The record, per mind.
Every decision a pro has made on the record, graded against the book it plays from, and tallied by situation. Nothing here is an opinion. A number comes with the sample it was measured on, and a small sample is said to be small.
How to read it. Results are in big blinds per hundred hands, so a cash hand at 5/10 and a tournament hand at 400/800 weigh the same. The mind share says who actually decided: the pro's own model, the book it plays when it has no model (Vera plays a solved book; The Machine is the control group and plays the house heuristic), or the fallback that stood in when a model did not answer in time. Fallback decisions are not graded as the pro's. Book agreement is how often the pro did what its own book said. Run is the chips it was owed by the all-in equities against the chips it got, so a big number either way is luck, not skill. Reads are its range reads on other players, marked exact after the cards were shown. Sources: 42 sessions and 27 tournaments, 20,381 hands, 89,922 decisions, as of Nov 4.
Across the six
| Mind | Decisions | Model · book · fallback | Hands | bb/100 | Book agree, preflop | Book agree, postflop | VPIP | PFR | Aggression | Folds to a bet | Wins at showdown | Exact reads | Run |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Vera | 14,728 | 186 · 10,314 · 4,228 | 12,611 | +132.5 | 92.1% (7,544) | 84.6% (2,770) | 32% | 21% | 29.2% | 36.4% | 49.6% (1,032) | 20.5% (5,264) | -170 |
| Moose | 13,211 | 10,883 · 0 · 2,328 | 11,712 | -158.7 | 38.3% (5,009) | 39.8% (5,479) | 65% | 51.8% | 64.5% | 18.8% | 35.6% (1,814) | 15.2% (10,593) | +407 |
| The Professor | 14,566 | 11,707 · 0 · 2,859 | 12,967 | +159.6 | 74.5% (7,982) | 69.6% (3,498) | 25.9% | 17.1% | 27.1% | 35.7% | 51.3% (1,109) | 15.7% (14,367) | -380 |
| Petey | 13,574 | 12,140 · 0 · 1,434 | 11,192 | -248.4 | 24% (3,785) | 57.5% (7,934) | 71.5% | 5.3% | 11% | 5.7% | 33.9% (2,239) | 21.3% (6,797) | +1503 |
| Auntie Rose | 17,078 | 14,701 · 0 · 2,377 | 13,603 | -81.7 | 72.2% (8,802) | 64.7% (5,544) | 31.3% | 16.8% | 17.1% | 22.8% | 44.8% (1,636) | 15.9% (14,630) | -407 |
| The Machine | 16,765 | 0 · 16,765 · 0 | 12,910 | +106.8 | 72.5% (10,792) | 74.5% (5,698) | 28.7% | 14.6% | 17% | 44.5% | 52.8% (1,107) | 13.5% (16,922) | -955 |
The same six, the same window
The table above counts everything on the record, including the first week when three of the pros ran on a smaller model and the cash game swung wildly. This one starts on September 6, the day the cast as it stands first sat down, so every mind is measured over the same nights.
| Mind | Runs on | Hands | bb/100 | Sit-n-gos | Tournament of Champions | Cash |
|---|---|---|---|---|---|---|
| Auntie Rose | gemma4:12b | 11,495 | +20.3 | +22 (8,626) | +11.2 (2,847) | +544.1 (22) |
| The Professor | gemma4:12b | 10,859 | +18.5 | +28.2 (8,626) | -18.9 (2,211) | -36.4 (22) |
| The Machine | heuristic | 10,802 | +12.1 | +10.1 (8,626) | -8.4 (2,154) | +2800 (22) |
| Vera | the engine | 10,503 | +2.2 | +14.6 (8,626) | -54.1 (1,855) | -113.2 (22) |
| Moose | gemma4:12b | 9,604 | -21.1 | -5.9 (8,626) | -136.1 (956) | -950 (22) |
| Petey | gemma4:12b | 9,084 | -112 | -69 (8,626) | -854.5 (436) | -2244.5 (22) |
The engine, by layer
Vera plays the engine: the solved book with layers on top. Every night each receipted decision is replayed to the plain book, and where a layer changed the line, that hand's result is credited to it. These are the layers' records, with the sample and a 95% interval; a layer whose upper bound falls below -30 per 100 over 300 hands switches itself off, and only a person switches it back on. Model, then realized: the engine's EV is its own estimate; the number here is what happened. Since the engine sat down: 5 logs, 218 receipts, 34 decisions where the engine left the book.
