This book splits cleanly on the trailing year, and the split doesn't match the week's RSI readings. TSLA, SBUX, and CROX are all up on the trailing year and TSLA's regime just upgraded out of collapse for the first time in this dataset — real repairs. HIMS, LULU, DPZ, NKE, and BRBR are all negative over the trailing year, including HIMS, which posted the book's strongest weekly return (+20.07%) sitting on top of a -23.16% trailing year — don't read that week as strength.

Data as-of the 2026-08-21 close for all eight names — no staleness this cycle. No 2026-08-21-dated scan exists for this lane; the comparison baseline below is the August 14 scan.

TSLA is the clean repair story this cycle. Its regime upgraded from collapse to downtrend — the first regime change out of collapse this book has shown across the cycles tracked here — as its RSI climbed to 58.0 from 48.7 and its 30-day loss narrowed further to -2.98% from -13.23%, extending what's now a multi-cycle pattern of real, cumulative improvement. TSLA's trailing year is modestly positive at +13.35%, even though it still trades 10% below its 200-day with a death-cross on its longer moving averages — a trend under test, genuinely improving, but not yet fully confirmed at every horizon. SBUX remains the steadiest name in the book: RSI cooled slightly to 55.0 from 58.0, still basing, still up 20.91% over the trailing year and comfortably above its 200-day. CROX had the sharpest technical cooling in the book this cycle — RSI fell to 38.6 from 50.1, its 30-day loss deepened to -10.35% — but it's still up 45.7% over the trailing year and trades 20.8% above its 200-day. Read against the week alone, CROX looks like the worst name in the book; read against the year, it's a deep pullback inside a trend that hasn't broken.

The other five names in this book are genuinely weaker once the trailing-year window is applied, regardless of how their individual weeks read. HIMS is the clearest case for taking the long-horizon check seriously: its 7-day return of +20.07% is the strongest in the entire book this cycle, and its RSI climbed to 58.1 from 42.8 — on the week alone, this reads as the best story in the book. But HIMS is down 23.16% over the trailing year with an alpha1y of -44.97, among the weakest annual reads in this scan. A hot week does not undo that; the underlying trend is still weaker than the week suggests. LULU shows a milder version of the same pattern — its 7-day return flipped positive (+1.25%, was -7.02%) and RSI recovered to 51.9 from 48.6, but it's down 38.91% over the trailing year and trades 22% below its 200-day with a death-cross. NKE's regime downgraded fresh to collapse this cycle, and here the technicals and the trailing year agree: down 46.5% over the year, 22.5% below its 200-day, a death-cross in place. DPZ is similar — still tagged downtrend despite carrying a positive 30-day return of +6.88%, but its trailing year is negative at -23.5% and it sits below its 200-day, which resolves the multi-cycle disconnect this name had been carrying between its monthly gains and its regime tag: the annual data now argues the regime tag was the more honest read all along.

BRBR remains the clearest veto in the book: thesis explicitly rated broken, RSI at 35.3 (the lowest in the book), down 74.38% over the trailing year and 43.7% below its 200-day. There's no long-horizon counter-argument here — every window agrees this name is weak.

• TSLA: regime upgraded collapse → downtrend, the first upgrade out of collapse this book has carried; RSI 48.7 → 58.0; 7-day climbed to +6.02% (was +4.17%); 30-day narrowed further to -2.98% (was -13.23%).
• HIMS: RSI 42.8 → 58.1; 7-day flipped to +20.07% (was -10.89%), the strongest weekly move in the book; still -23.16% over the trailing year with a deeply negative alpha.
• LULU: RSI 48.6 → 51.9; 7-day flipped to +1.25% (was -7.02%); still -38.91% over the trailing year and below its 200-day.
• NKE: regime downgraded to collapse (was downtrend); RSI 42.6 → 46.0, roughly flat; trailing year -46.5%, one of the weakest annual reads in this book.
• CROX: RSI 50.1 → 38.6, the sharpest technical cooling in the book; 30-day loss deepened to -10.35% (was -1.24%); still +45.7% over the trailing year.
• DPZ: RSI 53.5 → 51.3, roughly flat; 30-day still positive at +6.88% (was +11.39%); trailing year now confirmed negative at -23.5%, resolving the prior price-vs-label disconnect.
• SBUX: RSI 58.0 → 55.0, mild cooling; still +20.91% over the trailing year, the steadiest name in the book.
• BRBR: RSI 36.0 → 35.3, still the lowest in the book and thesis rated broken; -74.38% over the trailing year, the weakest read in this book.

Regime labels describe chart structure, not a buy/sell call, and this cycle is the sharpest example in this scan set of why the trailing-year check matters before writing any trend verdict. Taken on 7-day returns alone, HIMS looks like the best story in the book and CROX looks like the worst — the trailing year says close to the opposite. HIMS is down 23.16% over the year with a deeply negative alpha, so its hot week reads as noise sitting on top of real weakness. CROX is up 45.7% over the year and well above its 200-day, so its rough week reads as a pullback inside an intact trend, not a breakdown. TSLA's fresh regime upgrade out of collapse is the headline structural change in the book — a real, multi-cycle repair finally confirmed by the tag, even though the name still trades below its 200-day. DPZ's long-standing disconnect between strong monthly gains and a stuck downtrend tag is now resolved: the trailing year is negative, so the tag was the more accurate read the whole time. BRBR has no long-horizon counter-argument in either direction — every window agrees it's the weakest name in this book.

Tier A — Genuinely intact, positive trailing year (3): TSLA (fresh upgrade out of collapse), SBUX (steadiest in the book), CROX (roughest week, but still up 45.7% on the year).

Tier B — Hot week, negative trailing year (2): HIMS (strongest 7-day in the book, but -23.16% on the year), LULU (mild bounce, still -38.91% on the year).

Tier C — Weak on every horizon (3): DPZ (disconnect resolved negative), NKE (fresh collapse tag, both windows agree), BRBR (thesis broken, weakest read in the book).

None of the eight names reads as a clean fresh entry this cycle. SBUX remains the steadiest confirmed structure, still up on the trailing year. TSLA's regime upgrade out of collapse is the most consequential structural change in the book, though it still trades below its 200-day. CROX's rough week sits on top of a still-strong trailing year, worth a second look once the technical cooling stabilizes. HIMS and LULU's weekly bounces don't change the negative trailing-year picture — neither reads as an entry yet. BRBR's thesis stays rated broken.

• Tape data (price/RSI/SMA/returns/1-year, SMA200): full-scan pre-computed summaries as of the 2026-08-21 close — cultural-thesis.