{ "schema_version": "1.3", "rebuild": "v2", "provenance_class": "dealt_hand", "table_size": 6, "id": "tt-4bet-isolate-the-equity-thief", "format": "multi_option", "free": true, "updated_at": "2026-08-08T00:00:00Z", "source": { "session_id": "SmallBetsandHandReview", "verdict_audio": "quiz/audio/tt-4bet-isolate-the-equity-thief.m4a", "verdict_beats": "s1414-s1415 (t=5589.01-5599.61) + s1474-s1476 (t=5838.02-5844.00) — the beats already registered in tools/cut_quiz_audio.py. Clip NOT re-cut this rebuild: 17.53 s, all five sentences are tyler + spk_confident, all five are about THIS node (hero's preflop decision facing the button's 3-bet and the big blind's cold-call), and the only number aired is this hand's own equity delta. Echo-checked 2026-08-08 (faster-whisper small.en re-ASR of the shipped m4a, plus medium.en on the raw session span) — see aired_transcript. Screen during beat 1: the replayer is on this hand, rewound to the deal (own frame t=5590, 'Pot: $3', all six hands face-up); one blip-second earlier, at t=5589.0, the title bar reads a DIFFERENT marked hand (#4397907833) — Tyler is stepping between marked hands and lands on this one as he starts the sentence. Screen during beat 2: the same replayer parked at the big blind's decision ('Pot: $25') with Tyler's HoldEq window floating over it, seeded with this hand's three hand classes TT / AQ / 42o (own frame t=5841).", "lesson_slug": null, "master_slug": "SmallBetsandHandReview_master", "master_t": 5105.8, "src_t": 5589.0, "ser_sentences": [1412, 1414, 1415, 1416, 1425, 1426, 1428, 1430, 1433, 1434, 1438, 1439, 1442, 1443, 1444, 1445, 1452, 1453, 1459, 1461, 1464, 1465, 1467, 1469, 1470, 1473, 1474, 1475, 1476, 1478, 1486, 1490, 1491], "hand_record": { "lane": "REPLAYER (FRAME_READ)", "site": "Ignition hand reviewed in the PokerTracker 4 replayer ('Powered by PokerTracker' badge, bottom right). Frames are READ-ONLY provenance and are never aired — the title bar carries the poker room's name and the felt carries a Run It Once watermark.", "hand_id": "#4397840616", "hand_started": "2023/01/01 8:13 pm", "stake": { "source": "own frames — replayer title bar at t=5570/5583-5588/5590-5599/5841: 'Playback for Ignition Hand #4397840616 (2) - $2 NL FAST (6 max) - Holdem'; and, decisively, the action log's own posts at t=5570 (zoom f/log5570.png): 'P3-3b14f0b1a64a8816c85938cc87b4a783 posts the small blind 1.00' / 'P4-0a698e245855a2be41a77fbbc5822fc9 posts the big blind 2.00'.", "sb": 1.0, "bb": 2.0, "usd_per_bb": 2.0, "note": "The $-to-bb conversion in this fixture is licensed by the blind posts printed in THIS hand's own log, not by the session title and not by any tool sidebar. Every bb figure below is the log's dollar figure / 2." }, "frames": [ "t=5570 (scratchpad dh3/tt-4bet-isolate-the-equity-thief/f/t5570.jpg; zooms log5570.png, hero5570.png, p1.png, p2.png, p4.png, p5.png) — hand start, 'Pot: $3', ALL SIX hole hands face-up in the 4-colour deck: Hero T(blue=♦) T(red=♥) $208.38; P1 T(grey=♠) 5(red=♥) $186.27; P2 A(green=♣) Q(red=♥) $223.98 with the green dealer button beside the seat; P3 Q(♣) 3(♦) POST SB; P4 4(grey=♠) 2(blue=♦) POST BB; P5 9(♠) Q(♦) $286.60.", "t=5583-5588 (f/s5583..s5588.jpg, title strip f/titlestrip.png) — same hand, log fully scrolled: the complete preflop skeleton is on screen. t=5585 is the post-preflop state ('Pot: $52', hero tagged CALL, $17 in front of hero / P2 / P4, $1 from P3) with the replayer's three-way equity readout 'Hero - 48% / P2-f81177c16... - 38% / P4-0a698e245... - 15%'. t=5588 has advanced to the deal ('Flop: [Kc 6h 5d]', 'Pot: $52', Hero 55% / P2 23% / P4 21').", "t=5589.0 (f/t5589.jpg) — ONE second of a different marked hand: title bar 'Playback for Ignition Hand #4397907833 (2) - $2 NL FAST (6 