01
Too much information
Trackers measure everything. Study time is finite.
AUTOMATED DATABASE REVIEW♠♥♦♣
100,000
hands.
Three problems
worth fixing.
Turn your PokerTracker or Holdem Manager database into a prioritized review of the decisions that deserve your study time.
Post-session analysis only. No real-time assistance.
ALPHA IN PREPARATION
PRIORITY #1High confidence
River Bluff Catching
EST. IMPACT
−0.47
DECISIONS
186
- ■ sound folds102
- ■ thin calls45
- ■ expensive39
OF 412 OPPORTUNITIES · 84,281 HANDS
K♠7♦4♣2♥9♠
- POKERTRACKER 4
- HOLDEM MANAGER 3
- RAW HAND HISTORIES
- NLHE 6-MAX CASH
- NL10 — NL200
- EVIDENCE
- CONFIDENCE
- PRIORITY
WHY MOST PLAYERS STRUGGLE TO IMPROVE
02
Not every deviation matters
A red number is a signal, not a verified leak.
03
Know what to fix first
Rank a few issues by evidence, confidence, and cost.
WHAT SURVIVES THE FILTER
84,281
HANDS READ
12
CANDIDATE AREAS
3
VERIFIED PRIORITIES
WHERE THE COST SITS · BB/100POSITION × STREET
COST BY STREET
PREFLOP−.02
FLOP−.04
TURN−.07
RIVER−.24
BB on the river carries half the total cost. BB · river · −0.24 of −0.47.
EVIDENCE THRESHOLDOPPORTUNITIES IN SPOT
37
TURN PROBES IP
412
RIVER BLUFF CATCHING
<50
INSUFF.50
LOW SIGNAL150
MEDIUM400+
INVESTIGATE HANDS
INSUFF.50
LOW SIGNAL150
MEDIUM400+
INVESTIGATE HANDS
Same database, same export. One finding earns a claim, the other does not — the file size never decides.
That subtraction is the product.
50 combos carry 71% of the loss.
GOOD STATS DON'T MEAN GOOD DECISIONS
The number was fine.
The hands were not.
Two players share Fold to 3Bet = 58% and leak in opposite ways. A tracker reports one figure for the whole region. We read it cell by cell, then show the part that is actually costing money — 50 combos carry 71% of the loss.
DEFEND — SOUNDDEFEND — EXPENSIVEFOLD
SAMPLE FINDING
River Bluff Catching
EST. IMPACT
−0.47
bb/100
Calling frequency stays high versus large river bets after the hero has missed.
The sample shows a repeating tendency to continue on the river in spots where the opponent's value range is concentrated. The finding is ranked first because the estimated cost and the opportunity count are both material.
EST. IMPACT
−0.47
bb/100
DECISIONS
186
of 412 opps
EXPENSIVE
39
calls
DECISIONS
186
OPPORTUNITIES
412
DECISION COMPOSITION
- ■ sound folds 102
- ■ thin calls 45
- ■ expensive 39
REPRESENTATIVE HANDS
BB · SRPCalled a large river bet after missing−78bb
CO · 3BPCalled river after a missed draw−92bb
ACTION PLAN
Train fold/call decisions versus large river bets on missed-draw textures, then re-measure the same filter on new hands.
BB vs BTN Defence
940 of 2,104 · HIGH
3-Bet Pots OOP
268 of 615 · MEDIUM
Turn probes in positionINSUFFICIENT
EVIDENCE
EVIDENCE
THE REFUSAL LOG
Nine we
refused to claim.
Seven spots had too few opportunities to support a finding. Two were real and too cheap to be worth a study block. Published, not hidden.
Turn probes in position37INSUFFICIENT
River overbet defence24INSUFFICIENT
SB cold-call frequency41INSUFFICIENT
Flop donk bets12INSUFFICIENT
4-bet pots in position18INSUFFICIENT
Turn check-raise defence33INSUFFICIENT
River overcalls vs small bets46INSUFFICIENT
UTG open frequency1,840LOW IMPACT
Flop cbet sizing, SRP2,310LOW IMPACT
3 survived.OF 12 CANDIDATES
HOW IT WORKS
01
Upload
A PT4, HM3, or raw hand-history export. Local file, post-session.
02
Diagnose
Candidate areas detected from decisions, not from a stat grid alone.
03
Verify
Opportunity count and pattern consistency set the confidence level.
04
Prioritize
Survivors ranked by estimated cost × opportunities.
THEN FUTURE HANDS ARE RE-MEASURED
Priority #2 stays sealed until #1 verifies.
3 PRIORITIES · 4 WEEKS
TRAIN → PLAY → RE-MEASURE
WEEK 1TRAINING
River Bluff Catching
Missed-draw textures 40 spots · drilled
WEEK 2TRAINING
River Bluff Catching
Large bets only 40 spots · drilled
WEEK 3VERIFY
Play & re-measure
Same filter new hands only
WEEK 4LOCKED
BB vs BTN Defence
Unlocks when #1 verifies
BEFORE · 186 DECISIONS
EXPENSIVE CALLS 21%
6,284 NEW
HANDS
AFTER · 96 DECISIONS
EXPENSIVE CALLS 9%
−12pp
Target decision quality improved on hands the review never saw. Not a winrate claim — 6,284 hands cannot carry one.
WORKS WITH YOUR TRACKER
Keep your tracker.
Add a performance layer.
Export the database you already built. We are not rebuilding PT4 reports, and HM3 Leak Explorer stays a useful statistical starting point.
PokerTracker 4Holdem Manager 3Raw hand histories
FREE TOOLS
Poker Stats CheckerFREE →
Aggregate triage on your VPIP, PFR, 3Bet and showdown numbers. No GOOD/BAD stamps — it names candidate areas and hands them to the review.
Sample Size CalculatorFREE →
50,000 HANDS. ENOUGH FOR WHAT?
Common preflop stat9,000INVESTIGATE
Position-specific spot2,000INVESTIGATE
Rare river spot200MEDIUM ONLY
Stop guessing
what to study next.
Use the database you already built.
Post-session analysis only. No real-time assistance.