{ "schema_version": "1.3", "rebuild": "v2", "provenance_class": "dealt_hand", "table_size": 3, "id": "qt-protection-lead-paired-turn", "format": "multi_option", "free": true, "updated_at": "2026-08-08T00:00:00Z", "source": { "session_id": "ExploitativeSmallPotPoker", "verdict_audio": "quiz/audio/qt-protection-lead-paired-turn.m4a", "verdict_beats": "s523 + s539 + s527, the beats already registered in tools/cut_quiz_audio.py (clip NOT re-cut this rebuild). Echo-checked 2026-08-08 (faster-whisper small.en): re-ASR of the shipped 11.82s m4a returns exactly those three beats \u2014 see aired_transcript. Nothing in the clip contradicts the rebuilt hand: 'queen-ten' is hero's holding, 'the board pairing' is the 6s turn pairing the 6c, 'the small bet' is the 3.76 BB lead into 12 BB (31.3%), and no card, position, pot, size, percentage, population figure or operator name is aired. At the first verdict word (s523 t0=1933.92) the typed hand record for THIS hand is on screen \u2014 own frame t=1934.", "lesson_slug": null, "master_slug": "ExploitativeSmallPotPoker_master", "master_t": 1678.52, "src_t": 1933.9, "ser_sentences": [ 519, 520, 521, 523, 524, 526, 527, 535, 538, 539, 540, 543, 544, 545, 546, 549, 550, 551, 552, 553, 554, 560, 563 ], "hand_record": { "record_type": "TYPED HAND-HISTORY NOTE (hand-converter output pasted into Eric's Obsidian vault, note '5.20.2025'), read off the screen share. Frames are READ-ONLY, never aired. This is NOT a solver/equity tool: no range grid, no EV table, no frequency report is anywhere in frame at any of the five sampled anchors \u2014 the numbers are irregular dealt-hand numbers (118.73 / 103.25 / 3.76 / 11.4) and the pot is raked (12 BB pot returns 11.4 BB).", "note_verbatim": [ "Hero (BB): 96 BB", "UTG (UTG): 118.73 BB", "BTN (BTN): 103.25 BB", "Hero posts BB 1 BB", "Dealt to Hero: Q[blue tile = diamonds] T[dark tile = spades]", "UTG raises to 3 BB, fold, Hero calls 2 BB", "Flop (6 BB, 2 players): 6[green = clubs] T[red = hearts] 5[blue = diamonds]", "Hero checks, UTG bets 3 BB, Hero calls 3 BB", "Turn (12 BB, 2 players): 6[dark = spades]", "Hero bets 3.76 BB, fold", "Hero wins 11.4 BB" ], "note_caption": "CAPTIONS FOLLOW THEIR HAND in this file. This hand's own captions are the three lines directly BELOW the block, between it and the next block (own frame t=2088, scratchpad dh3/qt-protection-lead-paired-turn/f/p2088.jpg): 'vs the headsup monkey I think' / 'solver hates this line lol' / 'donk lead makes more sense when lowest card pairs turn'. The third line is dispositive for the direction: it is Eric's written version of what he says aloud about THIS hand at s535 ('Isn't it better if it's a five of spades, not the six of spades?'), and it cannot describe the block below it, whose board is 9s4s6h with a 3c turn \u2014 nothing pairs there (own frame t=2095). The line ABOVE this block, 'defend at mdf? villain was 80vpip', is therefore the PREVIOUS hand's caption (the BTN hand Tyler rules on at s515-s518); Eric reads it aloud at s519-s520 (t=1924.3) as the closing beat of that hand, nine seconds before Tyler opens this one at s523. None of these caption lines is displayed or aired. [Stage-5 correction 2026-08-08: the pre-Stage-5 fixture asserted the opposite (caption-ABOVE) and so attached the 80-VPIP HUD read to THIS hand's opener. Its supporting argument was also wrong \u2014 the previous hand's hero being the preflop raiser does not stop him defending at MDF postflop, which is exactly the call Tyler approves at s515 ('I think the call's OK here, Eric') while explaining that the in-position player gets over-bluffed.]", "stake": "NOT ON SCREEN. No currency figure, no table title, no