max) - Holdem', hero Q-Q $205. Recorded because the first aired word lands here; by t=5590.0 the title bar is #4397840616 again and stays there through the whole beat.", "t=5590-5599 (f/t5590.jpg, f/t5600.jpg + log zoom f/log5600.png) — this hand rewound to the deal ('Pot: $3', P5 green = to act) for the whole of the aired ruling; the log pane at t=5600 shows the skeleton through 'Flop: [Kc 6h 5d]'.", "t=5841 (f/t5841.jpg; zoom f/holdeq.png) — HoldEq (v1.1.7) floating over the same replayer (title bar still #4397840616). Range 1 'TT' Equity 47.62% / Wins 47.56% / Tie 0.35%; Range 2 'AQ' 38.14% / 38.02% / 0.35%; Range 3 '42o' 14.19% / 14.07% / 0.35%; Board and Dead boxes empty. The felt beneath is parked at the big blind's decision: 'Pot: $25', 'Odds: 1.67:1 - 37%', hero $203.38 behind with $5 out, P2 $206.98 behind with $17 out, P4 $177.04 behind and highlighted to act.", "t=5620 / 5900 / 5960 / 6010 (f/xlogstrip.png) — log-scroll sweep across the rest of the teaching window: the replayer never advances past 'Flop: [Kc 6h 5d]', so no postflop action line is log-verbatim from my frames." ], "log_verbatim": [ "Ignition Hand #4397840616 begins at 2023/01/01 8:13 pm", "Seat 2 (P2-f81177c168bf4544db18cdf8b019d70d) is the dealer", "P3-3b14f0b1a64a8816c85938cc87b4a783 posts the small blind 1.00", "P4-0a698e245855a2be41a77fbbc5822fc9 posts the big blind 2.00", "P5-22a452baf9370802cec25dbf2484e9f1 folds", "Hero raises to 5.00", "P1-d8a9ea25e2903f4f1cb0819ee5dc8d59 folds", "P2-f81177c168bf4544db18cdf8b019d70d raises to 17.00", "P3-3b14f0b1a64a8816c85938cc87b4a783 folds", "P4-0a698e245855a2be41a77fbbc5822fc9 calls 15.00", "Hero calls 12.00", "Flop: [Kc 6h 5d]" ], "table_structure": "6-max ring, all six seats dealt in. No ante and no straddle — the log's only posts are the 1.00 small blind and the 2.00 big blind. Dealer = Seat 2 = P2. Preflop action order in the log is P5, Hero, P1, P2, P3, P4, so the seat map is UTG=P5, HJ=Hero, CO=P1, BTN=P2, SB=P3, BB=P4. Screen geometry agrees: the green dealer disc sits on P2 (right seat), the next seat clockwise (P3, bottom right) posts the small blind and the one after (P4, bottom left) posts the big blind. Hero is the HIJACK, not the cutoff — UTG folds BEFORE him and the cutoff folds BEHIND him. Six-handed, so no heads-up button rule applies.", "decision_node": "Hero's second preflop decision: he has opened to 5.00 (2.5 bb), the button has 3-bet to 17.00 (8.5 bb), the small blind's 1.00 (0.5 bb) is dead in the middle, and the big blind has cold-called the 3-bet for 17.00 (8.5 bb). Pot at the decision = 1.00 + 5.00 + 17.00 + 17.00 = $40.00 = 20 bb; 12.00 = 6 bb more for hero to call. The replayer prints the immediately preceding node itself at t=5841 ('Pot: $25' = 1 + 2 + 5 + 17, before the big blind acts).", "stacks": { "note": "All figures read off the seat plates; bb = dollars / 2 per the stake block. 'behind' = chips left at hero's decision.", "hero": "start $208.38 = 104.19 bb; behind after the 2.5 bb open $203.38 = 101.69 bb", "BTN (P2)": "start $223.98 = 111.99 bb; behind after the 8.5 bb 3-bet $206.98 = 103.49 bb", "BB (P4)": "start $179.04 = 89.52 bb (frame t=5841 shows $177.04 behind after the 2.00 post, t=5588 shows $162.04 behind after the 17.00 call); behind at hero's decision $162.04 = 81.02 bb", "CO (P1)": "$186.27 = 93.14 bb (folded)", "SB (P3)": "start $241.22 = 120.61 bb, $240.22 = 120.11 bb behind after the dead 1.00 (folded)", "UTG (P5)": "$286.60 = 143.30 bb (folded)", "effective_at_decision": "vs the big blind 89.52 bb, vs the button 104.19 bb — the money that matters for the equity-theft argument is the big blind's ~89.5 bb, which is why no single 'N BB deep' figure is displayed." }, "equities_on_screen": { "three_way_all_in_at_this_node": "replayer