The Record

Fatigue mechanism (England 3-day turnaround post-90'+12 collapse) reactivates only if the semifinal XIs actually take the pitch; rotation neutralizes it. Trigger status 7/17: predicted XIs retain France 3/11, England 6/11 — NOT met pending official sheets ~20:00Z Sat.

Line at call: Trigger price at log: Polymarket 49.7¢ vs DK devig 45.6% (2026-07-16 ~23:00Z); by 7/18 04:35Z DK had moved to 49.7% and PM to 51.2% · Falsifier: England making zero substitutions before 65' falsifies the degradation mechanism regardless of result

Result: Official XIs: France retained 4/11 semifinal starters (Maignan, Rabiot, Olise, Mbappé), England 4/11 (Guéhi, Spence, Rice, Rogers) — trigger threshold ≥8 BOTH sides not met; card archives unfired per protocol. Match: England 6–4 France in regulation (Saka hat trick incl. 87' pen; Mbappé 48', 66'). Pre-registered falsifier ('zero England subs before 65 minutes') did NOT fire — Watkins entered at 46'. The venue story graded anyway: France 90' closed −125 (no-vig 52.6%, FanDuel 1:36pm ET) and LOST — every desk number (48.1/48.5/49.5) was below the close on the right side.

2026-07-18 — Rotation base rate beat the fatigue mechanism, exactly as the steelman argued — but the market kept moving France-ward into a 4-goal French loss. Sometimes the pass is the win.

Desk prices Mahomes W1 start at 27.0% (turn-1 22.5% +4.5pp conceded on the positional base-rate gap): reported ~9.0-month ACL/LCL return vs sourced RTP means 11.6–13.6 months and 62.7% multi-ligament RTP.

Line at call: At log: PM 'Chiefs Week 1 starting QB' bid 31 / ask 88 (untradeable 57¢ spread), 2026-07-18 ~04:33Z · Falsifier: Live-fire 11-on-11 snaps in Preseason Week 1 (Aug 13–15) without a mechanically restricted throwing motion

Result: Withdrawn by the desk 7/28: the trigger's price gate was met (firm bid 0.78 >= 0.35) but the liquidity gate was not — only $18 traded all week against an 18c spread ('a ghost town with newly painted signs'), and the 7/24 full medical clearance struck the card's biological-timeline mechanism directly. Desk probability revised 27% -> 62%; the number continues to grade on The Calls.

Desk prices LAR Under 11.5 at 56.5%: an 11.5 ceiling in a parity-engineered league prices the team for perfection; the card fires only if the public overextends the Over past −135.

Line at call: At log: LAR win total 11.5 (moved from 10.5 since March, book+timestamps in pack) · Falsifier: A 6-0 or better LAR start before the trigger fires

2026-08-06 — TRIGGER FIRED: Over 11.5 at -140 at BetMGM AND Caesars (board stamped 8/5 8:11pm ET; BetRivers -134; DK moved its number to 10.5 — key-number corroboration). Card armed as logged 7/17: desk prices LAR Under 11.5 at 56.5%. No book currently prints Under 11.5 juice, so no executable entry to quote; the call grades on the season.

2026-09-28 — DATA POINTS 2-3: W 28-6 vs NYG (Mon 9/21, SoFi — discounted: Dart lost on opening drive, Winston 11/27-111-INT); L 30-26 at DEN (Sun 9/27 — 16-0 halftime lead surrendered, Stafford 55 att/2 INT incl. 66-yd pick-6, Nix 1-yd keeper 0:47 left). LAR 1-2. Desk number REVISED 56.5 -> 67.2 (mechanism: negative game-script divergence + Garrett IR ~wk8 + Nacua hip; from 1-2 the over needs 11-3). Steelman walk-back to 63.8 on +750 SB print CHALLENGED (data predated turn 1, double-counted) and retracted. Entry stands +130 (printed 9/1). Next: @PHI Sun 10/4, BUF Mon 10/12 — pre-stated watch: Stafford attempt count, third-down D without Garrett; 0-2 -> number crosses 85.

2026-10-14 — STRESS TEST GRADED (pre-stated 9/28): W 24-20 at PHI (Sun 10/4) + L 34-37 OT vs BUF (Mon 10/12) = SPLIT -> rule 'split = on schedule at 67.2' HELD after editor challenge reversed an 84.5 override and a 74.3 fold. LAR 2-3; over now needs 10-2. Observables: Stafford attempts 51/54 (fired both); 3rd-down D 0/12 & 2/7 = thesis REFUTED on that instrument (failure was early-down run D: BUF 234 yds) -> instrument RETIRED, replaced by first-and-10 rush success >=4 yds (floor 15, bar 55%; baseline BUF 67%, PHI 53%); pre-committed steps +3.0 (either of ARI 10/18, LV 10/25) / +6.5 (both). New-and-outside-the-hinge steps: LT Jackson 2-game suspension +2.1; kicker swap +0.4. NUMBER: 69.7. Next grade after LV.