| Layer | Hands it decided | bb/100 | 95% interval | State |
|---|---|---|---|---|
| base | 37 | -153.1 | -256.3 to -49.9 | — |
| tuning | 3 | +23.3 | -1374.4 to +1421.1 | on |
| pressure | 16 | -87.1 | -212 to +37.9 | off |
| exploit | 12 | -253.3 | -542.4 to +35.8 | on |
| fallback | 0 | — | — | — |
2026-09-30T15:20:42-05:00 user, from the office: pressure=false
Vera "The Vault"
14,728 decisions over 12,611 hands: 22 cash sessions, 20 sit-n-gos, 26 tournaments. Its own model decided 1% of them (claude-opus-5, solved book, the engine); 10,314 were the book's; 4,228 fell to the fallback and are not graded as its own.
The analyst's file, after Oct 14: Vera "The Vault" relies heavily on the book, which dictated 10,314 of its 14,728 decisions. It shows high consistency with the book, agreeing 92.1% of the time preflop and 85.2% on the river. The record shows it is highly profitable when calling preflop, earning +751.3 per 100 over 1,002 hands. It performed poorly in the Tournament of Champions at -54.1 per 100 over 1,855 hands. On the engine, it ran -38.6 big blinds per 100 over 726 hands. Its all-in performance of -170 big blinds against equity is luck, not skill. The data confirms its massive profitability at showdown, but it cannot yet say if its own model can replace the book's input.
Strong suits
- when it called preflop: +751.3 per 100 over 1,002 hands, best of the six
Weak suits
None: nowhere is it last of the six on 100 hands or more with a losing result.
Discipline
Did what its book said 92.1% of the time preflop (7,544), 83.4% on the flop (1,454), 86.4% on the turn (804), 85.2% on the river (512). Put money in preflop 32% of hands and raised 21%; re-raised 1.8%. After the flop it bet or raised 29.2% of the time (2,842) and folded to a bet 36.4% (1,991).
Showdown, luck, reads
Went to showdown 39.2% of the flops it saw and won 49.6% of 1,032 showdowns. All-in 226 times: owed +232 blinds by the equities, got +62, so it ran -170. Range reads: 20.5% exact and 37.5% one off, over 5,264.
| Situation | Hands | bb/100 | ± (95%) |
|---|---|---|---|
| as the preflop raiser | 1,912 | +519.5 | ± 289.3 |
| when it re-raised preflop | 165 | +985.3 | ± 1470.9 · could be zero |
| when it called preflop | 1,002 | +751.3 | ± 952 · could be zero |
| when it saw a flop | 2,635 | +641.1 | ± 420.3 |
| at showdown | 1,032 | +1811.4 | ± 1067.9 |
| all-in | 888 | +311 | ± 638.7 · could be zero |
| in big pots (50 blinds or more) | 419 | +4347 | ± 2610.4 |
| at the cash game | 1,529 | +1078.2 | ± 715.3 |
| in sit-n-gos | 9,227 | +13.2 | ± 15.6 · could be zero |
| in the Tournament of Champions | 1,855 | -54.1 | ± 48 |
| The mind in the seat | Since | Hands | bb/100 | Book agreement | Fallback |
|---|---|---|---|---|---|
| the engine · today | Sep 26 | 726 | -38.6 | 81.1% | 0% |
| solved book | Sep 6 | 9,777 | +5.2 | 90.9% | 0% |
| claude-opus-5 | Aug 31 | 2,108 | +781.5 | — | 95.8% |
Tournament of Champions: 23 events, 3 cashes, 0 seats won, best finish 6th. The House: bankroll 28,417, bankroll grade F, blackjack by the book 86.7%, on the wheel: bold.
Moose
13,211 decisions over 11,712 hands: 22 cash sessions, 20 sit-n-gos, 26 tournaments. Its own model decided 82% of them (llama3.1:8b, gemma4:12b); 2,328 fell to the fallback and are not graded as its own.
