{ "schema_version": "1.3", "rebuild": "v2", "provenance_class": "dealt_hand", "table_size": 6, "id": "fold-nine-high-flush-river-jam", "format": "call_fold", "free": true, "updated_at": "2026-08-08T00:00:00Z", "source": { "session_id": "RiverCalls", "verdict_audio": "quiz/audio/fold-nine-high-flush-river-jam.m4a", "verdict_beats": "RE-CUT 2026-08-08. New beats s411 + s412 + s413 as three SEPARATE spans (tools/cut_quiz_audio.py), 19.8 s. The old cut spanned s411..s412 as one beat, so it carried the 10.9 s of dead tape between them (SER has no words in 3648.40-3659.31) and then ended on the dangling '...this is a really clear' (s414) — the payoff word 'fold' at 3681.06 is SER-attributed to ERIC (s415) and may not air. s414 dropped: it only restates s411's threshold and cannot end cleanly. All three aired sentences are spk='tyler', spk_confident=true. Echo-checked 2026-08-08 (faster-whisper small.en) — see aired_transcript. At the first aired word (t=3635.58) the PokerTracker replayer is parked on THIS hand's river node (own frame f_3636).", "lesson_slug": null, "master_slug": "RiverCalls_master", "master_t": 3033.3, "src_t": 3635.6, "ser_sentences": [411, 412, 413], "hand_record": { "record_type": "A hand actually played by Tyler and reviewed on tape in the PokerTracker 4 replayer. Frames are READ-ONLY source, never aired.", "hand_id": "#4458334810", "stake": "$2 NL FAST (6 max)", "stake_source": "replayer title bar verbatim, own frame f_3636 (t=3636, the first word of the aired verdict): 'Playback for Ignition Hand #4458334810 - $2 NL FAST (6 max) - Holdem'. Log header line, own frame f_3230: 'Ignition Hand #4458334810 begins at 2023/04/26 7:17 pm'.", "units_note": "SCALE IS ON SCREEN: the replayer displays every amount natively in BB and labels it so — log 'posts the small blind 0.50 BB' / 'posts the big blind 1 BB' / 'raises to 2 BB' / 'bets 39.07 BB', felt chips '57.83 BB' / '39.07 BB' / '292.42 BB', seat plates '66.77 BB' / '0 BB', headline 'Pot: 389.32 BB'. The fixture copies those BB figures; no $-to-bb conversion was performed anywhere, and the title bar's '$2 NL' is left uninterpreted (no on-screen mapping from it to the BB figures exists, and none is needed).", "video_t": [3225, 3690], "frames": [ "scratchpad/dh3/fold-nine-high-flush-river-jam/frames/f_3225.png, f_3230.png (t=3225/3230) — the action log scrolled to its TOP: hand id, date, dealer line, both blind posts and the complete preflop (see log_verbatim)", "f_3300.png (t=3300) — flop decision node: centre pot '6 BB', HJ bet chip '1.83 BB', hero raise chip '7.69 BB', 'Pot: 15.51 BB', 'Odds: 2.65:1 - 27%', hero '124.07 BB', HJ '316.51 BB'", "f_3636.png + zooms z_title_3636 / z_felt_3636 / z_log_3636 (t=3636) — THE DECISION NODE, on screen at the first aired verdict word (s411 t0=3635.58): board 7(red heart) A(black spade) 3(black spade) 8(black spade) 6(blue diamond); hero 9(spade) 5(spade); 'Pot: 389.32 BB'; 'Odds: 3.04:1 - 25%'; centre pot 57.83 BB, hero bet 39.07 BB, HJ 292.42 BB; 'Hero - 98% / 66.77 BB'; 'P4-d8c1174c7... - 2% / 0 BB'; folded seats P1 193.36 BB, P3 109.99 BB, P5 175.84 BB, P6 98 BB", "f_3600.png, f_3660.png, f_3675.png (t=3600/3660/3675) — the same river node still on screen across the whole aired window (verdict audio ends at s413 t1=3670.35); at t=3600 an Anki 'Add' card and a HoldEq window are open OVER the replayer (see unverifiable_claims — a different question, never displayed)" ], "log_verbatim": [ "Ignition Hand #4458334810 begins at 2023/04/26 7:17 pm", "Seat 6 (P6-682990ad23cb80a02a6dfd92ca7107e2) is the dealer", "P1-e6bbfbf13b4eaa1a1cb015fef5c6dfef posts the small blind 0.50 BB", "Hero posts the big blind 1 BB", "P3-10916937a8704b1b9e5b6aa3e20dd7ae