client window appears at any sampled anchor \u2014 the record is BB-native end to end.", "stake_source": "n/a \u2014 and none is needed: no $/chips-to-bb conversion was performed anywhere in this fixture. Every figure in the felt (1 / 3 / 3 / 3.76 / 6 / 12 / 96 / 118.73 / 103.25) is copied verbatim from a line that already reads 'BB' on screen (own frames t=1926, 1934, 1947, 1980, 2075).", "units_note": "The converter prints every amount in big blinds ('Hero posts BB 1 BB', 'UTG raises to 3 BB', 'Flop (6 BB, 2 players)', 'Hero bets 3.76 BB'). The fixture copies those BB figures unchanged.", "video_t": [ 1924, 2090 ], "frames": [ "t=1926 (scratchpad dh3/qt-protection-lead-paired-turn/f_1926.jpg + z_1926.png) \u2014 mid-scroll: end of the PREVIOUS note ('BB wins 10.93 BB'), Eric's caption 'defend at mdf? villain was 80vpip', then the top of THIS note with all three stack lines and 'Hero posts BB 1 BB'.", "t=1934 (f_1934.jpg + note_1934_zoom.png + note_cards_zoom.png) \u2014 ON SCREEN AT THE FIRST VERDICT WORD (s523 t0=1933.92): the complete note, every line transcribed in note_verbatim.", "t=1947 (f_1947.jpg) \u2014 same note, unchanged, at s527 ('your bet's just designed to get him to fold').", "t=1980 (f_1980.jpg + z_1980.png + cards_hero.png / cards_flop.png / cards_turn.png) \u2014 same note at the mechanism beat s539; card tiles sampled at 10x for the 4-colour read.", "t=2075 (f_2075.jpg + z_2075.png) \u2014 same note still on screen at s560 ('it's all about the ratios of queen ten to 6x'), i.e. the whole discussion s523-s563 runs over this one record.", "t=1870 / t=1905 (z_1870.png, z_1905.png) \u2014 the PREVIOUS note in the same file, which DOES print 'SB posts SB 0.5 BB, BB posts BB 1 BB'. This is the control that makes the missing SB line in this hand's note dispositive rather than an omission.", "STAGE-5 INDEPENDENT RE-DERIVATION 2026-08-08 (own frames, scratchpad dh3/qt-protection-lead-paired-turn/f/): t=1900 (previous note WITH its 'SB posts SB 0.5 BB' line and 5.5 BB flop pot \u2014 the control), t=1925 (mid-scroll: previous note's tail 'BB wins 10.93 BB', the caption 'defend at mdf? villain was 80vpip', then this block's top border and first line 'Hero (BB): 96 BB'), t=1934 / t=1940 (whole block, 4x zoom \u2014 every line in note_verbatim, card tiles legible), t=1995 and t=2075 (same block unchanged through the mechanism beats), t=2088 (this hand's three caption lines and the start of the NEXT block), t=2095 ('Ignition - $2 NL - Holdem' 5s8s BB-vs-SB block, board 9s4s6h / 3c). Every load-bearing fact in this fixture \u2014 BB-native scale, 3 dealt with no SB post, Qd Ts, 6c Th 5d / 6s in that order, 3 / 2 / 3 / 3 / 3.76, pots 6 and 12 \u2014 was re-read off those frames before the fixture was opened, and agrees." ], "table_structure": "THREE PLAYERS DEALT, NO SMALL BLIND POSTED. The record lists exactly three stacks (Hero BB, UTG, BTN), prints exactly one blind post ('Hero posts BB 1 BB' \u2014 no SB line, unlike the neighbouring note), and shows exactly one preflop fold. The arithmetic closes it: UTG 3 + Hero 3 = the printed 'Flop (6 BB, 2 players)'; a folded small blind would have made it 6.5. Preflop order UTG -> BTN -> BB and postflop order BB -> UTG follow from the seat names, so the small-blind SEAT sat empty (dead small blind); whether further seats existed at the table is not on screen. No straddle, no ante anywhere in the record.", "stacks_bb": { "hero_start": 96, "hero_at_decision": 90, "utg_start": 118.73, "utg_at_decision": 112.73, "btn_start": 103.25, "effective": "96 \u2014 hero is the short stack of the two live players; 96 - 6 invested = 90 behind at the turn decision.", "note": "Only