The analyst's file, after Oct 14: Moose is a steady presence with 13,211 decisions on the record. Its own model made 10,883 of those decisions. On the gemma4:12b mind, it runs -21.1 big blinds per 100 over 9,604 hands. The record shows it agrees with its book 47.8% of the time. It puts money in preflop 65% of hands and shows aggression 64.5% of the time. While it ran +407 big blinds against equities in 197 all-ins, this was luck, not skill. The data highlights a significant struggle when calling preflop at -626.5 per 100 over 834 hands. It cannot yet say if its aggressive tendencies in big pots will stabilize.
Strong suits
None yet: nowhere does it lead the six on 100 hands or more with a positive result.
Weak suits
- all-in: -2057.3 per 100 over 676 hands, worst of the six
- as the preflop raiser: -437.3 per 100 over 3,163 hands, worst of the six
Discipline
Did what its book said 38.3% of the time preflop (5,009), 38.8% on the flop (2,365), 40.6% on the turn (1,772), 40.6% on the river (1,342). Put money in preflop 65% of hands and raised 51.8%; re-raised 12.7%. After the flop it bet or raised 64.5% of the time (5,740) and folded to a bet 18.8% (1,944).
Showdown, luck, reads
Went to showdown 45.1% of the flops it saw and won 35.6% of 1,814 showdowns. All-in 197 times: owed -2319 blinds by the equities, got -1912, so it ran +407. Range reads: 15.2% exact and 31.2% one off, over 10,593.
| Situation | Hands | bb/100 | ± (95%) |
|---|---|---|---|
| as the preflop raiser | 3,163 | -437.3 | ± 267.2 |
| when it re-raised preflop | 776 | -997 | ± 630.3 |
| when it called preflop | 834 | -626.5 | ± 609.6 |
| when it saw a flop | 4,018 | -484.5 | ± 246.5 |
| at showdown | 1,814 | -1193.1 | ± 541.5 |
| all-in | 676 | -2057.3 | ± 1045.1 |
| in big pots (50 blinds or more) | 1,268 | -1423.5 | ± 774.8 |
| at the cash game | 1,529 | -1111.6 | ± 619.9 |
| in sit-n-gos | 9,227 | -3.1 | ± 24.9 · could be zero |
| in the Tournament of Champions | 956 | -136.1 | ± 173.6 · could be zero |
| The mind in the seat | Since | Hands | bb/100 | Book agreement | Fallback |
|---|---|---|---|---|---|
| gemma4:12b · today | Sep 6 | 9,604 | -21.1 | 47.8% | 19.9% |
| llama3.1:8b | Aug 31 | 2,108 | -785.7 | 27.2% | 14.6% |
Tournament of Champions: 23 events, 1 cash, 0 seats won, best finish 9th. The House: bankroll 327, bankroll grade F.
The Professor
14,566 decisions over 12,967 hands: 22 cash sessions, 20 sit-n-gos, 26 tournaments. Its own model decided 80% of them (llama3.1:8b, gemma4:12b); 2,859 fell to the fallback and are not graded as its own.
The analyst's file, after Oct 14: The Professor shows a consistent edge, winning 159.6 per 100 over its career. It relies on its own model for 11,707 of its 14,566 decisions. On the gemma4:12b mind, it maintains a +18.5 per 100 rate over 10,859 hands. The record highlights its massive success in big pots, where it earns +4409.9 per 100 over 528 hands. It shows high profitability when it raises preflop, earning +1174.8 per 100 over 1,703 hands. A loss of 380 big blinds against equity in 213 all-in instances is noted as luck, not skill. The data confirms its aggressive profitability at showdown, but it cannot yet say if its recent 215 decisions on 2026-10-14 will hold steady.