2026-10-23 — GAME 6: L 16-18 vs ARI (Sun 10/18; ARI closed +13.5, largest upset of the season; Ryland 6/6 incl. 43 yd at 0:00 after a 21-yd completion + 15-yd unnecessary-roughness penalty on Wallace put ARI at the LA 29 with 0:41). LAR 2-4; over needs 10-1. INSTRUMENT (pre-committed 10/14): ARI first-and-10 rushes 8/17 >=4 yds = 47% < 55% bar -> DID NOT FIRE (recomputed 3 ways); +3.0 still live vs LV; +6.5 gone. Attempt observable 33 -> did not fire. NUMBER: 69.7 -> 95.9 = arithmetic base 93.9 (flat p=.65 per game, named as a conservative convention; analyst's own game-by-game table implied 99.5 and was struck) + bounded steps: Adams (hamstring, 6-10 wks, IR) +1.5 to 11/25; Ford (season) + Turner (IR) +1.2 to 11/15; Landman (concussion) +0.2 to after LV; kicker +0.4 voided -> +0.3 (Grupe torn ACL on kickoff tackle 9 min into his debut; McManus = 3rd kicker in 3 weeks) less 0.15 for McManus's record (new in steelman, accepted); JACKSON CORRECTION: suspension cut to 1 game on appeal, announced Wed 10/14 7:38pm ET BEFORE our 10/14 issue published; we printed 2 games/+2.1 -> refunded half, -1.05, owned in print. Steelman -0.10 for a 'December Garrett timeline' STRUCK (no timetable exists; 5th documented fold). Conditional +0.6 if Nacua (calf, Q) inactive. PRE-COMPUTED: LV win (3-4) -> 93.35 before instrument; LV loss (2-5) -> 98.65 -> card to maintenance (tripwire 98.0). Market: no book lists 11.5; DK relisted 9.5 (O+108/U-136, via ESPN 10/20); Polymarket >11.5 wins 11.5c (thin); FPI 8.4 wins; Bovada SB +1100.

2026-11-03 — GAMES 7-8: W 35-21 at LV (Sun 10/25) + W 24-9 vs LAC (Sun 11/1). LAR 4-4; over needs 8-1. INSTRUMENT (pre-committed 10/14, window ARI/LV): LV 7 first-and-10 rushes, 3 >=4 yds (42.9%) - BELOW THE 15-ATTEMPT FLOOR, could not fire under any reading (ESPN summary, ESPN core feed, nflverse identical; LAC for the record 4/11 = 36.4%); window CLOSED, +3.0/+6.5 dead by expiry. Attempt observable 29/25 did not fire. NUMBER: 95.9 -> LV pre-computation 93.35 -> LAC re-derived under the named convention: base 87.9 (P(>=8 of 9 at .65) = 12.11%) + Adams +1.5 (expiry 11/25 -> 11/28) + Ford/Turner +1.2 (to 11/15) - Jackson 1.05 (permanent); Landman +0.2 expired; kicker step retired (McManus 9/9) = 89.55, printed 89.6. Written at zero: Garrett -0.5 at activation (analyst wrote -0.6; p=.655 on the final seven gives -0.49 and the arithmetic prints; window opens Wed 11/4); Stafford CONDITIONAL +5.5 if inactive at WSH (ESPN Q right arm; OC Register ice on right elbow; McVay 'some soreness there'; no diagnosis; p=.25 for that game) -> 95.05. Steelman held. PRE-COMPUTED WSH (Sun 11/8 1:00pm ET; ESPN/DK LAR -5.5, 47.5): win -> 84.75; loss -> 98.45 -> MAINTENANCE (98.0 tripwire; no weekly repricing thereafter; graded season end). No instrument through the bye. Market: Polymarket >11.5 16.5c midpoint (10/23 book, $952); no book lists the total (Bovada carries 24 teams, not LAR; DK unreachable, last 9.5 10/20); FPI 9.3 wins, 43.4% playoffs, remaining SOS 1st; Bovada SB +1200.

Cousins-veteran-floor: scheme and target concentration support the price. Survived a formal thesis review 8/7 on schematic verification.

