Finding a PLO leak requires a repeatable sample of decisions and a useful way to rank errors. A memorable losing hand is not automatically a leak, and a low accuracy score is not automatically the most expensive problem.
Define a leak as a repeated decision pattern
Examples include opening too loose from one position, defending the big blind with dominated structures or 3-betting a category too aggressively. The pattern matters more than one isolated result.
Collect decisions in a controlled spot
Use drills that hold position and action sequence constant. This reduces noise and makes structural hand differences easier to see.
Measure EV loss
Count how much value each error surrenders compared with the best available action. Group the losses by spot and category.
Separate broad and category leaks
A broad leak affects the overall strategy, such as raising too tight from the button. A category leak affects a structural group, such as overcalling unpaired double-suited hands in one defence spot.
Confirm the cause
Inspect the full range and compare exact combinations around the boundary. Look for the card feature you are overvaluing or missing.
Drill, then retest
Train the category in isolation, return to the complete spot and later retest in a mixed session. Improvement must transfer beyond the narrow drill.
| Signal | Possible interpretation | Next step |
|---|---|---|
| Low accuracy, low EV loss | Many close mix errors | Study the pattern without over-prioritising it |
| High accuracy, high EV loss | Few but severe mistakes | Review the exact pure-action failures |
| One weak position | Range construction issue | Run a position-specific drill |
| One costly category | Structural misunderstanding | Compare boundary combinations and drill the category |
| Errors across all spots | Foundational concept gap | Return to starting-hand structure and position |
Use played hands as supporting evidence
A Session Journal can reveal situations that feel repeatedly difficult. Use those hands to generate study questions, but validate the strategy through range, EV and equity analysis rather than results alone.
Frequently asked questions
How many mistakes make a leak?
There is no universal count. Look for a repeated pattern with enough decisions to distinguish it from random variation, and consider the total EV surrendered.
Should I fix the lowest accuracy score first?
Not automatically. Prioritise the combination of frequency, EV loss and strategic importance.
What is a category leak?
It is a repeated error involving a structural hand class, such as certain single-suited connected hands in one position.
How do I know a leak is fixed?
Retest the complete spot after focused practice and confirm that the correct process transfers to new combinations, not only the memorised drill.