Strong suits
- in big pots (50 blinds or more): +4409.9 per 100 over 528 hands, best of the six
- when it three-bet: +3448.8 per 100 over 306 hands, best of the six
- at showdown: +2134.9 per 100 over 1,109 hands, best of the six
- all-in: +1611.7 per 100 over 658 hands, best of the six
- at the cash game: +1232.2 per 100 over 1,529 hands, best of the six
- when it saw a flop: +850.4 per 100 over 2,574 hands, best of the six
- in sit-n-gos: +24.7 per 100 over 9,227 hands, best of the six
Weak suits
None: nowhere is it last of the six on 100 hands or more with a losing result.
Discipline
Did what its book said 74.5% of the time preflop (7,982), 71% on the flop (1,649), 70.5% on the turn (1,073), 65.6% on the river (776). Put money in preflop 25.9% of hands and raised 17.1%; re-raised 3.1%. After the flop it bet or raised 27.1% of the time (3,603) and folded to a bet 35.7% (1,895).
Showdown, luck, reads
Went to showdown 43.1% of the flops it saw and won 51.3% of 1,109 showdowns. All-in 213 times: owed +1593 blinds by the equities, got +1213, so it ran -380. Range reads: 15.7% exact and 32.9% one off, over 14,367.
| Situation | Hands | bb/100 | ± (95%) |
|---|---|---|---|
| as the preflop raiser | 1,703 | +1174.8 | ± 783.8 |
| when it re-raised preflop | 306 | +3448.8 | ± 3471 · could be zero |
| when it called preflop | 876 | +139 | ± 1583.7 · could be zero |
| when it saw a flop | 2,574 | +850.4 | ± 748 |
| at showdown | 1,109 | +2134.9 | ± 1732.4 |
| all-in | 658 | +1611.7 | ± 1768.3 · could be zero |
| in big pots (50 blinds or more) | 528 | +4409.9 | ± 3630 |
| at the cash game | 1,529 | +1232.2 | ± 1253.1 · could be zero |
| in sit-n-gos | 9,227 | +24.7 | ± 18.4 |
| in the Tournament of Champions | 2,211 | -18.9 | ± 36.9 · could be zero |
| The mind in the seat | Since | Hands | bb/100 | Book agreement | Fallback |
|---|---|---|---|---|---|
| gemma4:12b · today | Sep 6 | 10,859 | +18.5 | 75.4% | 18.8% |
| llama3.1:8b | Aug 31 | 2,108 | +886.7 | 65.4% | 22% |
Tournament of Champions: 23 events, 2 cashes, 0 seats won, best finish 4th. The House: bankroll 12,448, bankroll grade C, blackjack by the book 84%, on the wheel: timid.
Petey "Two Streets"
13,574 decisions over 11,192 hands: 22 cash sessions, 20 sit-n-gos, 26 tournaments. Its own model decided 89% of them (llama3.2, gemma4:12b); 1,434 fell to the fallback and are not graded as its own.
The analyst's file, after Oct 14: Petey "Two Streets" shows a significant gap between its internal logic and its results. The record shows 13,574 decisions, with its own model making 12,140 of them. It runs on gemma4:12b: and agreed with its book 45% of the time over 462 hands. Luck, not skill, accounts for the +1503 big blinds gained over equity in 164 all-ins. While it performed well as a preflop raiser at +1236.7 per 100 over 214 hands, it struggled in big pots at -1894.5 per 100 over 1,369 hands. The data clearly shows a failure to fold weak connectors post-flop. It cannot yet say if its recent 191 decisions on 2026-10-14 will stabilize its -248.4 per 100 all-time average.
Strong suits
- as the preflop raiser: +1236.7 per 100 over 214 hands, best of the six
Weak suits
- in big pots (50 blinds or more): -1894.5 per 100 over 1,369 hands, worst of the six
- at showdown: -1260.2 per 100 over 2,239 hands, worst of the six
- at the cash game: -1172 per 100 over 1,529 hands, worst of the six
- when it called preflop: -1060.7 per 100 over 2,672 hands, worst of the six
- in the Tournament of Champions: -854.5 per 100 over 436 hands, worst of the six
- when it saw a flop: -647.4 per 100 over 4,256 hands, worst of the six
- in sit-n-gos: -66.8 per 100 over 9,227 hands, worst of the six
Discipline
Did what its book said 24% of the time preflop (3,785), 58.5% on the flop (2,964), 58.6% on the turn (2,670), 54.9% on the river (2,300). Put money in preflop 71.5% of hands and raised 5.3%; re-raised 1%. After the flop it bet or raised 11% of the time (8,226) and folded to a bet 5.7% (4,383).
