US Semiconductors — End-of-Day Verdict
0 ACQUIRE, 1 ANTICIPATE, 0 DISPOSE | NVDA QUEUED-ANTICIPATE buy-stop @224.39, (0.5x anticipate slot x 0.75 valuation modifier = 5.8% of book), expires 2026-08-12 - fills only on strength, zero capital at risk otherwise | ACQUIRE is a genuine NO ACTION: both STRONG BUYs (AAOI cv90, COHR cv76) are HARD-BLOCKED by the quality gate (Quality_Pass False; AAOI FCF - / ROE -3.9% / F_Score 4, COHR FCF -) and both are same-day sympathy moves off AAOI's 08-06 Q2 print (L6) | AMD HOLD 172, escalation NOT fired (483.36 > 475.00, +1.76%, narrowed from +3.01%), stop 475.00 re-armed | TSM HOLD, TRAIL RAISED 385.57 -> 389.22, catalyst-aware hold into the ~08-10 TSMC July revenue print
Recommendations
1 to acquire · 0 to dispose
| Ticker | Name | Close | Action | Action Price Where to act — a buy-limit on a pullback or a buy-stop above resistance, not the current price. Sanity-check vs Close: on a held name, an action price stranded above the close means the queued level was set when the stock was higher — stale, awaiting reprice. | Stop The risk-defined exit, set below the entry for a long (~1.2–1.8×ATR). A held long whose Stop sits at/above the Close is already below its stop — a stale box awaiting reprice. | 1wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | 4wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | 12wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | Rationale |
|---|---|---|---|---|---|---|---|---|---|
| NVDA | NVIDIA Corporation | 223.96 | BUY-STOP — QUEUED @224.39 | 224.39 | 201.51 | 238.12 (+6.1%) | 270.17 (+20.4%) | 240.02 (+7.0%) | We are queueing a buy on NVIDIA that only executes if the stock keeps going up. The order sits at, a fraction above Friday's close of, and it triggers on strength rather than on a dip - if the shares stall or fall, nothing is bought and no money leaves the account. NVIDIA had its best week in months, up more than 10%, after SpaceX told investors it will build its data centres - on Earth and in orbit - exclusively on NVIDIA chips, and expects to take a meaningful slice of next year's GPU supply. That puts a very large new customer behind a business already growing at roughly 94% a year, and it is why NVIDIA is the strongest reading on our twenty-name board (conviction 88, trend score 93) and the only one of them that also passes our financial-quality screen. The shares are not cheap on any conventional measure, so we are deliberately buying a starter position - about under 6% of the book - with a stop at that caps the loss at roughly six-tenths of one percent of the portfolio against roughly twice that in upside. | Operational: ANTICIPATE slot (ADR-0012), gate MET - BUY - ANTICIPATE printed in 2 of the last 3 snapshots (08-05, 08-07) and cv88 clears the 55 floor. Sizing 77.68% x x 20% buy_pct = -> x0.75 valuation modifier (MoS_FV -2.16, the -3 < MoS <= -1 tier) = -> x0.50 anticipate-slot factor = -> @ 224.39 = val_adj SUPPRESSED to 0 under G2 / learnings note B (DCF anchor against a price is a broken hyper-growth artifact and must not originate the -10 floor haircut); the size modifier is deliberately left live, same asymmetry as the 08-06 card. Quality gate PASSES (Quality_Pass True, FCF +1bn, ROE 81.7%) - the only bullish row on the board that does. AnticipationUp_VCP|buy has NO scorecard cell (n=0) so bias 0 is an absence of evidence, not an endorsement. SUPERSEDES the 08-06 day-limit @<=220.00, which expired unfilled (08-07 low 220.66) - repriced to the live printed Entry_Price per the provenance rule, NOT carried forward. Entry band 222.48-226.30 (224.39 +/- 0.25xATR 7.63). f=0.30 MANDATORY on the 1wk (L5, calib_1w -0.548). 12wk 240.02 exceeds the DCF anchor AND prints below the 4wk - inverted curve, systemic finding #3, weakest of the three numbers. NVDA Q2 FY27 is 08-26, inside the 4wk horizon. Auto-withdraw if the settled close prints conviction < 65, the label reverts to WATCH/NO TRADE/AVOID, or expires_on 2026-08-12 passes unfilled. · news |
