Friday night, session over. You close the tables, open your tracker, and do what you always do: filter your biggest losing pots. There it is. The set that got runner-runner flushed in a $22. The AK you shoved against the widest range in existence and he happened to have exactly the two aces. You replay the hand three times, shake your head, curse the RNG, and close the software thinking you just studied.
You didn’t study. You just stewed.
That’s the trap that keeps most $11–$109 players stuck. We confuse pain with learning. The hand that hurt the most becomes the one we analyze the most, and the selection filter turns emotional instead of technical. But the pot you lost in a cooler has nothing to teach you. KK vs. AA all-in preflop is the game working as intended. There’s no leak there. Just variance.
The problem is that while you’re burning energy reviewing coolers, your real leaks slip right by. That BB call you make on autopilot. The 3bet you never make out of position. The bubble spot where you folded too tight and left EV on the table without even noticing. Those don’t hurt. That’s exactly why you never look at them.
Reviewing hands properly isn’t about revisiting a loss. It’s about separating what you control from what the deck controls — and attacking only the first.
The Mistake That Kills 90% of Your Analysis
Here’s the core issue: result is not decision. They’re two separate things your brain insists on treating as one.
You called a river with a bluff catcher, the guy showed a bluff, you won. Great feeling. You file that away as a “good call” and move on. But the result told you nothing about the quality of the decision. Maybe he only bluffs that spot 20% of the time and your call was -EV in the long run. You won despite the decision, not because of it. That’s outcome bias in its purest form — judging the line by what happened, not by what was correct.
Flip it around and it’s obvious. That hand you won that you never should have played in the first place? That turn call you made and a river miracle bailed you out? That’s the hand that needs review. It hid a leak under a good result. And you’ll never look at it, because it didn’t hurt. The other one hurt — the one you played perfectly and lost anyway.
That’s exactly why filtering by biggest losing pots is a statistical trap. You’re selecting your study sample based on emotion, and emotion targets results. The KK that lost to AA all-in preflop in a $22 has exactly zero things to teach you. The line was correct, the money went in good, the deck did the rest. Studying that is wasted time dressed up as dedication.
The player’s mind wants closure. It wants to understand why it lost. But poker doesn’t give you closure on a single hand — it gives you closure over a sample. If you want to study for real, you need to separate what you felt from what you did. Pain is a terrible filter for technical relevance.
There’s deeper work here that goes beyond any single session — how you structure your study in general. If that interests you, we broke down the method in how to study poker efficiently in 2026.
The Right Filter: By Decision, Not by Pot
So what filter actually works? Recurring spot, not isolated big pot.
Instead of “biggest pots lost,” filter by situation. 3bet pots out of position. Big blind defense against a late-position open. Bubble spots with 15–25bb effective. These are the places where you make the same decision dozens of times a week — and where a small mistake compounds into a real hole in your winrate.
The math is simple and brutal. Frequency beats magnitude. The $40 cooler happens once and you can’t control it. But that marginal BB call for $0.50 that you make on autopilot, repeated 200 times over a month? That’s $100 leaking through a crack you don’t even register as a crack. Magnitude is scary. Frequency is what kills you.
When you filter by decision, the tracker stops being a wall of misery and becomes a leak map. You start seeing patterns: “wow, I’m defending BB with K7o offsuit against a UTG raise every single time.” That’s actionable. That changes tomorrow.
How many hands before a pattern is real?
One important caveat before you start drawing conclusions from three hands. A spot needs volume to mean anything. Five BB defenses tell you nothing — that could be pure variance in your sample. You want at least a few dozen occurrences of the same spot type before you can say “this is a leak of mine.”
Otherwise you fall into the opposite of outcome bias: overadjustment. You see a ghost pattern in noise and rewrite your entire strategy because of six hands. Volume first, conclusions second.
Separate Technical from Emotional
Every hand you played badly has one of two roots. Either it was a strategy error, or it was a mental game leak. And the fix is completely different for each.
The question that separates the two is straightforward: did you not know the right line, or did you know it and not execute?
If you didn’t know — you weren’t clear on his range, you weren’t sure if a cbet was standard there, you sized wrong out of ignorance — that’s a technical error. Fix it with study. Solver work, range analysis, theory review, talking through spots with better players. It’s a knowledge gap and you fill it with knowledge.
But if you knew and did something different anyway, the problem isn’t in your game. It’s in your head. The classic example: that river call you made on tilt, after taking two bad beats in a row, knowing perfectly well that his range was loaded with value there. That’s not a range leak. It’s a mental leak. You can study theory eight hours a day and it won’t fix it — because the knowledge was already there. What failed was execution under emotional pressure.
Mixing up these two categories is what keeps competent players stagnant. They make a tilted decision, classify it as a technical error, and go study solvers to fix a problem solvers can’t touch. Studying the wrong thing with the wrong tool.
A useful framework here is the Mental Hand History, which Jared Tendler developed as a structured process for mapping the emotional layer beneath a bad decision. The idea is to treat mental leaks with the same rigor you apply to technical ones — write down what you felt, identify the distorted logic behind it, and define the correction. If you want to understand the full structure, it’s in our summary of Tendler’s work.
And when you notice that your error patterns cluster around decisions made from the bottom of the pit — those lines you only take when you’re tired or frustrated — it’s worth understanding the concept of C-game and how to keep it away from the table.
