Find the Decisions Behind
Your HM3 Leaks
Leak Explorer is a useful statistical starting point. The next question is which hands created the number, and whether the deviation is worth fixing.
Leak Explorer surfaces and drills into leak signals. The review decides which decision families deserve action.
REQUEST EARLY ACCESS · ALPHA IN PREPARATION
REQUEST EARLY ACCESS · ALPHA IN PREPARATION
POST-SESSION ONLY · NO REAL-TIME ASSISTANCE
What HM3 Leak Explorer already does well
Leak Explorer is already a strong leak-analysis starting point. It can compare stats to winning players, show leak severity, weigh opportunity counts, and drill into positions, hole cards, and hands.
LEAK EXPLORER ALREADY HELPS WITH
- Statistical comparison
- Leak severity
- Opportunity-aware analysis
- Hand and position drill-down
NEXT REVIEW ANSWERS
- Is there a repeating decision family?
- How confident is the finding?
- Is the pattern material enough for this study cycle?
- Which issue should come first?
- What should I train?
- How will I re-measure it?
Keep Leak Explorer as the statistical starting point. The review layer adds evidence, priority, and a training loop.
Why benchmark deviations are only the beginning
A benchmark signal can start an investigation. A deviation is not automatically a decision leak.
NOT YET A PPC FINDING
- Defined decision family
- Finding-level confidence
- Study-cycle materiality
- Review-level priority
- Training target
Leak Explorer already ranks and drills into leak signals. Those objects are not yet a finding with explicit confidence and a study-cycle priority.
Which hands create the statistic?
HM3 can already drill into the hands behind a leak signal. The PPC review uses that hand-level evidence differently: define the decision family, count the relevant opportunities, inspect representative decisions, then decide whether the pattern deserves a finding.
- LEAK EXPLORER SIGNAL
- DEFINE THE SPOT
- FILTER HISTORICAL HANDS
- COUNT OPPORTUNITIES
- INSPECT REPRESENTATIVE DECISIONS
- REPEATING PATTERN?
Opportunity counts already exist in Leak Explorer. Counting them here is part of good evidence, not a capability unique to PPC.
Does the deviation actually cost meaningful EV?
A repeated deviation can still be small, low-impact, not trainable enough, or not a current priority. Priority is not the largest percentage deviation and not the most frequent warning.
- REPETITION
- ESTIMATED IMPACT
- CONFIDENCE
- TRAINABILITY
Leak Explorer already supplies a useful severity signal. PPC does not replace that score. The review combines available signals with decision-family evidence, confidence and trainability to decide what deserves the current study cycle.
Which leak deserves attention first?
Leak Explorer can rank leak severity. PPC's priority answers a different question: what should enter this player's next study block? Two candidates can both be visible and still not deserve the same study time.
CANDIDATE A
- HIGH FREQUENCY
- LOWER MATERIALITY
CANDIDATE B
- LOWER FREQUENCY
- HIGHER ESTIMATED COST
- STRONGER EVIDENCE
This comparison is PPC's study-priority model. It is not a claim that Leak Explorer does not rank leaks, and it is not a measured HM3 result.
A signal can resolve to priority, no action, or insufficient evidence
These are outcome states of the review model — not your results.
OUTCOME MODEL
Enough evidence and enough expected impact to act.
A signal exists, but it is not worth the current study cycle.
The sample in the defined spot cannot support a strong claim yet.
Turn a leak into a training objective
Once a priority is verified, define one practice block, play new hands, then re-measure the same spot. Train the defined decision family. Training is post-session. There is no live table prompt and no HUD coach.
- VERIFIED PRIORITY
- DEFINE ONE PRACTICE BLOCK
- PLAY NEW HANDS
- RE-MEASURE SAME SPOT
Measure the same spot on future hands
Do not judge success from short-term winrate. Re-measure the same decision family on new opportunities.
- IMPROVING
- UNCHANGED
- MORE DATA NEEDED
Those labels are conceptual follow-up states. They are not percentages from an HM3 customer graph.
Leak Explorer can stay the starting point. Export the data you already have.
You do not need to move or replace your HM3 database.
- 01
Open Holdem Manager 3.
- 02
Select the player and database you want reviewed.
- 03
Export the relevant hand-history or report data. Leak Explorer can stay as the statistical starting point.
- 04
Save the export locally.
- 05
Return here and select the file.
The current Analyze preview only checks whether the selected file type looks supported. It does not parse or analyze the file yet. The selected file stays on this device during the current preview. Full automated analysis is not live.
Full database analysis is not live yet. We're opening early access to players with real PT4, HM3 and hand-history databases.
Early users may be asked to share anonymized exports manually while the automated analyzer is being built.
Questions
Is this a Leak Explorer replacement?
No. Leak Explorer is a strong way to see statistical deviations. This layer asks whether those deviations are decision leaks, how confident the sample is, and what to train first.
What happens after the first leak list?
The intended loop is diagnose, train one priority, play, then re-measure the same spot on new hands.
See the broader HM3 workflow
From HM3 history to a short study plan — without replacing the tracker.
POKER LEAK FINDERSee the generic leak methodology
What a poker leak is, and why a strange number is not automatically one.
SAMPLE REPORTSee the current illustrative sample report
The current sample uses a separate PokerTracker-source illustrative dataset, but demonstrates the same priority / evidence / refusal-state report model.
METHODOLOGYSee how evidence is required
Confidence, sources of truth, and why no action is valid.
Don't stop at the deviation.
Inspect the decisions.
Leak Explorer already surfaces and drills into leak signals. The review decides which decision families deserve the next study block.
Post-session analysis only. No real-time assistance.