance is brutal. The standard deviation of MTT results is several hundred buy-ins per session. Even a winning tournament player can run 200+ buy-ins below expectation.

Minimum bankroll guidelines:

• Recreational/semi-serious: 100 buy-ins for the level you regularly play.
• Serious grinder: 200 buy-ins, with strict moving-down rules at 50% bankroll loss.
• Satellite/qualifier-heavy schedule: 150 buy-ins (variance is similar to MTTs).

A $50 buy-in MTT regular needs $5,000-$10,000 of poker bankroll. A $215 PLO MTT regular needs $20,000+. Going below these numbers means accepting a non-trivial probability of going broke even with positive long-run EV.

Track your bankroll with the TryBluff bankroll tracker — it logs MTT and cash sessions separately, calculates ROI, and shows your variance distribution over time.

Read the full bankroll guide →

Sit-and-gos (SNGs) and multi-table tournaments (MTTs) share the same chip-stack-to-money mechanics, but the strategic emphasis is different.

MTT specifics:

• Long structures (15+ minute levels) reward postflop play in early stages.
• Pay structures are top-heavy — the top 1-2% of the field win 30%+ of the prize pool. Playing for top 3 matters more than min-cashing.
• Fold equity persists through more levels because there are more opponents to fold.

SNG specifics (especially 9-max single-table):

• Short structures (8-10 minute levels). You hit the push/fold zone within 30-40 minutes.
• Pay structures are flatter — typically 50/30/20 for top 3. This makes ICM pressure intense from very early.
• Less postflop play; more mathematical preflop decisions. Solver Nash ranges are nearly all you need.

The strategic skill in MTTs is in the deep- and mid-stack play; SNG skill is in the precision of your push/fold ranges and your reads on opponents' calling tendencies.

Satellites are tournaments where the prize is a seat into a bigger event, and they require a strategic adjustment that breaks ICM intuition.

The key insight: in a flat satellite (e.g., top 10 of 100 win seats), every winning seat is identical. Once you have enough chips to lock the seat, the value of additional chips is zero. Playing tight as a chip leader on the bubble of a satellite is mathematically correct in ways that would be wrong in a regular MTT.

Satellite-specific adjustments:

• Big stacks fold even premium hands when they cover all bubble stacks — preserving the seat is worth more than any chip-EV gain.
• Medium stacks tighten up dramatically; only call all-ins with hands that are crushing.
• Short stacks should jam wider than usual — your fold equity is enormous because nobody wants to bubble.

A short-stack reshove range of 25%+ is correct in many satellite spots that would be tight in a regular MTT. The ICM math works in reverse — calling has terrible value because eliminating a player gains you almost nothing, but shoving steals huge fold equity.

1. Limping in late position — limping forfeits fold equity. Open or fold.
2. Calling 3-bets out of position with marginal hands — postflop variance OOP is the silent killer.
3. Set-mining when shallow — set mining requires 15-20:1 implied odds, which means 200bb+ stacks. At 50bb you cannot set-mine 22.
4. Stack-off mistakes on the bubble — even small chip-EV mistakes have huge dollar costs near the money.
5. Calling ICM-tight in big-pay-jump spots — the cost of busting is non-linear, and most players overestimate their fold equity.
6. Failing to adjust to ICM at the final table — most amateurs play final tables at chip-EV and leave thousands of dollars on the table.
7. Insufficient bankroll — playing $215 MTTs with a $5,000 bankroll is gambling, not professional poker.
8. Tilting from variance — tournament variance produces inevitable 50-200 buy-in downswings even for winners.

What is the difference between cash game and tournament strategy?

In cash games, every chip is worth its face value in dollars, and your bankroll is independent of your current stack. In tournaments, chips have non-linear money value (ICM), and the prize pool's pay structure means you sometimes correctly fold +chip-EV hands for negative money EV. Tournaments also feature increasing blinds that force action; cash games stay at fixed stakes.

How do you play poker tournaments to win?

Master the three phases — deep stack tight-aggressive, mid stack steal-and-resteal, short stack solver push/fold. Apply ICM ruthlessly on the bubble and at final tables. Manage your bankroll with at least 100 buy-ins. Use tools like our ICM calculator for deal evaluation and the equity calculator to check your shoves against realistic calling ranges.

What stack depth changes my strategy the most?

The biggest strategic shifts happen at 40bb (where you transition from postflop play to push/fold/reshove zones), at 20bb (where you stop opening for raises and start jamming), and at 8bb (where push/fold becomes 90%+ of decisions). Below 8bb you are fundamentally just playing solver Nash equilibrium.

How important is ICM in tournaments?

ICM is the single most important concept in tournament poker after basic preflop ranges. Once you reach the money bubble or final table, ignoring ICM is the difference between break-even and winning: playing chip-EV at money-EV decision points systematically forfeits real-money equity, because the dollar cost of busting near a pay jump is non-linear in chips.

What is the best bankroll for tournament poker?

For serious play, 200 buy-ins of the level you regularly enter, with strict moving-down rules at 50% bankroll loss. Recreational players can get away with 100 buy-ins. Tournament variance is severe — 100+ buy-in downswings happen to winners regularly.

How do I improve my tournament poker skills?

Check push/fold spots against calling ranges with the equity calculator, study ICM with the ICM calculator, review hand histories with the AI coach, and play volume to learn population tendencies. Read solver outputs for spots that confuse you. Track your results with the bankroll tracker to identify which buy-in levels and game types are most profitable for you.

• ICM Calculator — Plug in stacks and payouts, see exact equity for any tournament spot.
• Equity Calculator — Check shoves and calls against realistic calling ranges.
• AI Poker Coach — Submit hand histories, get detailed strategic feedback.
• Hand Analyzer — Replay hands, identify mistakes, share with players.
• Bankroll Tracker — Log MTT results, monitor ROI, manage variance.
• Tournament Finder — Browse live tournaments by location, buy-in, and structure.

Related guides:

• Master poker cash game strategy →
• Get personalized poker coaching →
• Read the full poker strategy hub →