folds", "P4-d8c1174c7ac4de5e27cf8b8753df9da9 raises to 2 BB", "P5-922692657f59fc9702d4b9e9407d9b15 folds", "P6-682990ad23cb80a02a6dfd92ca7107e2 folds", "P1-e6bbfbf13b4eaa1a1cb015fef5c6dfef calls 1.50 BB", "Hero calls 1 BB", "Flop: [7h As 3s]", "P1-e6bbfbf13b4eaa1a1cb015fef5c6dfef checks", "Hero checks", "P4-d8c1174c7ac4de5e27cf8b8753df9da9 bets 1.82 BB", "P1-e6bbfbf13b4eaa1a1cb015fef5c6dfef folds", "Hero raises to 7.68 BB", "P4-d8c1174c7ac4de5e27cf8b8753df9da9 calls 5.86 BB", "Turn: [8s]", "Hero bets 18.23 BB", "P4-d8c1174c7ac4de5e27cf8b8753df9da9 calls 18.23 BB", "River: [6d]", "Hero bets 39.07 BB", "P4-d8c1174c7ac4de5e27cf8b8753df9da9 raises to 292.42 BB" ], "table_structure": "6-max, six seats occupied, no straddle and no ante (the log's only posts are the 0.50 BB small blind and the 1 BB big blind). Dealer = Seat 6 = P6, so P1 = SB, Hero = BB, P3 = UTG, P4 = HJ, P5 = CO, P6 = BTN. The flop is THREE-WAY (SB, Hero, HJ); the SB check-folds to hero's flop check-raise, and everything from the turn on is heads-up.", "stacks_bb": { "hero_at_decision": 66.77, "villain_at_decision": 0, "hero_start_derived": "133.75 — 66.77 behind at the river node plus 2 + 7.68 + 18.23 + 39.07 invested; cross-checks against the flop node's '124.07 BB' (133.75 - 2 - 7.68 = 124.07, own frame f_3300)", "villain_start_derived": "320.33 — 0 behind plus 2 + 7.68 + 18.23 + 292.42; cross-checks against the flop node's '316.51 BB' (320.33 - 2 - 1.82 = 316.51, own frame f_3300). The HJ covers hero by 186.58 BB, which is why part of his shove is uncallable — see cosmetic_invention.", "effective_at_the_deal": 133.75 }, "price_at_the_node": "Hero is all-in for his last 66.77 BB if he calls. The callable pot is 57.82 settled + his own 39.07 + the HJ's matched 105.84 = 202.73 BB, so 66.77 to win 202.73 = 3.04:1 = 24.8% breakeven. PokerTracker prints exactly that at this node: 'Odds: 3.04:1 - 25%' (own frame f_3636) — its odds engine already caps the HJ's shove at what hero can be called for, even though its headline pot figure ('Pot: 389.32 BB') does not.", "rounding_note": "PT cent-rounding artifacts, max drift 0.01 BB: the felt's centre chip prints '57.83 BB' where the log's own amounts settle to 57.82 (6 + 15.36 + 36.46), and PT's callable pot is 202.74 where the fixture's is 202.73. The flop chip labels print '1.83'/'7.69' where the log says 'bets 1.82 BB'/'raises to 7.68 BB'; the fixture follows the log, which is internally consistent (1.82 + the logged 'calls 5.86 BB' = 7.68).", "result": "UNKNOWN — no showdown is visible. The HJ's cards are face down on the felt and the replayer is parked at his raise for the whole aired window, so nothing about the outcome is claimed anywhere in this fixture." } }, "context": "$2 NL FAST 6-MAX · HJ OPENS, 3 TO THE FLOP · HERO 133.75 BB AT THE DEAL", "hero": { "pos": "BB", "cards": ["9s", "5s"], "stack_bb": 66.77, "avatar": "lono_ghoul" }, "seats": [ { "pos": "UTG", "avatar": "ghoul_2919", "folded": true }, { "pos": "HJ", "avatar": "ghoul_4514", "stack_bb": 0, "villain": true }, { "pos": "CO", "avatar": "ghoul_5189", "folded": true }, { "pos": "BTN", "avatar": "ghoul_1780", "folded": true, "button": true }, { "pos": "SB", "avatar": "ghoul_226" } ], "streets": [ { "name": "preflop", "actions": [ {"pos": "UTG", "act": "fold"}, {"pos": "HJ", "act": "open", "to_bb": 2}, {"pos": "CO", "act": "fold"}, {"pos": "BTN", "act": "fold"}, {"pos": "SB", "act": "call"}, {"pos": "hero", "act": "call"} ], "pot_bb": 6 }, { "name": "flop", "board": ["7h", "As", "3s"], "actions": [ {"pos": "SB", "act": "check"}, {"pos": "hero", "act": "check"}, {"pos": "HJ", "act": "bet", "to_bb": 1.82}, {"pos": "SB", "act": "fold"}, {"pos": "hero", "act": "raise", "to_bb": 