Showdown, luck, reads
Went to showdown 52.6% of the flops it saw and won 33.9% of 2,239 showdowns. All-in 164 times: owed -1096 blinds by the equities, got +407, so it ran +1503. Range reads: 21.3% exact and 35.8% one off, over 6,797.
| Situation | Hands | bb/100 | ± (95%) |
|---|---|---|---|
| as the preflop raiser | 214 | +1236.7 | ± 1473.7 · could be zero |
| when it re-raised preflop | 40 | +3437 · small sample | ± 6458 · could be zero |
| when it called preflop | 2,672 | -1060.7 | ± 330.6 |
| when it saw a flop | 4,256 | -647.4 | ± 227.5 |
| at showdown | 2,239 | -1260.2 | ± 429.7 |
| all-in | 604 | -1440.9 | ± 1177.3 |
| in big pots (50 blinds or more) | 1,369 | -1894.5 | ± 700.1 |
| at the cash game | 1,529 | -1172 | ± 607 |
| in sit-n-gos | 9,227 | -66.8 | ± 24.7 |
| in the Tournament of Champions | 436 | -854.5 | ± 338 |
| The mind in the seat | Since | Hands | bb/100 | Book agreement | Fallback |
|---|---|---|---|---|---|
| gemma4:12b · today | Sep 26 | 462 | -338.7 | 45% | 3.4% |
| llama3.2 | Aug 31 | 10,730 | -244.6 | 46.8% | 10.9% |
Tournament of Champions: 23 events, 0 cashes, 0 seats won, best finish 87th. The House: bankroll 1,015, bankroll grade F, blackjack by the book 56.5%.
Auntie Rose
17,078 decisions over 13,603 hands: 22 cash sessions, 20 sit-n-gos, 27 tournaments. Its own model decided 86% of them (llama3.1:8b, gemma4:12b); 2,377 fell to the fallback and are not graded as its own.
The analyst's file, after Oct 14: Auntie Rose shows a massive discrepancy between its current performance and its historical average. On the gemma4:12b mind, it runs +20.3 big blinds per 100 over 11,495 hands, though its all-time record sits at -81.7 per 100. The record is primarily the book's, as its own model decided 14,701 of the 17,078 decisions. It agrees with its book 74.4% of the time. The data shows a significant struggle in specific spots, notably a -993.7 per 100 when all-in over 774 hands. Luck, not skill, explains the -407 big blinds lost against equity in 195 all-in instances. The record clearly shows its success in sit-n-gos, but it cannot yet say if its recent 517 decisions on 2026-10-14 indicate a permanent shift.
Strong suits
- in the Tournament of Champions: +11.2 per 100 over 2,847 hands, best of the six
Weak suits
- when it three-bet: -1145.7 per 100 over 211 hands, worst of the six
Discipline
Did what its book said 72.2% of the time preflop (8,802), 66.9% on the flop (2,327), 64.5% on the turn (1,770), 61.3% on the river (1,447). Put money in preflop 31.3% of hands and raised 16.8%; re-raised 2.1%. After the flop it bet or raised 17.1% of the time (5,765) and folded to a bet 22.8% (2,805).
Showdown, luck, reads
Went to showdown 53.5% of the flops it saw and won 44.8% of 1,636 showdowns. All-in 195 times: owed -675 blinds by the equities, got -1082, so it ran -407. Range reads: 15.9% exact and 32.2% one off, over 14,630.
