Three weeks. You played well. You reviewed your hands, made the tough folds, shoved the right spots, called when pot odds demanded it. And your graph is still red.
Not a little red. Red enough to hurt.
You open your tracker and look at the all-in EV line. It’s climbing. Your actual results line sits at the bottom, glued to the floor like a dog that refuses to get up. The gap between the two is money that was rightfully yours by mathematical right — and it simply never came.
Then that question creeps in and won’t let go. “Am I playing bad without realizing it? Or is this just bad luck?”
That doubt is poison. Because it lives on both sides. Sometimes you blame variance to avoid facing a glaring leak. Other times you tear apart your entire game looking for a mistake that doesn’t exist, when the only problem was the coin landing wrong.
Variance is the most misunderstood force in poker. Everyone talks about it. Almost nobody truly understands what it is, how long it lasts, or how brutal it gets specifically in MTTs.
This article breaks it down with real numbers, real examples, and real stack sizes. Not so you can accept variance with a zen smile. But so you stop confusing bad luck with incompetence — and vice versa. Because that confusion costs more EV than any bad beat.
What Variance in Poker Really Is
Let’s strip away the mysticism. Variance is the gap between the result you should have gotten and the result you actually got in the short run. That’s it. It’s math, not a curse.
When you shove AK against QQ pre-flop, you’re roughly a 43% underdog to the 57% favorite. Lose that flip and you didn’t play badly. You played correctly and the die landed on the wrong side. That’s variance operating in real time.
Here’s a costly confusion worth clearing up: variance isn’t luck. Luck is the emotional word you use when you take a bad beat at 3 a.m. Variance is the statistical description of how results are distributed around the expected mean. Over the long run it trends toward zero. In the short run it makes you want to throw your mouse through the wall.
And MTT is the format where it screams loudest. Why? Three reasons that stack on top of each other. The field is huge, so deep runs are rare by definition. The payout structure is top-heavy, concentrated at the very top. And to get there you need to win a string of all-ins back to back, each one a new coin flip.
In cash games you rebuy and keep playing at the same table. Your edge shows up hand by hand. In MTTs you can play a perfect six-hour session and get eliminated in a flip at the final table. Every bit of EV you accumulated evaporates on one river card.
That’s the nature of the beast. Now let’s look at why it’s so ferocious specifically here.
Why MTT Is Poker’s Most Variance-Heavy Format
Top-heavy payout structure
Look at the numbers for any large MTT. Roughly 15% of the field cashes. That sounds generous. But the real money — the kind that pays your bills and makes your ROI exist — is concentrated in the top three spots.
A min-cash gets you roughly your buy-in back. Sometimes less after rake. The money that matters is at the very top of the payout table, and getting there requires an unlikely chain of things going right.
The practical effect is harsh. You can have a genuine 20% ROI — an excellent number — and spend months in the red. Not because you got worse. Because the deep runs that sustain that ROI simply haven’t shown up yet. They will come. Just on their own timeline, not yours.
Field size and deep run frequency
The bigger the field, the rarer the deep run and the more violent the swings. That’s intuitive when you stop to think about it, and completely ignored when you’re in the middle of a downswing.
Take an example. A $22 with 2,000 entries: to reach the final table you need to outlast 1,991 players. Now a $109 with 500 entries: only 491 stand in your way. The higher buy-in has a smaller field, so final tables come more often, meaning smoother variance in terms of tournament count.
Not that the $109 is easier. The field is tougher. But the variance structure is different. Fields of 2,000+ entries produce deep run droughts that make competent players question everything. You could enter 400 large-field tournaments without a single final table and still be playing A-game.
The all-in gauntlet
Here’s the heart of the matter. To win a 1,000-person MTT you need to win, what, seven, eight, ten all-ins? And being a favorite isn’t enough. You need to be a favorite and win — several times in a row.
Each all-in is an independent variance event. You win a 55/45 flip. Great. Now you need to win another. And another. Multiply 0.55 by itself six times and see how small the probability of running the table gets. Even playing flawlessly, the math conspires against a happy ending in the vast majority of cases.
That’s why bad beats hurt so much more in MTTs than in cash games. It’s not just one lost hand. It’s the entire tower you spent six hours building collapsing all at once. The all-in that eliminates you erases all the EV you accumulated up to that point.
Add all three factors together and you have poker’s most variance-heavy format by a wide margin. Accepting that isn’t weakness. It’s honest bookkeeping.
How to Measure and Understand Your Downswing
MTT downswings are measured in buy-ins, not days
The first mistake every player makes during a downswing: counting time. “I haven’t been profitable in three weeks.” Wrong. Time tells you nothing about variance. What matters is how many buy-ins worth of sample you have and how deep the hole is in buy-ins.
And here comes the number nobody wants to hear: a downswing of 200 to 300 buy-ins in MTT is normal. Mathematically expected. It’s not a sign you’re broken. It’s just Tuesday.
The higher your ROI, the shallower and shorter your downswings tend to be. A player with a high ROI in a mid-stakes field will still see droughts of 150+ buy-ins with uncomfortable regularity. A player grinding large, top-heavy fields can face 400 buy-in downswings with zero technical errors. Knowing that in advance makes the hole a lot less terrifying.