The 4-Step Process
Alright. You’ve filtered by decision, separated technical from mental. Now how do you actually analyze a hand without sliding back into results-thinking? Four steps.
Step 1 — Reconstruct context before judging
Before deciding whether the line was good or bad, you need to know the situation it happened in. Effective stack. Your position and your opponent’s. Tournament stage and ICM pressure. What reads you had on that villain up to that point.
Without that, you’re judging in a vacuum. A 12bb shove on the bubble under heavy ICM pressure is a completely different decision from the same 12bb shove early in the tournament with zero ICM. Skipping this step is the most common mistake — the player looks at the hand as if it happened in neutral space, forgetting that context defines half the answer.
Step 2 — Define the range on each street
This isn’t about memorizing solver outputs. It’s about reasoning. On each street, ask: what hands does my line represent, and what does his range look like?
On the flop, what part of his range continues against your cbet? On the turn, what gets there? On the river, against your bet, what calls and what folds? You don’t need exact numbers. You need the rough distribution — value-heavy, draws, air. That’s what turns “I thought he had” into “his range on this line is mostly X.”
Step 3 — Ask why you took the line, not whether it worked
This is the step where most players drift back into outcome bias without realizing it. The wrong question is “did it work out?” The right question is: was this decision +EV against the range I assigned in step 2?
If the river call was profitable against his distribution, it was a good call — even if he showed the nuts this time. If it was -EV, it was a bad call — even if you won. The result of the specific hand is irrelevant to that evaluation. You’re judging the process, not the runout.
Step 4 — Tag the pattern, not the hand
Last step, and the one that actually drives improvement. Use a tagging system in your tracker. “Marginal BB defense.” “3bet OOP without a plan.” “Bubble shove too tight.” Every time a hand fits a pattern, it gets the tag.
One tagged hand is an anecdote. Ten hands with the same tag is a confirmed leak, with volume, ready to attack. You stop studying individual hands and start studying tendencies in your own game.
Applied example, running all four steps at once. 4bet pot, $55 buy-in, 30bb effective. You 4bet from the BTN with AKs, villain called on the SB. Context: mid-tournament, no meaningful ICM, villain is a solid reg who called your 4bet instead of shoving — valuable information. His range: calling a 4bet out of position with 30bb as a reg usually means QQ–TT, AQs, sometimes KK slowplay; he’s shoving AA/KK premium at that stack depth, so the call skews toward medium pairs and AQ. Flop K72 rainbow. You bet, he called. Why: against the range you assigned, value betting top pair top kicker is mandatory, and his call makes sense with QQ–JJ and AQ that whiffed the flop. Pattern: if you notice you’re consistently losing your way on the turn in this type of board after 4betting, the tag is “4bet pot turn planning.” Not the hand. The pattern.
This kind of structured review is the direct application of Deliberate Practice — study with a target, feedback, and correction, not passive repetition. We unpacked that in Deliberate Practice applied to poker.
When to Use a Solver — and When Not To
Opinion that’ll annoy some people: using a solver at $11–$55 is usually wasted study time. Not saying solvers are useless. Saying that memorizing exact solver frequencies at these buy-ins solves a problem that doesn’t exist at your table.
The reason is straightforward. A solver gives you the GTO answer against an opponent who also plays GTO. Your opponents at $22 don’t play GTO — not even close. They over-fold some spots, over-call others, and hold ranges no solver would ever assume. Memorizing that you should cbet at 67% frequency on a specific board type is useless when the guy in front of you folds 90% of the time there — you should be cbetting 100% and printing, not following theoretical balance.
The correct use of solvers is conceptual. Ask “why does the solver cbet less on connected boards and more on dry boards?” and understand the principle behind it. Then take that principle to the table and adjust against the real villain. Concepts travel across situations. Memorized frequencies don’t — they die the moment your opponent deviates from GTO, which at your buy-in is constantly.
Study the why. Forget the percentage.
Common Review Mistakes
The stumbles that come up most often, in a short list to check yourself against:
- Reviewing only losses. Already covered, but worth repeating because it’s the hardest habit to drop. Winning hands hide leaks just as large as losing ones.
- Analyzing with fresh tilt. Opening your tracker ten minutes after busting is a recipe for pulling on emotion instead of technique. Let it cool. Review the next day.
- Judging without reconstructing stacks. Evaluating a decision without knowing the effective stack and tournament stage is guessing in the dark.
- Copying solver output without understanding it. “The solver does this” is not analysis. If you don’t know why, you learned nothing — you just memorized a move that doesn’t transfer.
Conclusion
There’s a side effect to reviewing hands properly that nobody tells you about. It changes how you feel when you lose.
When your review filter stops being pain and becomes pattern, a loss loses its power to consume you. You take a brutal bad beat and instead of stewing, the automatic question becomes “was the decision +EV?” If yes, there’s nothing to carry. Next hand. The cooler stops being a personal offense and becomes what it always was: variance doing its job.
Proper analysis isn’t a post-session punishment. It’s a process habit. You stop looking for someone to blame — the RNG, the fish who called, plain bad luck — and start hunting for tendencies. Tendencies you can fix. Blame, you can’t.
The difference between the player stuck at $22 for two years and the one moving up to $109 is rarely raw talent. It’s what each one does with their tracker after the tables close. One stews. The other maps.
Want to see your own patterns? Poker Playbook analyzes your grind with AI and shows you exactly where you’re leaking EV without realizing it. Try it free at pokerplaybook.pro