Line at call: ADP 37.8 entry (FantasyFootballCalculator PPR 12-team, 2026-07-28) · Falsifier: Pre-registered 2026-08-12: under 65% first-team snap share OR over 30% inline-blocking rate in the preseason opener kills the call immediately; separately, ADP 38.5 or richer required by the 8/18 print.

Line at call: ADP 13.4 entry (FFC PPR 12-team, 7/28); ADDED at 14.6 on 8/12 after the naming cheapened the board against analyst consensus · Falsifier: Exits pre-registered 8/12: under 12.5 by the 8/20 print or out; emergency exit past 15.0

Line at call: Read logged 8/12; doubled down vs the trade-showcase counter on dead-cap gravity; test chain pre-registered 8/23 (game-3 starter, fallback Wk1 first snap) · Falsifier: The org naming the veteran the permanent starter

Line at call: ADP 67.5 entry (FFC PPR 12-team, 7/28); peak adverse 55.2 (8/12); hard exits pre-registered 8/18: WIN >=67.5 / LOSS <63.0 / push between, grading on the Sept 2 print · Falsifier: Activation + board round-trip below 63.0 by the grade

Rebuilt model: this committee hikes by default unless the data hands it an off-ramp; the +29k payroll print (July re-revised negative), Waller's 'hikes do not need to come at consecutive meetings' (10/8), and a minutes document whose only forward sentence says 'most... by year end' with no 'October' in it constitute the off-ramp. Hot 10/14 headline CPI was gasoline-led; market spiked 30c and faded to flat.

Line at call: Polymarket Oct hike-25 YES 22.5c (bid .22/ask .23), 2026-10-15 01:57 UTC; December hike-25 78.5c · Falsifier: F1: December hike contract daily close <60c before the blackout (+9). F2: October contract daily volume >3.5M shares AND close-to-close move >+15c (+14). Warsh fireside hinge (Thu 10/15 11:30pm ET): A recent-data + immediate-action language +18 / B explicit sidelining language beyond the minutes -6 / C restatement or global macro 0.

2026-10-14 — UNDERWRITTEN at 11% hike (SKIP call) after the minutes were read, per the 9/19 commitment. Falsifier 2 and Warsh Branch B rebuilt under editor challenge (first drafts kept in calibration #13). Entry-honesty sentence pre-written in print. This is the desk's third Fed call on the dovish side of the market - the side of both prior misses.

2026-10-14 — POST-PRINT: Sept 29-30 collapse attributed (Williams 2pm 'no need for urgency… late this year'; Aug PCE core 3.0 vs 3.3 exp; Jefferson 'may take more time') - all omitted from the pack (Board calendar not voter-complete; BEA unopened) and owned in print. LAST MARK 11 -> 6; record number stays 11. Warsh hinge: 'late this year' now Branch C (0); Branch B needs explicit naming of this meeting. Falsifiers unchanged.

2026-10-23 — PRE-MEETING GRADE OF INSTRUMENTS (immutable): F1 Dec close <60c before blackout -> Dec closed 78.5/78.5/76.5 (lowest print 75.5) DID NOT FIRE. F2 Oct >3.5M shares AND c2c >=+15c -> largest up-day +2c DID NOT FIRE. WARSH HINGE (Thu 10/15 11:30pm ET Bangkok fireside w/ Georgieva; no transcript; 6 outlets agree): 'I don't prejudge meetings. I don't believe in saying now what I might do at the meetings that we have in the balance of this year'; 'new chapter at the Fed... not in that business anymore'; 'world has broadly conformed to our expectations'; no CPI reference, no skip/pause/wait/hold -> BRANCH C, 0. Tape: Oct 27.5 (6pm-11:41pm) -> 21 (11:50pm) -> 12.5 (3:20am); Dec 78.5 -> 77.5. VOTER SWEEP: Hammack 10/15 hawkish/no guidance; Williams 10/13 'one more... no reason to believe we would need to do more'; BOWMAN 10/14 5:40am ET (pre-CPI) 'I don't feel any urgency for taking further action at this point' - ABSENT from the 10/14 pack (calendar 'Discussion' with no text), owned in print. Data since log HOT: PPI +1.3 m/m (largest since Mar 2022), core PPI +0.7, import prices +1.1, flash PMI mfg 56.0 w/ selling prices sharpest since Aug 2022, UMich 1y 4.7/5-10y 3.5 (highest since May), claims 194k, Brent 116.72. THU 10/22 collapse 12.5 -> 5.5 (Dec 78.5 -> 67 -> 71.5) attributed to NON-FED flight-to-quality (Meta capex, NYT Iran-strike report, Houthi-Saudi attacks, chip rout; 2-yr -18.5bp biggest since Aug 2025; 10-yr through 5%; oil UP 4%; claims hawkish and ignored). FedWatch 7.2% hike at 10/22 close; Reuters poll 94/100 hold, 100/100 Dec. Market now 7.5c (Dec 72.5). LAST MARK 6 -> 4 on Bowman (not in file at 6); steelman pushed to 9 on data already weighed -> challenge restored 4 ('I allowed the narrative friction to simulate new mass'). Record stays 11. Both 2:01pm notes pre-written in print. December pre-registration printed (dissent base 65/80/95; language +15/+10/-20; presser classes +10/-15; cap 95; Dec >65 Wed & >60 Thu 10/29; 2-yr -30bp nullifier). Wallet: 474,305 Oct-hike @30.8c vs 7.5 (-$110,607); +263,805 @12.82c on 10/16, 10/20 (50 fills), 10/21 (12 fills @14:16Z) - inside blackout, BEFORE the 10/22 halving; no sells since 10/1. GDP + Sept PCE both Thu 10/29 8:30 (BEA).