readout at t=5585 — Hero 48% / P2 38% / P4 15% (these ARE the exact holdings: every hand is face-up on the felt); Tyler's HoldEq re-run at t=5841 — TT 47.62% / AQ 38.14% / 42o 14.19%, typed as the three hand CLASSES, not the specific suits (his 'put this exact combo in' at s1473 refers to narrowing 42 to 42o, and the Range fields read 'TT' / 'AQ' / '42o'). Class-level and combo-level agree to 0.05 pts here, so nothing displayed turns on the difference.", "independent_recompute": "my own exhaustive enumeration (scratchpad eq.py, all C(46,5)=1,370,754 runouts): T♦T♥ 47.67% / A♣Q♥ 37.59% / 4♠2♦ 14.75%. Heads-up, all C(48,5)=1,712,304 runouts: T♦T♥ 56.95% / A♣Q♥ 43.05%. 56.95 - 47.67 = 9.28 points — the '9%' Tyler reads out." }, "result": "The real hero CALLED the 3-bet ('Hero calls 12.00') and the deal came K♣ 6♥ 5♦. The quiz therefore grades the should-have, not the line taken — which is exactly how Tyler frames it and how Eric receives it (s1478: 'I just wouldn't have realized I could four-bet go with it'). No action past the deal is log-verbatim: across every frame sampled from t=5570 to t=6010 the replayer never advances beyond 'Flop: [Kc 6h 5d]', so the outcome is deliberately not recorded and not displayed.", "gated_material_deliberately_left_off_surface": "Two screens Tyler leans on near these beats are rule-7 gated and appear nowhere in this fixture: the population 'Cold Call 3bet [TOT]' range grid he narrates at s1441 ('what they've overcalled in the big blind here versus two players over the last half million hands'), and the GTO Wizard 3-bet frequency at s1424-1426 with the 'tens are good roughly 75% of the time' figure derived from it. The direction survives in the displayed text ('his cold-calling range is stuffed with pocket pairs', 'the button 3-bets wide'); the magnitudes do not." } }, "display_stake": "$1/$2", "context": "ERIC'S OWN HAND · $1/$2 6-MAX · PREFLOP", "hero": { "pos": "HJ", "cards": ["Td", "Th"], "stack_bb": 101.69, "avatar": "lono_ghoul", "button": false }, "seats": [ {"pos": "CO", "avatar": "ghoul_2624", "folded": true}, {"pos": "BTN", "avatar": "ghoul_3143", "stack_bb": 103.49, "button": true}, {"pos": "SB", "avatar": "ghoul_349", "folded": true}, {"pos": "BB", "avatar": "ghoul_2992", "name": "REC", "stack_bb": 81.02, "villain": true}, {"pos": "UTG", "avatar": "ghoul_4191", "folded": true} ], "streets": [ { "name": "preflop", "actions": [ {"pos": "UTG", "act": "fold"}, {"pos": "hero", "act": "open", "to_bb": 2.5}, {"pos": "CO", "act": "fold"}, {"pos": "BTN", "act": "3bet", "to_bb": 8.5}, {"pos": "SB", "act": "fold"}, {"pos": "BB", "act": "call"} ], "pot_bb": 20 } ], "prompt": "T♦T♥ · THE BUTTON 3-BETS AND THE BIG BLIND COLD-CALLS — 4-BET, FLAT OR FOLD?", "pot_bb": 20, "options": [ {"id": "fourbet", "label": "4-BET"}, {"id": "flat", "label": "FLAT"}, {"id": "fold", "label": "FOLD"} ], "correct": "fourbet", "option_feedback": { "fourbet": "Correct — the cold-caller is, in Tyler's words, 'completely nonsensical in value', so your tens are far ahead of the junk he came along with, and the button is 3-betting a range wide enough that tens beat it more often than not. That is what makes tens stackable here. The 4-bet also charges the recreational for the equity he is helping himself to: three-handed, all-in right now, your tens are about 48%; against the ace-queen alone they are about 57%. He had four-deuce offsuit.", "flat": "This is what actually happened, and it is the leak. Flatting keeps the equity thief in the pot: three-handed your tens are about 48%, against the ace-queen alone they are about 57%. Calling hands him nine points of your equity for nothing, and his cold-calling range is stuffed with pocket pairs that beat you when they connect. If tens are good enough to