hero's stack is put on the felt. The validator's depth check (I7) assumes one declared depth for every seat carrying stack_bb, and these stacks are unequal (96 / 118.73 / 103.25), so the villain's stack is recorded here rather than displayed." }, "result": "Hero bets 3.76 BB on the turn, UTG folds, Hero wins 11.4 BB (note verbatim). Recorded for provenance; not displayed, not graded, and NOT aired \u2014 s528 ('He did') is deliberately not in the clip." } }, "context": "CASH GAME \u00b7 3 DEALT, NO SMALL BLIND \u00b7 96 BB EFFECTIVE \u00b7 BB vs UTG", "hero": { "pos": "BB", "cards": [ "Qd", "Ts" ], "stack_bb": 90, "avatar": "lono_ghoul" }, "seats": [ { "pos": "UTG", "avatar": "ghoul_108", "villain": true }, { "pos": "BTN", "avatar": "ghoul_612", "button": true, "folded": true } ], "streets": [ { "name": "preflop", "actions": [ { "pos": "UTG", "act": "open", "to_bb": 3 }, { "pos": "BTN", "act": "fold" }, { "pos": "hero", "act": "call" } ], "pot_bb": 6 }, { "name": "flop", "board": [ "6c", "Th", "5d" ], "actions": [ { "pos": "hero", "act": "check" }, { "pos": "UTG", "act": "bet", "to_bb": 3 }, { "pos": "hero", "act": "call" } ], "pot_bb": 12 }, { "name": "turn", "board": [ "6c", "Th", "5d", "6s" ], "actions": [] } ], "prompt": "Q\u2666T\u2660 \u00b7 TOP PAIR \u00b7 THE 6 PAIRS THE TURN AND YOU'RE FIRST TO ACT \u2014 CHECK, LEAD SMALL, OR LEAD POT?", "pot_bb": 12, "options": [ { "id": "check", "label": "CHECK" }, { "id": "lead", "label": "LEAD 3.76 \u2014 31% POT" }, { "id": "leadpot", "label": "LEAD 12 \u2014 POT" } ], "correct": "lead", "option_feedback": { "check": "Tyler's own default lives here \u2014 'mathematically it's a really complicated idea and it's not that easy to get your head wrapped around, which is why I'm usually, like, just check.' But checking hands back both jobs at once: he gets to check behind and realise his equity for free, and when he does bet he gets to bluff optimally. The 3.76 BB lead is what takes those two things away.", "lead": "Correct \u2014 'the small bet here is just designed to keep him from bluffing optimally against your hand and giving him equity.' Equity denial plus bluff denial, both bought for 3.76 BB into 12. And the paired board is what makes it safe rather than scary: your defending range credibly holds sixes, so 'he just can't go ham with a pair of jacks.'", "leadpot": "Too big. Betting the whole 12 BB is how you'd want to play the jacks, not the top pair \u2014 Tyler runs it in reverse: 'if you knew you had queen ten, how would you want to play a pair of jacks here?\u2026 And bet big on the turn, right?' The small lead does the same protection work and caps the damage: 'you only lose four when he has jacks, as opposed to losing ten.'" }, "verdict_text": "I'm OK with queen-ten. The board pairing \u2014 the small bet here is just designed to keep him from bluffing optimally against your hand and giving him equity. Your bet's designed to get him to fold.", "aired_transcript": "Oh, I'm okay, queen-ten. Like, the board pairing, the small bet here, is just designed to keep him from bluffing optimally against your hand and giving him equity. I mean, your bet's just designed to get him to fold.", "aired_transcript_note": "Verbatim small.en re-ASR of quiz/audio/qt-protection-lead-paired-turn.m4a (11.82s; beats s523 + s539 + s527 per tools/cut_quiz_audio.py). All three beats are WHOLE-sentence beats with no word trims, so s523's leading 'Oh,' IS aired. Stage-5 re-ASR 2026-08-08 (faster-whisper small.en, cuda) returned exactly: 'Oh, I'm okay, Queen Tan.' / 'Like, the board pairing, the small bet here, is just designed to keep him from bluffing optimally against your hand and