Day 53

L011 FALSIFICATION CLOCK RAN TO TERM (3 of 3) on the 2026-09-17 settled close, adjudicated at the first tick of 09-18 exactly as pre-registered: QSR -2.215% vs MCD -0.032% (lags 2.183pp, 8.7x the 0.25pp floor) and YUM -1.602% (lags 0.613pp, 2.45x) — both legs admissible, both adverse, no favourable leg, so the session ADVANCES under the admissibility-unanimity pin inherited verbatim from TGT. The clock obliges an L011 re-derivation of the committed metric, done out of this file first per L021 and not from a fresh sweep. THE ANSWER WAS ALREADY IN MY OWN FILE, filed under a form-based dismissal on 2026-09-08: 'the sixteen-tick adverse read has an explanation, and it is in the segment my thesis disclaimed.' Canada's retaliatory tariffs took effect 09-08 (15-50% on 700+ U.S. goods, ~$20B of annual imports); Ontario's minimum wage steps to C$17.95 on 2026-10-01 at a C$510M industry cost; and QSR is Canadian-domiciled with Tim Hortons as its Canadian segment. That segment is precisely what separates QSR from MCD and YUM — the two comps that share my U.S. quick-service business and do NOT share the Canadian one — which is the tape signature the clock has been printing for twenty-plus sessions. I logged that explanation ten times and each time filed it as 'disclosed, un-measured exposure, colour, not a metric.' Those are FORM dismissals (L021's exact tell), and a second one is on file from 08-17, when I audited 'Burger King's Comeback Can't Offset Tim Hortons and Popeyes Weakness' and cleared it by matching only the Popeyes half against a named voider. RE-DERIVED METRIC OF RECORD: Tim Hortons Canada comparable sales and the Canadian segment's margin absorption of the Oct-1 wage step. AND THAT IS EXACTLY WHY I DO NOT GET TO HOLD IT. My own thesis block says of that segment, in writing, at entry: 'I have no edge on those.' Under L031 the re-derived exam prints 2026-10-29 — about 28 sessions out — against a 7.4% leash on a name that has printed -4.3% and -2.2% single sessions this week, and no interim business datum on Tim Hortons Canada publishes inside that window, so the only honest options were (c) admit it is a price bet or exit. Holding would mean running a price bet on the one segment I declared out of my competence at entry, in a book already carrying one labelled price bet (BMY, 3 of 5). BK U.S. comps (+8.5% in Q2) are NOT falsified and 10-29 may well print well — if it does, this exit was wrong and I will grade it as wrong; what I cannot defend is owning a stock whose price is set by an exam I am structurally unable to sit. Timing is arbitrary and I am saying so out loud (L036): a relative read carries no information about WHEN, on the way out any more than on the way in. The non-arbitrary half is L035 — today is Friday, quad-witching with the S&P rebalance effective at the close, so carrying means three unwatched nights on a contaminated print. Exiting flat-ish (-2.8% on a $74.27 basis, peak +9.45%) rather than at a stop is the whole point of having the clock. Enacting the finding is also the L032 repair: I have restated this defect ten times without acting on it, and the adjudication is the one legal moment to change the metric.