How to Run a Hand-History Review That Actually Fixes Leaks

Most hand reviews are entertainment disguised as coaching. A student brings their most interesting hands, you talk through them, everyone nods, the hour ends, and nothing changes, because the hands that are fun to discuss are almost never the hands that are costing the student money. A review that actually fixes leaks looks different, and it's a skill you can run like a system.

Here's the system.

The core mistake: reviewing the wrong hands

When you ask a student to "bring some hands," they bring the dramatic ones: the big coolers, the hero calls, the spots where something memorable happened. These are the worst possible teaching material, for two reasons. First, big memorable pots are usually high-variance spots where the line barely mattered. Second, they're the hands the student has already thought about the most, so there's the least new ground to cover.

The leaks that drain a student's win rate are boring and frequent: the small preflop misclicks, the auto-c-bets, the river overfolds, the spots they play on autopilot a hundred times a session without ever questioning. Those are invisible to the student precisely because they're routine. Your job is to find the routine, not relive the drama.

Start with the data, not the hands

Before you look at a single hand, look at the numbers. A student's tracker tells you where the money is leaking before they say a word, and it points you at the hands worth reviewing instead of the hands they want to show you.

You're looking for the stats that are out of line for their game and their stakes: the tendencies that suggest a systematic leak rather than a one-off. A number that's far from where it should be is a thread to pull, and pulling it leads you to a category of hands the student plays wrong repeatedly. That category is your review. You're not reviewing hands; you're reviewing a pattern, using specific hands as evidence.

This is the difference between "let's look at this cooler" and "your tracker says you're folding too much on rivers in single-raised pots. Let's look at ten of those and figure out why."

Find the pattern, not the one-off

A single hand played wrong is a mistake. The same hand played wrong ten times is a leak, and leaks are what you fix, because fixing one changes a hundred future spots.

So when you find a candidate leak in the data, pull multiple examples of it. Show the student the same situation recurring across different hands. The power isn't in any one hand. It's in the student seeing "oh, I do this every time." That recognition is what creates change. One hand is arguable; the pattern is undeniable.

This is also why database review done on your own time (where the student sends you their hands and you go through them whenever, instead of on a live call) is so powerful for finding leaks: across thousands of hands you can spot patterns that never surface in a live session looking at a handful of cherry-picked spots.

Make the student do the thinking

The instinct of a new coach is to explain: to deliver the correct answer clearly and completely. But a student who is told the answer forgets it; a student who arrives at it remembers. The review should be mostly questions.

When you reach a decision point, stop and ask: what are you thinking here? Why this size? What's your read on their range? What happens if you do the other thing? Make them reconstruct their reasoning out loud. Often the act of explaining their own play to you is when they catch their own mistake, which is worth ten times more than you catching it for them.

Save your explanation for after they've worked it. Then you're correcting a thought they actually had, not depositing information into a passive listener. The hands they fix themselves, with your guidance, are the ones that stick.

Don't drown them: one or two leaks per session

A review can surface a dozen things wrong. Resist the urge to name them all. A student who leaves with a list of twelve fixes changes zero of them; a student who leaves with one or two clear, specific things to work on actually works on them.

Pick the highest-impact leak (usually the most frequent one, since frequency times severity is what actually costs money) and go deep on it. Make sure they understand not just that it's wrong but why, and what to do instead, concretely enough that they'll recognize the spot next time they're in it. One leak truly fixed per session, session after session, transforms a player. A dozen leaks mentioned and none fixed is just a nice conversation.

End with something they can actually do

A review that ends with "so play better in those spots" has failed. End with a specific, checkable assignment: a concrete adjustment to make, a spot to watch for, sometimes a small piece of off-table work: a range to drill, a few minutes in a solver on the exact node you covered, a number to watch in their tracker next week.

The assignment does two things. It turns insight into habit, which is where actual improvement lives. And it gives the next session a starting point: "last week we worked on your river folds. How did it feel this week, what did the data say?" Now your coaching is a continuous arc instead of a series of disconnected hours, and the student can feel themselves improving, which is what keeps them booking.

The session, start to finish

Put it together and a leak-fixing review looks like this:

You review the student's data before the session and arrive with a hypothesis about their biggest leak. You open by pulling several hands that show the pattern. You walk through them asking questions, making the student reconstruct and examine their own reasoning. When the pattern is undeniable to them, you explain the why and the fix concretely. You resist naming the other eleven things you noticed. You close with one specific assignment and a marker to check next week.

That's a review that changes a player. It's less fun than swapping cooler stories, and it's the difference between a coach a student keeps for a year and a coach they try once.