7.68}, {"pos": "HJ", "act": "call"} ], "pot_bb": 21.36 }, { "name": "turn", "board": ["7h", "As", "3s", "8s"], "actions": [ {"pos": "hero", "act": "bet", "to_bb": 18.23}, {"pos": "HJ", "act": "call"} ], "pot_bb": 57.82 }, { "name": "river", "board": ["7h", "As", "3s", "8s", "6d"], "actions": [ {"pos": "hero", "act": "bet", "to_bb": 39.07}, {"pos": "HJ", "act": "allin", "to_bb": 105.84} ] } ], "prompt": "9♠5♠ · YOU MADE A FLUSH — BUT EVERY FLUSH HERE PLAYS THE BOARD'S A♠ · HE JAMS OVER YOUR RIVER BET — CALL OR FOLD?", "pot_bb": 202.73, "options": [ {"id": "call", "label": "CALL"}, {"id": "fold", "label": "FOLD"} ], "correct": "fold", "option_feedback": { "call": "You check-raised the flop and barrelled into a made flush, so folding it feels absurd — and the 3.04:1 price is inviting. But the A♠ belongs to the board, which means it is in everybody's flush; what ranks yours is your own top spade, the nine. Three spades are out there, so he needs two of his own to have a flush at all — and 22 of the 28 two-spade combinations still available contain a ten or better and beat you, against just six that don't. Tyler's read of the raise: he probably value-raises every hand better than our hand. You need about 25% of that raising range to be hands you beat — and the only candidates are pure bluffs and the six worse flushes, which have no reason to raise at all.", "fold": "Correct — Tyler: 'he probably value-raises every hand better than our hand, so he needs to raise this river like 30% of the time before we call. Who the hell raises this river 30% of the time?' Read the board first: it is unpaired, so no full house exists, and every straight — including the 9-8-7-6-5 you also hold — loses to a flush, so the only hands that beat you are two-spade flushes containing a ten or better. Those are precisely the hands he raises for value. So the call needs bluffs, and lots of them: you are putting in your last 66.77 BB to win 202.73, about 25%, which means a quarter of everything he raises has to be hands you beat. That is a river-raise frequency almost nobody runs." }, "verdict_text": "He probably value-raises every hand better than our hand, so he needs to raise this river like 30% of the time before we call. Who the hell raises this river 30% of the time? That river-raise frequency is insane.", "aired_transcript": "This sounds kind of funny because essentially, he probably value-raises every hand better than our hand, right? So he needs to raise this river like 30 % of the time before we call. I mean, who the hell raises this river 30 % of the time? I mean, that river raise frequency is insane.", "aired_transcript_note": "Verbatim faster-whisper small.en re-ASR of the re-cut quiz/audio/fold-nine-high-flush-river-jam.m4a (19.8 s, beats s411 + s412 + s413, 2026-08-08). Two notes: (1) small.en hears the opening 'This hand's kind of funny' as 'This sounds kind of funny' — SER's own word-level transcription has \"hand's\" at 3636.26 with conf 0.65, the lowest-confidence word in the clip; nothing displayed depends on it, and verdict_text simply starts after that clause. (2) Each beat ends on Tyler's trailing 'I' and the next opens on 'mean', which reads as 'I mean' across the 0.35 s gap — the same join the previously live clip had. No cards, no positions, no bet sizes and no population magnitudes are aired; the only figure is Tyler's own required river-raise frequency for this price.", "principle": "When three of a suit are on the board and the ace of that suit is one of them, every flush plays the same shared ace — what ranks yours is your own top card in the suit. With the nine as that card, a river raise