continue with here, they are good enough to raise with.", "fold": "Too weak — the cold-call is what makes tens strong here, not weak. The overcaller is 'completely nonsensical in value' and the button 3-bets wide, so your tens hold the biggest single share of the equity of the three hands in the pot. Folding hands that share away. Tens are a 4-bet here, not a fold." }, "verdict_text": "When this overcaller comes in, tens is stackable — because the overcaller is just completely nonsensical in value. He stole 9% of your equity here. So it's awfully important to take him out of the pot.", "aired_transcript": "So there's something you probably should know here, which is basically when this overcaller comes in, tens is stackable. Because the overcaller is just completely nonsensical in value. Yeah, 9%. He stole 9% of your equity here. So it's awfully important to take him out of the pot.", "aired_transcript_note": "Re-ASR of quiz/audio/tt-4bet-isolate-the-equity-thief.m4a (17.53 s; beats s1414-s1415 + s1474-s1476 per tools/cut_quiz_audio.py), faster-whisper small.en, run 2026-08-08. Raw output verbatim: 'So there's something you probably shouldn't know here / Which is basically when this over color comes in tens of stackable because the over color is just completely nonsensical and value / Yeah, 9% he sold 9% of your equity here. So it's awfully important to take him out of the pot'. Four ASR garbles are normalised above against the SER words: 'shouldn't know'→'should know' (s1414), 'over color'→'overcaller' (x2), 'tens of stackable'→'tens is stackable', 'nonsensical and value'→'nonsensical in value'. One judgement call: both small.en and medium.en (re-run on the raw session span 5832-5848) hear 'sold', and SER s1475 also transcribed 'sold', but the construction is 'sold 9% of your equity from you' nonsense — the parallel sentence s1461 ('I think this dude stole, like, 15% of your equity, Eric') and s1471 ('he sold 8% of his equity FROM you') show the verb is 'stole', softened in delivery. verdict_text renders it 'stole'. No card, no position, no bet size and no population figure is aired; the only aired number is this hand's own equity delta, which is why the pre-rebuild clip needed no re-cut.", "principle": "A loose cold-caller is not free dead money — he is buying a slice of your equity out of your pocket. When his range is junk and the 3-bettor's range is wide, raise again to charge him and to play the pot heads-up.", "pct_provenance": { "48": { "role": "equity_at_decision", "referent": "Hero's share if all three live players got the stacks in AT THIS PREFLOP NODE and ran it out — T♦T♥ against the button's A♣Q♥ and the big blind's 4♠2♦. Not a range equity, not a solver frequency, not a figure from any later street.", "source": "own frame t=5585 — the replayer's own three-way readout with every hand face-up: 'Hero - 48%' / 'P2-f81177c16... - 38%' / 'P4-0a698e245... - 15%'. Corroborated by Tyler's HoldEq run on screen at t=5841 (TT 47.62% / AQ 38.14% / 42o 14.19%) and by my own exhaustive enumeration of the exact combos over all 1,370,754 runouts: 47.67% / 37.59% / 14.75%." }, "57": { "role": "equity_at_decision", "referent": "Hero's share at the SAME node with the cold-caller removed — T♦T♥ heads-up against the button's A♣Q♥. This is the counterfactual the verdict turns on, not a different street and not a different sizing.", "source": "Tyler, s1467 ('10s here versus Ace Queen has about 57') and s1469 ('It's, like, 57 point something'), spoken over this hand. My own exhaustive enumeration over all 1,712,304 runouts: T♦T♥ 56.95% / A♣Q♥ 43.05%." }, "9": { "role": "fixture_math", "referent": "The equity the cold-caller takes off hero at this decision: hero's heads-up share (~57%) minus his three-handed