| AMD | Advanced Micro Devices, Inc | 483.36 | HOLD | — | 475.00 | n/a | n/a | 488.82 | We are holding AMD for a third session against a line we wrote down in advance: a decisive close below triggers a full exit. AMD closed at - still above the line, but the margin has narrowed from 3.0% to 1.8%, and Friday was an ugly session in which the stock opened at, touched, and gave the whole move back. The underlying quarter was strong - revenue up 50%, data-centre sales more than doubled - but the market has spent the week rewarding NVIDIA (up over 10%) and punishing AMD (down roughly 8%) for the same AI story, and our forward twelve-week estimate has now slipped to, barely above the current price. We are not selling on a near miss, but the position is on a short leash and the exit is pre-committed. | Operational: origination test run in full. Signal -> NO TRADE cv0 ('Consensus conflict + low conviction'), not a sell label, and NoTrade|sell is the book's only ACTIONABLE cell at bias -5 (n=8, hit_1w 0.25) - a direct empirical instruction against originating sells from NO TRADE rows here. L2 -> fails leg 1 (483.36 > 475.00). ADR-0012 -> PROFIT_ARMED fires and tick (c) NOW FIRES for the first time (12wk 488.82 < 1.02 x close 493.03, by), but the arm is FALSE - the position is -6.34% below cost with Gain_ATR -0.90 - and ADR-0012 is a profit-take with no profit to take (systemic finding #1, 13th straight session). ADR-0020 -> Peak Stop 526.88 vs live 475.00 = -9.85% against the 15% threshold; mandatory-trim trigger at a stop below 447.85, recorded for the 5th consecutive session. Chandelier 2.0xATR = 410.93, far below the standing line, so 475.00 is RETAINED not loosened, and re-armed unchanged for the 08-10 settled close. · news |
| TSM | Taiwan Semiconductor Manufacturing C | 420.04 | HOLD | — | 389.22 | n/a | n/a | 394.35 | We are keeping the small TSMC core and raising the safety net under it for the second session running. The shares rose to, and because our exit line is calculated from the price, that lifts it from to - stops only ever move one way. The reason to still be holding lands on Monday: TSMC reports July monthly revenue, the single most frequent read the market gets on whether AI capital spending is still accelerating. June was NT7bn, up 68% on the year, on full-year guidance already raised above 40%, and the last quarter delivered a record 77% jump in profit. Selling a deliberately small core position one day before that number would be churn, not risk management. | Operational: catalyst-aware hold BINDING (TSMC July revenue ~08-10, one session out; trim only on a confirmed lower high or a decisive stop break, never pre-trim into the catalyst). ADR-0012 tick (c) fires (12wk 394.35 < 1.02 x close 428.44) but the arm beneath it is the 25th CONSECUTIVE FALSE ARM (-4.10% below cost, Gain_ATR -1.17) - armed-but-not-fired resolves to the mandatory trail-raise, which is what is being done. Chandelier 2.0xATR = 420.04 - 2 x 15.408 = 389.22, above the standing 385.57, so it ratchets; the line now sits 7.92% under price. USC-5 satisfied: the ARE the defined core (78% was already cut on 08-03), not a discretionary majority. Technical is not broken - price is well above trail, 12wk improved 386.72 -> 394.35 - so USC-5's de-risk clause does not apply. NoTrade|sell bias -5 argues against originating a sell here in any case. · news |
Outcome & track record
Accuracy at the time of this record.
19 Graded decisions Past recommendations old enough to score against what actually happened (target hit / stopped out / closed). This is the evidence base — the bigger the number, the more history behind the calls. Click to inspect every one. See the decisions →
-0.548 1-week calibration How much of the eventual 4-week move typically shows up in the first week (0 = almost none, 1 = most of it). Use it to read the 1-week estimates: a low number means prices drift toward the target slowly, so don't over-read a soft week-1.
53 Published records How many dated end-of-day verdicts exist for this book — the length of the public track record you can browse in the Archive below.