from an opponent who value-raises every better flush needs a bluff frequency almost nobody runs, so it is a fold.", "pct_provenance": { "30": { "role": "required_frequency", "referent": "Tyler's own stated threshold for THIS call at THIS node: the river-raise frequency the HJ would have to be running before hero's 9♠5♠ can call. Spoken verbatim: 'he needs to raise this river like 30 of the time before we call' (s411, t=3644.36-3647.70) and 'who the hell raises this river 30 of the time' (s412, t=3659.45-3662.23). It is a required frequency Tyler assigns to this one opponent, not a measured population or database figure — no pool/database magnitude is stated, displayed or aired anywhere in this quiz.", "source": "SER RiverCalls s411/s412 (spk='tyler', spk_confident=true); aired verbatim in the verdict clip (see aired_transcript)" }, "25": { "role": "price", "referent": "Breakeven equity of the graded call: hero puts in his last 66.77 BB to win the 202.73 BB out there (57.82 settled + his own 39.07 + the HJ's callable 105.84), i.e. 66.77 / 269.50 = 24.8%, displayed as about 25%. PokerTracker prints the identical price at this exact node: 'Odds: 3.04:1 - 25%'.", "source": "own frame f_3636 (t=3636, 'Pot: 389.32 BB' headline + 'Odds: 3.04:1 - 25%') + fixture arithmetic" } }, "size_claims": { "river_bet": { "street": "river", "actor": "hero", "bet_bb": 39.07, "pct_of_pot": 67.6, "note": "Documentation only — no size ADJECTIVE is used anywhere in the displayed text. Hero's river bet of 39.07 BB into the 57.82 BB settled pot is 67.6% of pot. Log verbatim 'Hero bets 39.07 BB'.", "source": "own frames f_3230/f_3636 (action log) + felt chip '39.07 BB'" }, "river_raise": { "street": "river", "actor": "HJ", "bet_bb": 105.84, "pct_of_pot": null, "note": "Documentation only. The HJ shoved to 292.42 BB on the felt; 105.84 is the callable portion (hero's 39.07 plus his last 66.77) and is what the fixture renders — see cosmetic_invention. No size adjective describes it; the displayed word is 'jams'.", "source": "own frame f_3636 (log 'raises to 292.42 BB', felt chips '292.42 BB', HJ plate '0 BB', hero plate '66.77 BB')" } }, "claims": [ { "text": "he probably value-raises every hand better than our hand", "basis": "Tyler, spoken at the node and aired verbatim: s411 (t=3635.58-3648.40) 'this hand's kind of funny because essentially he probably value raises every hand better than our hand right so he needs to raise this river like 30 of the time before we call'.", "source": "SER RiverCalls s411 (tyler, spk_confident); own frame f_3636 shows the replayer parked on this river node at t0" }, { "text": "the only hands that beat 9♠5♠ here are two-spade flushes containing a ten or better", "basis": "Board 7♥ A♠ 3♠ 8♠ 6♦ read off own frame f_3636 and the log ('Flop: [7h As 3s]', 'Turn: [8s]', 'River: [6d]'). It is unpaired, so no full house or quads exist. Hero's five spades (A♠8♠3♠ from the board + 9♠5♠) outrank the 9-8-7-6-5 straight he also holds, so straights lose. Only three spades are on the board, so a rival flush needs TWO spades of its own; the villain's flush is A♠ 8♠ 3♠ plus his two, and it beats hero's A-9-8-5-3 exactly when his higher spade is above the nine. A straight flush is impossible: every spade run that reaches five cards needs the 9♠ or 5♠, and hero holds both.", "source": "own frame f_3636 (board + hero cards) + card removal" }, { "text": "22 of the 28 two-spade combinations still available contain a ten or better and beat you, against just six that don't", "basis": "Spades unaccounted for after the board (A♠3♠8♠) and hero (9♠5♠): K♠ Q♠ J♠ T♠ 7♠ 6♠ 4♠ 2♠ = 8, so C(8,2) = 28 two-spade combinations. Those made only of {7♠6♠4♠2♠} are