2026-11-03 — GRADED (statement Wed 10/28 2:00pm ET): the FOMC HELD at 3.75-4.00% by a 10-2 vote; Cook and Hammack dissented in favor of a 25bp hike (Cook's first dissent since 2022; Hammack's third of 2026). Oct hike-25 resolved NO (2.9c at 1:47pm -> 0.1c at 2:03pm). RECORD 11 -> Brier .0121 vs market .0506 at entry; LAST MARK 4 -> .0016. Fed record vs market 1-for-3. Pre-written skip note printed as written in the 11/03 issue. Statement diff vs Sept: 'solid' -> 'strong'; 'Today's policy action will support a timelier return' -> 'The accumulation of more information will help the Committee determine the policy stance appropriate to ensure a timely return of inflation'; dissent sentence added; neither 'by year end' nor 'additional firming'. Neither dissenter published an explanation (Reuters 10/30). Wallet (v2, 0.00% drift): October book realized -$167,940 (hike leg 626,084 sh @24.80c -> 0 = -$155,255; it bought 151,779 more at 4.8-6.0c Oct 26-28 plus 165,232 no-change NO at 5.8-6.6c).

The pre-registered model printed 10/23 (dissent base 65/80/95; statement phrases +15/+10/-20; presser classes +10/-15; cap 95; tape bars; 2-yr nullifier) run on the Oct 28 record outputs 55 (two dissents 80; neither statement phrase -20; both presser classes present, applied additively per the editor's literal ruling); named steps for evidence after 10/23 - Logan 'at least two more' +4, data +3, oil -2 - take it to 60. The desk underwrites the model as written; the analyst's on-record view that he would rewrite the -20 and -15 for January prints as calibration, not as a revision.

Line at call: Polymarket Dec hike-25 YES 72.5c (bid .72/ask .73), 2026-11-03 05:36 UTC; no-change 25.5c (ask .26); executable entry no-change YES at 26c (desk 40, breakeven 26) or hike-25 NO at 28c · Falsifier: Toward the hike: Oct payrolls > +120k (Fri 11/6) +4; Oct core CPI > +0.30% m/m (Tue 11/10) +5; Dec hike-25 close > 85c on 11/10 post-CPI +6 (last-mark). Toward the skip: Nov 18 minutes with 'several/many/most' participants favoring waiting beyond December or seeing the stance as possibly sufficient -4 (if 'most/many' judged a further increase likely appropriate at the next meeting or by year end +4; 'some/a few' = 0); zero voters explicitly endorsing a December increase before the 11/28 blackout -3; Dec close < 40c any day before the blackout -6 (last-mark). HINGE: Warsh Nov 16 field hearing by language class - A commitment to further action next meeting/this year +8; B concession to time or 'may be sufficient' -6; C no policy content / 'I don't prejudge meetings' 0; both A and B -> both applied. Record number grades 12/09 2:00pm ET regardless.

2026-11-03 — UNDERWRITTEN at 60 (SKIP) after the pre-registered model was run on the record; turn-1 59 included a -1 'dissenters' silence' step struck on editor challenge (silence is not evidence). Card value gate: skip 40 vs 26c ask clears by 14. Fourth consecutive dovish-side Fed call (that side 1-for-3). Wallet: tracked account long 217,004 Dec hike-25 @75.09c (505 fills 10/22-11/2; 0.00% drift) + 86,472 'another hike 2026' @80.4 + 50,000 NO 'exactly one hike' @85 = ~$275k at risk to collect ~$353k on a hike; portfolio $1.93M.