US Semiconductors — End-of-Day Verdict
NO ACTION - 0 ACQUIRE, 0 DISPOSE | AMD escalation TESTED and NOT fired (settled close 482.05 > 475.00, +1.48%) - HOLD 172, stop 475.00 re-armed unchanged | TSM 78% reduce FILLED 164 @ 419.32 - HOLD the 47-sh core into the ~08-10 TSMC July revenue print, stop 382.61 | NVDA BUY - ANTICIPATE cv83 DECLINED - anticipate slot needs prints, this is 1, and it is already 0.47xATR past its own entry on a news-driven +3.4% day
Recommendations
0 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 | 219.22 | BUY-STOP | 215.24 | 189.86 | n/a | n/a | 219.29 (+1.9%) | NVIDIA printed this book's first bullish signal in ten weeks - an AnticipationUp_VCP setup at conviction 83, its strongest bullish reading since inception, backed by a real competitive win: SpaceX named NVIDIA's Vera Rubin chips the exclusive silicon for its Starmind orbital-compute program, with Elon Musk calling the architecture 'the best available' and locking out AMD's rival MI450. We are not buying it today. The setup is an anticipation print, which under our rules needs to appear twice in three sessions before it earns a position - this is the first - and the shares have already run 3.4% past the model's own entry level on the day of the news itself. Paying up into a one-day headline gap is the mistake this book has made before. NVIDIA's own results are still three weeks away, so there is no reason to be early. OPERATIONAL: anticipate-slot gate fails ( required prints); Entry_Distance_ATR -0.47 = past trigger; L6 no-gap-chase + USC-4 no-chase queue both apply. val_adj suppressed to 0 per learnings note B / G2 (DCF 163.98 vs Target_12Week 219.29 = unreliable anchor); Quality_Pass True; AnticipationUp_VCP has no scorecard cell, bias 0. RE-ARM: a 2nd BUY - ANTICIPATE print at conviction >=55 on the 08-06 or 08-07 settled close opens the slot -> GTC buy-stop at THAT snapshot's then-printed Entry_Price (never 215.24 - provenance rule), 0.5x normal new-BUY, 3-session expiry. Alternatively a PullbackBuy_VWAP print (L1) admits it to the ordinary ACQUIRE pool. · news |
| AMD | Advanced Micro Devices, Inc | 482.05 | HOLD | — | 475.00 | n/a | n/a | 497.88 | We are holding AMD, and the reason was decided before today. Last session we committed to a single test - a decisive close below would trigger a full exit, judged on the settled close rather than the after-hours tape. AMD closed at, so the line held, by a thin 1.5%. The quarter behind the selloff was a record one: revenue 5bn up 50%, data-centre sales more than doubling to 7bn, and next-quarter guidance above consensus. The shares fell ~7% anyway on a jump in capital spending ( against a year ago), a softer second-half PC outlook, and a guide that beat the published consensus but missed the whisper number - on a stock that had already doubled this year. Wells Fargo, Jefferies and JPMorgan all raised their price targets after the print. AMD also lost the SpaceX Starmind order to NVIDIA outright, which is a genuine setback - but news of that kind can cancel a plan to buy, never on its own force a sale. The stop stays at and decides the next session. OPERATIONAL: escalation TESTED and NOT fired (482.05 > 475.00, +1.48% / +0.165xATR); re-armed unchanged for the 08-06 settled close, no re-litigation. L2 fails BOTH legs (above stop; 12wk 497.88 = +3.28%, clears the 1.02x dead-view threshold 491.69 by). PROFIT_ARMED is a FALSE arm (Unreal_PnL -6.59%, Gain_ATR -0.80) -> ADR-0012 profit-take does not apply, no tier. Trail-check: 2.0xATR chandelier 396.72 is far BELOW the standing 475.00, so the tighter existing line is RETAINED - not loosened, not ratcheted. Scorecard NoTrade|sell bias -5 (n=8, hit_1w 0.25, deteriorated from -3) argues directly against originating a sell from today's NoTrade label. ADR-0020 WATCH: Peak Stop 526.88, live 475.00 = -9.85%, no size cut since peak; a stop below 447.85 makes a 30% trim mandatory and veto-immune. · news |
| TSM | Taiwan Semiconductor Manufacturing C | 414.00 | HOLD | — | 382.61 | n/a | n/a | 376.25 | Yesterday's decision to cut TSM by 78% executed this morning at - a better price than the we planned on - leaving a deliberately small 47-share core. That core exists for a reason: TSMC reports July monthly revenue around 10 August, three sessions away, and it is the highest-frequency read on whether the AI capex build is still accelerating. The company raised full-year growth guidance above 40%, reclaimed a $2trn market value, and trades near 20 times forward earnings. The shares sit 8% above our stop. Cutting a position again, days before the event it was sized to survive, would be churn rather than risk management. We hold and let the print speak. OPERATIONAL: 08-04 reduce FILLED @ 419.32 (realised - / -4.27% vs 438.02 basis); this is the USC-5 defined-core prescription, not a deferred decision. Anchor row prints NO TRADE cv0 - NoTrade streak RESET to 1 by the 08-04 STRONG SELL - REDUCE, so ADR-0012 tick (d) does not fire; tick (c) fires on a dead 12wk (376.25 = -9.12%) but PROFIT_ARMED is the 23rd consecutive FALSE arm (Unreal_PnL -5.48%, Gain_ATR -1.46) so there is no profit to take. L2 fails leg 1 (414.00 > 382.61, +8.20%). Catalyst-aware hold: a trim through a <=5-session catalyst requires a confirmed lower high OR a decisive stop break - neither present. Scorecard NoTrade|sell bias -5 (n=8, hit_1w 0.25) penalises exactly this origination. Trail-check: 2.0xATR chandelier 381.09 is BELOW the recorded 382.61, so the stop is RETAINED. ADR-0020: watermark re-based to 382.61 with a 78% size cut already taken, erosion 0%. · news |
Outcome & track record
Accuracy at the time of this record.
18 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.