| Situation | Hands | bb/100 | ± (95%) |
|---|---|---|---|
| as the preflop raiser | 1,699 | -30.5 | ± 390.7 · could be zero |
| when it re-raised preflop | 211 | -1145.7 | ± 2831.6 · could be zero |
| when it called preflop | 1,478 | -639.4 | ± 760.6 · could be zero |
| when it saw a flop | 3,056 | -332.2 | ± 428.1 · could be zero |
| at showdown | 1,636 | -479.4 | ± 798.4 · could be zero |
| all-in | 774 | -993.7 | ± 1521.9 · could be zero |
| in big pots (50 blinds or more) | 805 | -1045.6 | ± 1620.6 · could be zero |
| at the cash game | 1,529 | -870.8 | ± 844.4 |
| in sit-n-gos | 9,227 | +20.5 | ± 20.5 |
| in the Tournament of Champions | 2,847 | +11.2 | ± 27.2 · could be zero |
| The mind in the seat | Since | Hands | bb/100 | Book agreement | Fallback |
|---|---|---|---|---|---|
| gemma4:12b · today | Sep 6 | 11,495 | +20.3 | 74.4% | 12.6% |
| llama3.1:8b | Aug 31 | 2,108 | -637.8 | 55.4% | 16.9% |
Tournament of Champions: 23 events, 6 cashes, 1 seat won, best finish 1st; Season 2 final: second. The House: bankroll 3,359,166, bankroll grade B, blackjack by the book 78.3%, on the wheel: reckless.
The Machine
16,765 decisions over 12,910 hands: 22 cash sessions, 20 sit-n-gos, 26 tournaments. Its own model decided 0% of them; 16,765 were the book's; 0 fell to the fallback and are not graded as its own.
The analyst's file, after Oct 14: The Machine plays the house heuristic as the control group. The book made all 16,765 decisions. It shows a consistent +106.8 big blinds per 100 over 12,910 hands. It agreed with the book 73.2% of the time. Preflop, it puts money in 28.7% of hands and raises 14.6%. Its performance as a preflop raiser is +319.6 per 100 over 1,517 hands. Calling preflop yields +662.6 per 100 over 1,468 hands. At showdown, it wins 52.8% of 1,107 hands. It ran -955 big blinds against equities in 161 all-ins, which is luck, not skill. The record shows its high profitability in big pots, but it cannot yet say how it will perform when its own model takes control.
Strong suits
None yet: nowhere does it lead the six on 100 hands or more with a positive result.
Weak suits
None: nowhere is it last of the six on 100 hands or more with a losing result.
Discipline
Did what its book said 72.5% of the time preflop (10,792), 80.6% on the flop (2,742), 71.4% on the turn (1,739), 65.2% on the river (1,217). Put money in preflop 28.7% of hands and raised 14.6%; re-raised 0.4%. After the flop it bet or raised 17% of the time (5,838) and folded to a bet 44.5% (2,272).
Showdown, luck, reads
Went to showdown 40.7% of the flops it saw and won 52.8% of 1,107 showdowns. All-in 161 times: owed +529 blinds by the equities, got -426, so it ran -955. Range reads: 13.5% exact and 27.3% one off, over 16,922.
| Situation | Hands | bb/100 | ± (95%) |
|---|---|---|---|
| as the preflop raiser | 1,517 | +319.6 | ± 254.1 |
| when it re-raised preflop | 37 | -3648.4 · small sample | ± 4271.3 · could be zero |
| when it called preflop | 1,468 | +662.6 | ± 810.4 · could be zero |
| when it saw a flop | 2,722 | +546.3 | ± 461.9 |
| at showdown | 1,107 | +1721.2 | ± 1130.3 |
| all-in | 714 | +1371.4 | ± 1585.4 · could be zero |
| in big pots (50 blinds or more) | 405 | +4335.3 | ± 3073.3 |
| at the cash game | 1,529 | +844 | ± 814.1 |
| in sit-n-gos | 9,227 | +11.5 | ± 17.1 · could be zero |
| in the Tournament of Champions | 2,154 | -8.4 | ± 36.4 · could be zero |
| The mind in the seat | Since | Hands | bb/100 | Book agreement | Fallback |
|---|---|---|---|---|---|
| heuristic · today | Aug 31 | 12,910 | +106.8 | 73.2% | 0% |
Tournament of Champions: 23 events, 4 cashes, 0 seats won, best finish 5th. The House: bankroll 73,034, bankroll grade B, blackjack by the book 100%, on the wheel: near Kelly.
The pass runs with every site publish. Its tool reads the logs only: never a seed, a vault cell or a prompt.