Analyze Thousands of Poker Hands —
Not Just One Spot
One interesting hand is a story. Recurring spots across a database are a study plan.
One hand is a story. A database is a pattern.
REQUEST EARLY ACCESS · ALPHA IN PREPARATION
REQUEST EARLY ACCESS · ALPHA IN PREPARATION
Post-session analysis only. No real-time assistance.
Why one-hand analysis misses recurring leaks
A single interesting hand can be worth reviewing. It cannot tell you whether the same decision keeps appearing across a session or a database.
ONE-HAND REVIEW
Useful for one situation.
- One action
- One line
- One interesting spot
- One decision context
BULK HISTORY REVIEW
Useful for recurring tendencies.
- Repeated actions
- Position-specific tendencies
- Street-specific patterns
- Recurring pot types
- Whether a pattern appears enough times to investigate
SINGLE-HAND CHAT
- Input
- One hand
- Output
- One explanation
- Best for
- One spot
BULK PERFORMANCE REVIEW
- Input
- Many hands
- Output
- Recurring patterns and priorities
- Best for
- Deciding what to study
One-hand review is not a weaker product. It answers a smaller question. Bulk history answers whether the spot keeps happening.
What we extract from hand histories
A hand history is source material. The useful object is a structured decision Hero actually faced — not a prompt pasted into a chat.
- POSITIONHero seat for the decision.
- STREETPreflop, flop, turn, or river.
- POT TYPESingle-raised, 3-bet, or other.
- DECISION FAMILYThe repeating situation being counted.
- OBSERVED ACTIONWhat Hero did when a choice existed.
- STACK CONTEXTEffective stack when the history records it.
- OPPORTUNITYA decision Hero actually faced.
FROM RAW HANDS TO STRUCTURED DECISIONS
Hand history becomes a structured decision.
The intended workflow is reconstruction, not commentary. Each step is a data transformation. This is not a live progress indicator.
- HAND HISTORY
- PARSED HAND
- HERO DECISIONS
- CONTEXT
- DECISION FAMILY
- HAND HISTORYSource text from a session, export, or history batch.
- PARSED HANDA normalized hand the product can inspect.
- HERO DECISIONSThe choices Hero actually faced.
- CONTEXTPosition, street, pot type, and stack when known.
- DECISION FAMILYA repeating family that can be counted.
Analyze decisions across entire sessions
The useful unit is not only one hand. It can be a session, a tracker export, a history batch, or a database review sample.
- A session
- A tracker export
- A hand-history batch
- A database sample
Current product scope is NLHE cash 6-max. The page describes the intended capability. The live Analyze flow is still a preview, not an automated parser.
Group hands by position, street and pot type
Once hands are decisions, they can be stacked by context. Grouping is how a database becomes inspectable instead of a pile of stories.
REPRESENTATIVE CONTEXT
One inspected context inside the family — not the slice the family counts belong to.
REPRESENTATIVE DECISION
hh-river-01
Called a large river bet after missing
FAMILY-LEVEL EVIDENCE
Counts belong to the decision family, not to a single position–street–pot cell.
River Bluff Catching
RELEVANT DECISIONS
186
of 412 opportunities
CONFIDENCE
HIGH
for this decision family
Find repeated situations
Recurrence is what makes a database useful. One representative decision is a story. A family with enough opportunities is something to investigate.
hh-river-01
BB · Single-raised pot
Called a large river bet after missing
preflop / flop / turn / river
River Bluff Catching
186 relevant decisions of 412 opportunities
High confidenceVerify representative decisions
A repeated pattern is inspected through representative decisions from the same family — not through invented cards, boards, or a replayer.
BB · Single-raised pot
Called a large river bet after missing
preflop / flop / turn / river
Similar weak bluff-catchers are being continued too frequently in this branch.
CO · 3-bet pot
Called river after a missed draw
preflop / flop / turn / river
Ace-high with no improvement after a passive turn — the same branch, a different runout.
Identify candidate leaks. Recurrence is not a verdict.
A pattern that repeats is still a candidate. It needs enough opportunities, enough confidence, and enough estimated impact before it becomes a study target.
River Bluff Catching
Repeated, evidenced, and material enough to inspect further.
UTG open frequency
Detected, then left off this cycle because the expected impact is small.
Turn probes in position
The opportunity count is too small to support a leak claim.
Recurring is not the same as wrong. A leak finder that cannot refuse a claim is just a warning generator.
Turn hands into training material
The product should compress thousands of hands into a few study targets — not write a paragraph for every hand.
Training on this site is the intended next step after a verified priority. This page is not a live training engine and does not advise current hands.
USEFUL OUTPUT
- Recurring candidate patterns
- Evidence and opportunity counts
- Representative decisions
- Confidence
- Estimated impact where supported
- Prioritized study targets
- Refusal states
- Follow-up filters
NOT THE OUTPUT
- Thousands of comments
- An AI paragraph per hand
- A solver line for every decision
FROM HISTORY TO PRIORITY
- HANDS
- DECISIONS
- GROUPS
- CANDIDATES
- EVIDENCE
- PRIORITY
- HANDS IN THE SAMPLE84,281
- FINDING OPPORTUNITIES412
- RELEVANT DECISIONS186
- RECURRING FAMILYOne decision family
River Bluff Catching
−0.47 bb/100
HIGH CONFIDENCE
Supported formats
Designed for PokerTracker 4 exports, Holdem Manager 3 exports, and raw hand histories.
- PT4PokerTracker 4
Use exports and hand histories from the database you already collect in PT4.
- HM3Holdem Manager 3
Keep HM3 as your tracker. Add a performance review on top.
- HHRaw Hand Histories
Use exported hand-history files directly.
The current Analyze preview can say a file type looks supported. That is weaker than claiming every export version is compatible, and it does not mean a live parser has run.
NOT CLAIMED
- A direct tracker database connection
- PokerStars or GG APIs
- Automatic cloud sync
POST-SESSION BOUNDARY
Post-session analysis only. No current-hand advice.
- Post-session analysis only
- No real-time assistance
- No HUD
- No advice on a hand you are playing
The current Analyze preview keeps the selected file on the device. That is preview behavior, not a permanent promise about future full analysis.
FROM THOUSANDS OF HANDS
Want to see what thousands of hands become?
The sample report shows how a large history compresses into a short priority list, plus the refusal states that keep the product honest.
DATABASE REVIEWHand-history analysis turns raw hands into structured evidence. Database review turns that evidence into a study plan.
This page is the input transformation: histories become grouped decisions. Database review is the output process: evidence becomes priorities.
LEAK FINDERHistories find recurring patterns. Leak finder asks which ones are worth acting on.
A repeated family is still a candidate. Leak finder is the later question: evidence, confidence, estimated impact, and whether to train.
Stop reviewing one interesting hand at a time.
Turn a session into patterns.
Turn the patterns into priorities.
REQUEST EARLY ACCESS · ALPHA IN PREPARATION
Post-session analysis only. No real-time assistance.