The green line doesn’t lie (but it does in the short run)
Your adjusted all-in EV line is the most honest instrument you have. It shows how you would have performed if every all-in had landed exactly at its probability. It’s your “fair” result.
When the green line climbs and your actual results lag behind, the reading is clear: you played well and ran bad. The gap between the two lines is pure variance — money that’s yours but hasn’t landed yet.
Be careful, though. The green line also lies in the short run, just in the opposite direction. With a small sample, it absorbs your own bad all-in decisions. If you’re shoving the wrong ranges but running hot on them, the green line looks great while lying to your face. It only becomes truthful with volume. Read the graph as a long-run trend, never as a verdict on a single session.
Sample size: the number nobody wants to hear
How many tournaments does it take to know if you’re truly profitable? Not hundreds. Thousands. Several thousand.
Two hundred tournaments tells you nothing. Five hundred doesn’t either. It’s uncomfortable, but ROI in MTT only stabilizes after a sample that most players underestimate by a factor of ten. That positive ROI you have over 300 tournaments? Could be variance disguised as skill. The negative ROI? Same thing, in reverse.
That’s why bankroll management isn’t optional in MTT. You need enough of a cushion to survive the time it takes for your sample to reveal who you actually are.
Variance vs. Leak: How to Tell the Difference
This is the question that eats at every player in a downswing. The same one from the intro. Am I running bad or playing bad?
The lazy answer is to look at results. Lost? Must have played badly. Won? Played well. That reasoning is poison, because results and decision quality are separate things in the short run. The only honest way to answer is to review decisions, not outcomes.
Pull up the big hands from the downswing and ask of each one: was this decision +EV at the moment I made it, given the information I had? Not “did it work out.” Was it correct? If ten decisions were solid and you lost all ten, that’s variance screaming. If half were questionable, you found your leak.
This is where Tommy Angelo’s Reciprocality works as a lens. The idea is simple and sharp: your profit comes from the difference between what you do and what your opponent does in the same spot. If you both play identically in a situation, it’s neutral in the long run. Your edge lives in the spots where you decide better than they do. Reviewing hands through that lens forces you to evaluate the decision in isolation from the card that fell.
And here’s the take that’s going to sting. Most players get it wrong on both sides simultaneously. They blame variance when there’s an obvious leak staring them in the face — to avoid admitting they need to study. And they blame their own play when it’s pure variance, dismantling a strategy that was actually right. The same player does both things in different weeks. Confusing the two is the most expensive hidden cost in poker, more expensive than any individual bad beat.
The Mental Cost of Variance
Here’s the truth the numbers don’t show. Variance rarely breaks your bankroll directly. It breaks your mind first, and a broken mind is what breaks the bankroll afterward.
The mechanism is ugly. You run bad for two weeks. Frustration builds. Downswing tilt sets in — that state where each bad beat lands harder than the last because it hits an already depleted emotional balance. Then desperation tilt, when you start forcing plays to “recover,” registering too much volume, moving up in stakes to accelerate the comeback, making decisions your A-game would never make.
And this is exactly where variance transforms into a genuine leak. The card drought wasn’t your fault. But the marginal call you make in hour three because you’re exhausted and angry — that one is. The frustration 3-bet against the reg who eliminated you yesterday — that one too. Variance pushed you into a mental state where you started playing worse, and now the downswing that was purely bad luck picks up a real error component.
Understanding the types of tilt and what triggers each one is what separates players who cross a downswing intact from those who come out the other side with their game in pieces. Variance you can’t control. Your response to it, you can.
How to Survive Variance in Practice
Bankroll as a shield
No motivational speech replaces having a cushion. Bankroll is the shield that lets you run bad without going broke. In MTT, the math is different from cash: you need significantly more buy-ins given the format’s brutal variance.
One hundred buy-ins is the floor for smaller fields and softer structures. Large top-heavy fields call for 200 or more. That’s not nit-level conservatism. That’s the number that keeps you at the table through the 300 buy-in downswing that the math guarantees is coming.
Focus on process, not results
Change the success metric for your session. Stop asking “how much did I make” and start asking “how well did I decide.” A session where you made good decisions and got eliminated in three flips is a winning session in the ways that matter.
That’s not self-help fluff. It’s the only metric you actually control. Results belong to variance in the short run. Decision quality belongs to you. Measuring what you control keeps your head straight when the graph refuses to cooperate.
Volume dilutes variance
Only the long run reveals your real edge. Every tournament you play well pushes the green line and the actual results line a little closer together. Not because luck “evens out” — it doesn’t even out anything. But because the sample grows and the percentage variance shrinks around the mean.
Volume isn’t about grinding until you break. It’s about understanding that your edge only shows up with enough hands played in the right mental state. A player who logs volume on tilt just accelerates the disaster. Volume with a sustainable grind is what turns theoretical edge into actual profit.
Variance is the entry fee to a game where skill pays. If there were no variance, the worst players would never win and would stop playing. It’s what keeps the fish at the table thinking they have a shot. You don’t want to eliminate variance. You want to survive it long enough for your edge to collect what’s owed.
Want to see your own patterns? Poker Playbook analyzes your grind with AI and shows where you’re leaking EV without realizing it. Try it free at pokerplaybook.pro