share (~48%) ≈ 9 percentage points. Single-hand equity arithmetic about these three exact holdings — not a population, database or pool figure.", "source": "Tyler s1473-1476, spoken over the HoldEq window at t=5841 after narrowing the cold-caller to 42o ('there must be a rounding error here, so I guess we should go look and put this exact combo in' → 'Yeah, 9%. He stole 9% of your equity here'). On his screen the delta is 56.25 (his earlier TT-vs-AQ run) − 47.62 = 8.6, which he reads out as 9. My own enumeration of the exact combos: 56.95 − 47.67 = 9.28 points, which also rounds to 9." } }, "size_claims": {}, "claims": [ { "text": "hero is the HIJACK — UTG folds before him, the cutoff folds behind him, the button 3-bets", "basis": "Log verbatim, in order: 'Seat 2 (P2-f81177c168bf4544db18cdf8b019d70d) is the dealer' / 'P3... posts the small blind 1.00' / 'P4... posts the big blind 2.00' / 'P5... folds' / 'Hero raises to 5.00' / 'P1... folds' / 'P2... raises to 17.00'. Action opens with P5, so P5=UTG and Hero is second to act = HJ, P1=CO, P2=BTN, P3=SB, P4=BB. The felt agrees: the green dealer disc is on P2 and the two seats clockwise of it post the blinds. The pre-audit fixture's 'CO' is wrong.", "source": "own frames t=5570 (log zoom f/log5570.png, felt) and t=5600 (log zoom f/log5600.png)" }, { "text": "hero holds T♦T♥", "basis": "4-colour deck at 4x zoom (f/hero5570.png): the first ten is on the blue diamond tile with a ♦ pip, the second on the red tile with a ♥ pip. The pre-audit fixture's T♠ does not exist in this hand — the T♠ is in the CUTOFF's hand (P1 holds T♠5♥, f/p1.png).", "source": "own frame t=5570" }, { "text": "the button 3-bets to 8.5 bb and the big blind cold-calls it; the pot at hero's decision is 20 bb with 6 bb to call", "basis": "Log verbatim 'Hero raises to 5.00' / 'P2... raises to 17.00' / 'P3... folds' / 'P4... calls 15.00'. At $2 per bb (the log's own big-blind post): 5.00=2.5 bb, 17.00=8.5 bb, the dead small blind 1.00=0.5 bb. Pot = 1 + 5 + 17 + 17 = $40 = 20 bb; hero owes 12.00 = 6 bb. The pre-audit fixture's 9 bb 3-bet and 21 bb pot are each off by half a big blind.", "source": "own frames t=5585/5600 (log), t=5841 (felt showing the preceding node, 'Pot: $25')" }, { "text": "the stake is $1/$2, so one big blind is $2", "basis": "The conversion is licensed by THIS hand's own log lines — 'posts the small blind 1.00' and 'posts the big blind 2.00' — read at 3x zoom, not by the session title and not by a tool sidebar. The replayer title bar independently names the game '$2 NL FAST (6 max)'.", "source": "own frames t=5570 (f/log5570.png), title strip f/titlestrip.png" }, { "text": "three-handed all-in at this node hero's tens are about 48%; heads-up against the ace-queen about 57%; the cold-caller costs him roughly nine points", "basis": "See pct_provenance['48'], ['57'], ['9']. Three independent sources agree: the replayer's own readout on the face-up holdings (48/38/15), Tyler's HoldEq run on the hand classes TT/AQ/42o (47.62/38.14/14.19), and my own exhaustive enumeration of the exact combos (47.67/37.59/14.75 three-way; 56.95 heads-up).", "source": "own frames t=5585, t=5841 (f/holdeq.png); SER s1467/s1469/s1474/s1475; scratchpad eq.py" }, { "text": "hero's tens hold the biggest single share of the equity of the three hands in the pot", "basis": "47.67% vs 37.59% vs 14.75% (my enumeration); 47.62 / 38.14 / 14.19 on Tyler's calculator; 48 / 38 / 15 on the replayer. Hero is the equity leader three-handed — which is why the fold option is graded wrong.", "source": "own frames t=5585, t=5841; scratchpad eq.py" }, { "text": "Tyler's ruling is 4-bet, and he is deliberately indifferent about the sizing", "basis": "s1414 'when this overcaller comes in, 10s is stackable'; s1445 