C(4,2) = 6 (7♠6♠, 7♠4♠, 7♠2♠, 6♠4♠, 6♠2♠, 4♠2♠) and lose to hero; the other 22 contain at least one of T♠/J♠/Q♠/K♠ and win. Reproducible card removal, not a database figure. Eric makes the same observation on tape at s371: 'there's no worse flushes, eight high flush maybe six high flush'.", "source": "own frame f_3636 (board + hero cards) + card removal; SER RiverCalls s371" }, { "text": "you are putting in your last 66.77 BB to win 202.73, about 25%", "basis": "Hero's plate reads '66.77 BB' and the HJ's '0 BB' at the node; the callable pot is 57.82 settled + hero's 39.07 + the HJ's matched 105.84 = 202.73, so 66.77 / (202.73 + 66.77) = 24.8%. PokerTracker's own readout at the same node is 'Odds: 3.04:1 - 25%' (202.73 / 66.77 = 3.04).", "source": "own frame f_3636 + fixture arithmetic (see pct_provenance['25'])" }, { "text": "you check-raised the flop and barrelled into a made flush", "basis": "Log verbatim: 'P4… bets 1.82 BB / P1… folds / Hero raises to 7.68 BB / P4… calls 5.86 BB / Turn: [8s] / Hero bets 18.23 BB / P4… calls 18.23 BB / River: [6d] / Hero bets 39.07 BB'. The turn 8♠ is the card that completes hero's flush.", "source": "own frames f_3230 / f_3636 (action log)" }, { "text": "the flop was three-way and the SB check-folded to the check-raise", "basis": "Log verbatim: 'P1… calls 1.50 BB / Hero calls 1 BB / Flop: [7h As 3s] / P1… checks / Hero checks / P4… bets 1.82 BB / P1… folds'. Eric flags the structure at s416: 'it's a check raise multi-way pod, it's pretty important information'.", "source": "own frames f_3225/f_3230 (log top) + f_3636; SER RiverCalls s416" } ], "unverifiable_claims": [ "'Who the hell raises this river 30% of the time / that river-raise frequency is insane' (s412-s413) is Tyler's qualitative judgement about how often opponents raise rivers. It is aired and displayed as his quote; no measured population or database figure backs it, and none is stated, displayed or aired.", "The replayer's seat readouts 'Hero - 98%' and 'P4-d8c1174c7... - 2%' are PokerTracker's equity display against UNKNOWN cards — the HJ's cards are face down and the hand never reaches showdown in the replay. They are NOT hero's equity against the raising range and are displayed nowhere in this fixture. (The same readout says 'Hero - 53%' at the flop node, f_3300.)", "The HoldEq window Tyler opens over the replayer during and after the verdict (own frames f_3600/f_3675: Board 'As8s7h6d3s', Range 1 'AA,77,33,A7s,A3s,AsKs,AsQs,AsKh,AsKc,AsKd…' vs Range 2 '9s5s' = 22.222% / 77.778%; at t=3600 Range 2 was 'Qs5s' = 9.524% / 90.476%) answers the Anki card he is typing at t=3600 — 'How good are we against his range on the river? / How often does he to raise here, before' — i.e. hero's river BET decision and a hypothetical Q♠5♠ holding. Different node, different question, and the range string changes while he edits it. Recorded for provenance; never displayed.", "Tyler's river-sizing discussion in the run-up (s379-s390: quarter-pot value-bet schemes, 'why are we doing this to ourselves') is about how hero should have bet the river, not about the raise he faces. Not displayed.", "Eric's 'it's a check-raise multiway pot, it's pretty important information' (s416) — a categorisation aside spoken after the verdict; the structural fact behind it IS rendered (three-way flop), but the quote is not displayed." ], "cosmetic_invention": [ "THE HJ'S RIVER RAISE IS RENDERED AT ITS CALLABLE SIZE. On the felt he raises to 292.42 BB with 0 BB behind, while hero has only 66.77 BB behind (own frame f_3636) — 186.58 BB of that shove can never be called and would be returned. The fixture renders the effective all-in, 105.84 BB (hero's 39.07 river bet