giving him equity.' / 'I mean, your bet's just designed to get him to fold.' \u2014 'Queen Tan' is small.en's mis-render of 'queen ten' (SER words w4928-4932 read 'queen ten'; the session-level ASR of the same audio also reads 'queen ten'). [Stage-5 correction: the pre-Stage-5 fixture dropped the 'Oh,' from this field and claimed the cut trimmed it \u2014 it does not.] No cards beyond 'queen ten' (hero's own hand), no positions, no board cards, no sizes, no percentages, no population figures and no result talk are aired \u2014 s528 'He did' is not in the clip. Beat 3 (s527, t=1946.7) is aired AFTER beat 2 (s539, t=1980.0), i.e. out of source order; both are Tyler about this same turn bet, so the splice states nothing the tape does not.", "principle": "When the board pairs a middle card and you hold a vulnerable top pair out of position, a small lead does two jobs a check cannot: it denies the rest of his range its free equity and it takes away his ability to bluff optimally \u2014 and because your defending range credibly holds the paired rank, he cannot raise you off it.", "pct_provenance": { "31": { "role": "size", "referent": "The graded option's own price: hero's turn lead of 3.76 BB into the 12 BB turn pot = 31.3% of pot, displayed as '31% POT' on the option label. It is the size actually bet at THIS node \u2014 not a figure carried in from another street, node or sizing.", "source": "own frames f_1934 / f_1980 / f_2075 \u2014 note verbatim 'Turn (12 BB, 2 players)' and 'Hero bets 3.76 BB'; 3.76 / 12 = 0.3133" } }, "size_claims": { "small": { "street": "turn", "actor": "hero", "bet_bb": 3.76, "pct_of_pot": 31.3, "note": "Every 'small' in the prompt, the options, the feedback, the verdict text and the aired clip refers to one bet: hero's 3.76 BB turn lead into the 12 BB pot = 31.3%. The adjective is Tyler's own \u2014 s539 'the small bet here', s553 'betting small here with queen ten'.", "source": "own frames f_1934 / f_1980 / f_2075 (note verbatim 'Turn (12 BB, 2 players): 6[spades] / Hero bets 3.76 BB, fold'); SER ExploitativeSmallPotPoker s539, s553" } }, "claims": [ { "text": "preflop pot at the flop = 6 BB, with no small blind in it", "lhs": "UTG 3 + Hero 3 (1 posted + 2 called) + 0 dead small blind", "op": "=", "rhs": "6 BB", "source": "note verbatim 'Hero posts BB 1 BB' (the ONLY post line) + 'UTG raises to 3 BB, fold, Hero calls 2 BB' + 'Flop (6 BB, 2 players)'. Control: the note directly above in the same file prints 'SB posts SB 0.5 BB, BB posts BB 1 BB' and a 5.5 BB flop pot (own frames t=1870 / t=1905), so the missing SB line here is real, not an omission. A folded SB would have made this pot 6.5 \u2014 the number the pre-audit fixture carried." }, { "text": "pot at the decision = 12 BB", "lhs": "6 + UTG 3 + Hero 3", "op": "=", "rhs": "12 BB", "source": "note verbatim 'Hero checks, UTG bets 3 BB, Hero calls 3 BB' + 'Turn (12 BB, 2 players)' (own frames f_1934 / f_1980)" }, { "text": "the villain's flop bet is half pot", "lhs": "3 BB into the 6 BB flop pot", "op": "=", "rhs": "50% of pot", "source": "note verbatim 'Flop (6 BB, 2 players)' + 'UTG bets 3 BB' (own frame f_1980). Not described with a size adjective anywhere in the displayed text; recorded so the felt's numbers are all accounted for." }, { "text": "the graded lead is 3.76 BB = 31% of pot", "lhs": "3.76 / 12", "op": "=", "rhs": "0.3133", "source": "note verbatim 'Hero bets 3.76 BB' into 'Turn (12 BB, 2 players)' (own frames f_1934 / f_1980 / f_2075)" }, { "text": "stacks: 96 BB effective, hero 90 BB behind at the decision", "lhs": "hero 96 \u2212 6 invested / UTG 118.73 \u2212 6 invested", "op": "=", "rhs": "90 BB / 112.73 BB behind; effective = 96 (hero is the shorter of the two live stacks)", "source": "note stack list verbatim 'Hero (BB): 96 BB / UTG (UTG): 118.73 BB / BTN (BTN): 103.25 BB' (own frames f_1926 / f_1934 / f_1980)" }, { "text": "you only lose four when he has jacks, as opposed to losing ten", "lhs": "hero's own risk on the turn: the 3.76 BB lead", "op": "vs", "rhs": "the ~ten Tyler says hero loses on this same turn when villain is left free to raise / bet big with jacks. Both figures are Tyler's own round numbers, both spoken about THIS node (turn, 12 BB pot), neither imported from another street, node or sizing. Tyler attaches no size and no unit to either number; the fixture quotes them as speech and never renders them as any option's price \u2014 the third option's price is the pot itself, 12 BB, read off the note.", "source": "SER s550-s554 verbatim: 'if you knew you had queen ten, how would you want to play a pair of jacks here?' / Eric: 'You'd want to raise.' / 'And bet big on the turn, right?' / 'what you're doing here by betting small here with queen ten is you're preventing him from doing those things' / 'So, you only lose four when he has jacks, as opposed to losing ten when he has jacks.' (all words conf \u2265 0.85 on the numerals). The quiz's third option is the exact pot, 12 BB; Tyler's 'ten' is his own round figure for that big-bet route and is quoted as speech, never rendered as the option's price." }, { "text": "the turn 6 pairs the MIDDLE card of the flop, not the bottom card", "lhs": "flop ranks T (high), 6 (middle), 5 (low); turn 6", "op": "\u21d2", "rhs": "the middle card pairs; the pre-audit prompt's 'THE BOTTOM CARD PAIRS THE TURN' was wrong", "source": "note tiles (own frame f_1980, cards_flop.png / cards_turn.png at 10x) + Eric on tape drawing exactly this contrast, s535 t=1965.09: 'Isn't it better if it's a five of spades, not the six of spades?'" }, { "text": "hero has TOP PAIR", "lhs": "Q\u2666T\u2660 on 6\u2663 T\u2665 5\u2666 6\u2660", "op": "=", "rhs": "a pair of tens \u2014 the highest board rank \u2014 with a queen kicker", "source": "note verbatim 'Dealt to Hero: Q T' + 'Flop \u2026 6 T 5' + 'Turn \u2026 6'; spoken s523 'I'm OK, queen ten', s553 'betting small here with queen ten'" }, { "text": "the mechanism that makes the small lead safe: hero's range credibly holds sixes", "lhs": "BB defending range on 6-T-5-6", "op": "\u21d2", "rhs": "the villain cannot raise big with an overpair", "source": "SER s549 'because you have a thread [threat] of a six in this range, so he just can't go ham with a pair of jacks' + s560 'It's all about the ratios of queen ten to 6x'" }, { "text": "Tyler endorses the bet that was made", "lhs": "s523 'I'm OK, queen ten' + s524 'I think this is fine' + s539 (mechanism) + s563 'Truthfully, I'm fine with it'", "op": "\u21d2", "rhs": "graded answer = the small lead", "source": "SER ExploitativeSmallPotPoker s523, s524, s539, s544-s545, s563; the endorsement is delivered against Eric's own doubt (s526 'this is a bad bet, isn't it, on the turn?', s532 'I don't know if I like this')" } ], "unverifiable_claims": [ "s549's 'a thread of a six in this range' is an ASR garble (word conf 0.38 on 'thread'); read as 'threat of a six', which is the only sense that fits s560's 'ratios of queen ten to 6x'. The feedback quotes only the unambiguous half of the sentence ('he just can't go ham with a pair of jacks', all words conf \u2265 0.90).", "s562 'I mean, that's why I don't decide doing it' is garbled and is NOT used anywhere in the fixture; the adjacent s546 ('which is why I'm usually, like, just check', all words conf \u2265 0.71) and s563 ('Truthfully, I'm fine with it', conf \u2265 