'I'm kind of happy to just put the money in and see what happens'; s1476 'it's awfully important to take him out of the pot'. Sizing: s1438 'I don't really have a strong preference on sizing schemes here', s1439 'Jam is guaranteed money, but you probably lose the recreational player most of the time'. The option is therefore labelled '4-BET' with no size attached — no 4-bet amount is spoken anywhere, so none is displayed (G7).", "source": "SER SmallBetsandHandReview s1414, s1438, s1439, s1445, s1476" }, { "text": "the mechanism: the cold-caller's range is junk, so tens beat what he continues with", "basis": "s1415 'the overcaller is just completely nonsensical in value'; s1430 'the other judgment call here is what this guy is over-flatted with'; s1433 'if you kind of break down that range, you're like, oh, he has like 7s a lot'; s1442-1444 'there's a bunch of pocket pairs here... I would be very comfortable with 10s'; s1453 'he probably folds 9s'.", "source": "SER s1415, s1430, s1433, s1442, s1443, s1444, s1453" }, { "text": "the button's 3-betting range is wide enough that tens are ahead of it more often than not", "basis": "s1425-1426: 'the 9% number means that 10s is good here roughly 75% of the time... you're going to have 2.4% of jacks are better, and there's 9% of hands, so you're good 74%'. The DIRECTION is displayed; the frequency and the 74/75% derived from it are not, because they are read off the GTO Wizard range grid Tyler has open (rule 7 — advertise the tool, gate the magnitude). Tyler also states the condition explicitly at s1428: 'I would not stack it at 6% 3-bet, because you're only good about half'.", "source": "SER s1425, s1426, s1428" }, { "text": "the villain tag 'REC' is Tyler's own word for the big blind", "basis": "s1434 'you end up with this scenario where you're like, I'm really good against this recreational player', spoken directly about the over-flatter he has just been describing (s1430, s1433). Also s1452 'I think that this guy doesn't know what's going on'. The pre-audit tags are removed: 'FISH' was ERIC's word (s1431 'I just assume it's a fish'), not Tyler's, and 'REG' on the button is nobody's — Tyler never characterises the 3-bettor beyond his range width.", "source": "SER s1430, s1433, s1434, s1452 (tyler, spk_confident); s1431 is eric" }, { "text": "the big blind actually held four-deuce offsuit", "basis": "Face-up on the replayer at t=5570: the 4 is on the grey spade tile with a ♠ pip, the 2 on the blue tile with a ♦ pip (f/p4.png) — 4♠2♦. Tyler names it on tape too, s1464: 'when the 4 and the deuce come out'. Stated only in the answer feedback, never on the felt, because hero cannot see it at the decision.", "source": "own frame t=5570; SER s1464" }, { "text": "hero is Eric", "basis": "The replayer seat is labelled 'Hero' in a session of Tyler reviewing Eric's play, and Tyler addresses him by name over this exact hand: s1461 'I think this dude stole, like, 15% of your equity, Eric'; s1475 'He stole 9% of YOUR equity here'. Eric answers in the first person about his own decision, s1478: 'I just wouldn't have realized I could four-bet go with it'.", "source": "SER s1461, s1475, s1478" } ], "unverifiable_claims": [ "The verdict is a SHOULD-HAVE: the real hero called the 3-bet ('Hero calls 12.00'). The quiz asks what he should have done, which is the frame Tyler and Eric are both using on tape (s1478), but the felt necessarily stops at the decision and never shows the line taken.", "Tyler's '4-bet' ruling is conditional in a way the two-word option label cannot carry: it holds because the button is 3-betting wide (s1428 — 'I would not stack it at 6% 3-bet, because you're only good about half') and because this particular big blind is a recreational. At a tighter 