plus his last 66.77), because that is the amount the hand was actually played for: it reproduces the price PokerTracker itself prints at this node ('Odds: 3.04:1 - 25%') and the callable pot its odds engine uses (202.74). Rendering the gross 292.42 would put a 389.32 BB pot on the felt and price the call at 253.35 to win 389.31 (39.4%) — a price nobody at that table ever faced. Consequence of the cap, stated plainly: the HJ's rendered line adds up as if he had started with hero's 133.75 BB; his real start was 320.33 BB and he covered hero by 186.58.", "Ghoul avatars on all six seats (lono_ghoul for hero) — the replayer shows anonymised hashes (P1-e6bbfbf1…, P3-10916937…, P4-d8c1174c…, P5-922692657…, P6-682990ad…), not faces.", "Seat LABELS UTG / HJ / CO / BTN / SB. The replayer prints hashes, not position names; the labels are derived from the log's own dealer line ('Seat 6 (P6-682990ad…) is the dealer') plus the blind posts (P1 = SB, Hero = BB), which fixes P3 = UTG, P4 = HJ, P5 = CO. The derivation is sourced, only the naming is cosmetic.", "The villain seat is deliberately unnamed. Tyler states no read on this opponent anywhere in the window, so the retired fixture's 'REG' tag is removed rather than re-sourced.", "PT cent-rounding, max drift 0.01 BB: the settled pot renders 57.82 (6 + 15.36 + 36.46 from the log) where the felt chip prints '57.83 BB', and the callable pot renders 202.73 where PT's is 202.74. Flop amounts follow the LOG (1.82 bet, raise to 7.68) rather than the felt chip labels ('1.83', '7.69'), because the log is internally consistent (1.82 + the logged 'calls 5.86 BB' = 7.68).", "No suit, position, card or stack was invented anywhere else — see sourced_not_invented." ], "sourced_not_invented": [ "Hero's exact cards 9♠5♠ — own frame f_3636, felt zoom z_felt_3636: two grey cards with spade pips, ranks 9 and 5, on the 4-colour deck this replayer uses (red = hearts, blue = diamonds, grey/black = spades, green = clubs).", "The full board with suits, 7♥ A♠ 3♠ / 8♠ / 6♦ — log verbatim 'Flop: [7h As 3s]', 'Turn: [8s]', 'River: [6d]' (own frames f_3225/f_3230/f_3636), corroborated by the 4-colour felt in z_felt_3636. LOAD-BEARING: the whole lesson is that exactly three spades are out and one of them is the ace.", "Every rendered bet size — 2 (HJ open), 1.82 (HJ flop bet), 7.68 (hero flop raise), 18.23 (hero turn bet), 39.07 (hero river bet) — action log verbatim, own frames f_3230 / f_3636.", "Positions: BTN = Seat 6 (log 'Seat 6 (P6-682990ad…) is the dealer'), SB = P1 (log 'posts the small blind 0.50 BB'), BB = Hero (log 'Hero posts the big blind 1 BB'), so the opener P4 is the HJ. LOAD-BEARING: the old fixture had this opponent in the CO.", "The three-way flop and the SB's flop check-fold — log verbatim. LOAD-BEARING: it is why the pot is the size it is, and it is the fact Eric singles out at s416.", "Hero's 66.77 BB behind and the HJ's 0 BB at the decision — seat plates, own frame f_3636. LOAD-BEARING: hero's stack is what caps the callable raise and sets the price.", "The price 3.04:1 / 25% — PokerTracker's own readout at the node, own frame f_3636. LOAD-BEARING: it is the number Tyler's '30% river-raise frequency' threshold is answering.", "Game type and hand identity '$2 NL FAST (6 max)', hand #4458334810, 2023/04/26 — title bar and log header, own frames f_3636 / f_3230.", "Hero's flop stack 124.07 BB and the HJ's 316.51 BB at the flop node — own frame f_3300; used to derive and cross-check both starting stacks." ], "concept": "low-flush-relative-strength", "concepts_secondary": [], "topic": "river-defense", "difficulty": 5, "decision": { "pot_settled_bb": 57.82, "bets": { "hero": 39.07, "HJ": 105.84 } } }