0.97) are clean and carry the same nuance, and both are represented \u2014 s546 inside option_feedback.check.", "'CASH GAME' in the context line is an inference, not a pixel: the record shows 0.5/1-style blinds with no ante and 96-119 BB stacks, and the neighbouring notes in the same file are the same format, but no stake, currency figure, site name or table title appears at any sampled anchor.", "Eric's caption 'villain was 80vpip' and his spoken s520 are a HUD read on the PREVIOUS hand's villain, not this one's (see hand_record.note_caption \u2014 captions follow their hand). Nothing is known about THIS hand's opener: no villain tag is rendered, none is aired, and Tyler never characterises him \u2014 the opponent-read rule requires reads Tyler states on tape, and his only words about the opponent here are the hypothetical 'a pair of jacks'. Eric's own captions on this hand ('vs the headsup monkey I think', 'solver hates this line lol') are Eric's guess and Eric's aside, not Tyler reads, and are likewise not displayed and not aired.", "Whether more than three seats existed at the table is not on screen. The record names an SB position that nobody occupies; the felt therefore renders the three dealt players and the context line declares the missing small blind.", "The hand's result (villain folds, Hero wins 11.4 BB) is provenance only \u2014 recorded in hand_record, never displayed and never aired." ], "cosmetic_invention": [ "Ghoul avatars, art only: hero lono_ghoul, UTG ghoul_108, BTN ghoul_612. The source is a text note; it has no table art at all.", "The empty small-blind SEAT is not drawn. The felt renders the three dealt players (hero BB, UTG, BTN) and the missing small blind is declared in the context line instead \u2014 the engine has no way to draw an unoccupied, unposted blind seat.", "Nothing else. No suit, position, size, stack or pot in this fixture is cosmetic \u2014 see sourced_not_invented." ], "sourced_not_invented": [ "Hero's exact cards Q\u2666T\u2660 \u2014 note tiles at 10x (own frame f_1980, cards_hero.png): Q on a BLUE tile = diamonds, T on a DARK tile = spades, in the 4-colour scheme the same note uses for the board. Spoken corroboration: s523 'queen ten'.", "The full board with suits 6\u2663 T\u2665 5\u2666 / 6\u2660 \u2014 note tiles at 10x (cards_flop.png: green 6 = clubs, red T = hearts, blue 5 = diamonds; cards_turn.png: dark 6 = spades). Spoken corroboration for the turn card: Eric s535 'the six of spades'.", "The turn pairs the MIDDLE flop card (6), not the bottom (5) \u2014 pixels plus Eric's explicit contrast at s535. The pre-audit prompt said 'THE BOTTOM CARD PAIRS THE TURN'; that is corrected.", "Every amount: preflop 3 / call 2, flop 3 / call 3, turn lead 3.76, pots 6 and 12, all read verbatim off BB-labelled lines. No conversion of any kind was performed.", "All three stacks 96 / 118.73 / 103.25 BB, and therefore the 96 BB effective depth.", "The 3-dealt, no-small-blind structure and the single preflop fold (BTN) \u2014 sourced from the stack list, the single post line, the one 'fold', and the printed 6 BB flop pot, with the neighbouring note's 'SB posts SB 0.5 BB' as the control.", "Hero is the BB, is the caller, and acts FIRST on the turn (UTG has position on him) \u2014 from the note's position labels and its own action order ('Hero checks, UTG bets 3 BB').", "Tyler's mechanism and every quoted phrase in the feedback, verdict and principle \u2014 SER s539, s544-s546, s549-s554, s560, s563 (tyler, spk_confident)." ], "concept": "small-lead-denies-equity", "concepts_secondary": [], "topic": "protection-betting", "difficulty": 4, "decision": { "pot_settled_bb": 12, "bets": {} } }