3-bet frequency the same hand is a flat or a fold, and the quiz cannot express that dial.", "Tyler's own EV walk-through a minute later (s1480-1491) prices a JAM specifically and concludes the spot is 'really cuspy... if he's at 9% placed perfectly against you', with the profit coming from the recreational's leaks rather than from the button. None of those figures (the pot after a jam, the good-when-called percentage, the dollar loss) appear anywhere in this fixture — they belong to a different sizing branch than the un-sized '4-BET' the quiz grades.", "'ERIC'S OWN HAND' in the context line: the hand history itself is anonymised (the seat is 'Hero', the opponents are hashes), so the identification rests on Tyler naming Eric while pointing at that seat (s1461) rather than on pixels alone.", "The postflop result is unknown to this fixture. Across every frame sampled from t=5570 to t=6010 the replayer never advances past 'Flop: [Kc 6h 5d]', so no outcome is recorded, claimed or displayed." ], "cosmetic_invention": [ "Ghoul avatars on every seat (hero lono_ghoul) — art only; the real table shows anonymised hashes (P1-d8a9ea25e…, P2-f81177c16…, P3-3b14f0b1a…, P4-0a698e245…, P5-22a452baf…).", "Seat LABELS UTG / HJ / CO / BTN / SB / BB. The replayer prints no position names; these are the standard 6-max names, and they are FORCED by the log's own action order (P5, Hero, P1, P2, P3, P4) plus 'Seat 2 (P2…) is the dealer'. The mapping is derived, not chosen — only the words are ours.", "The three folded seats (CO, SB, UTG) are rendered without stacks, to keep the felt on the three players still in the pot. Their real stacks are recorded in source.hand_record.stacks.", "Displayed stacks are ROUNDED to two decimals from the dollar plates at $2/bb ($203.38 → 101.69, $206.98 → 103.49, $162.04 → 81.02). No rounding exceeds half a cent.", "NOT cosmetic — every one of these is sourced and must not be changed: hero's T♦T♥ (the second ten is a HEART; the T♠ belongs to the cutoff), hero's seat (HIJACK), the 2.5 / 8.5 / 8.5 bb preflop amounts, the 0.5 bb dead small blind, the 20 bb pot, the 6-max structure with no ante and no straddle, and the button being the 3-bettor with the big blind cold-calling behind." ], "sourced_not_invented": [ "Hero cards T♦T♥ — own frame t=5570 at 4x zoom (f/hero5570.png), 4-colour deck: blue tile with a ♦ pip, red tile with a ♥ pip.", "Position map UTG=P5, HJ=Hero, CO=P1, BTN=P2, SB=P3, BB=P4 — forced by the log's action order plus the dealer line, and confirmed by the dealer disc on the felt.", "Preflop tree and sizes: 1.00 / 2.00 posts, 'Hero raises to 5.00', 'P2… raises to 17.00', 'P4… calls 15.00' — log verbatim at t=5570/5585/5600; pot 20 bb, 6 bb to call.", "Stake $1/$2 (1 bb = $2) — the blind-post lines in this hand's own log; title bar '$2 NL FAST (6 max)' agrees.", "Stacks at the decision: hero $203.38 = 101.69 bb, button $206.98 = 103.49 bb, big blind $162.04 = 81.02 bb — seat plates at t=5588 and t=5841.", "The three-way equity 48 / 38 / 15 — the replayer's own readout at t=5585 on the face-up holdings; Tyler's HoldEq run at t=5841 gives 47.62 / 38.14 / 14.19 on the hand classes TT / AQ / 42o; my own exhaustive enumeration of T♦T♥ / A♣Q♥ / 4♠2♦ gives 47.67 / 37.59 / 14.75.", "The heads-up counterfactual ~57% — Tyler s1467/s1469, verified by exhaustive enumeration (56.95%).", "The big blind's actual holding 4♠2♦ — face-up at t=5570 (f/p4.png); Tyler names it at s1464.", "The 'REC' read — Tyler s1434 ('this recreational player'), s1452 ('this guy doesn't know what's going on')." ], "concept": "four-bet-isolation", "concepts_secondary": [], "topic": "exploiting-recreationals", "difficulty": 4, "decision": { "pot_settled_bb": 0.5, "bets": { "hero": 2.5, "BTN": 8.5, "BB": 8.5 } } }