US Semiconductors — End-of-Day Verdict
1 DISPOSE - 0 ACQUIRE | TSM SELL 78% (164 of) @ ~417.17, 47-sh core retained behind 382.61 | AMD HOLD 172, prior trim WITHDRAWN (both ADR-0012 ticks invalidated), stop 475.00 held, decisive close < 475.00 re-fires the full exit | ACQUIRE empty - zero bullish rows on the board
Recommendations
0 to acquire · 1 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 |
|---|---|---|---|---|---|---|---|---|---|
| TSM | Taiwan Semiconductor Manufacturing | 417.17 | SELL 77.89% | 417.17 | 382.61 | 404.17 (-3.1%) | n/a | 373.84 (-10.4%) | Reduce three-quarters of the Taiwan Semiconductor holding into strength. The model has moved from six sessions of ambiguity to an explicit STRONG SELL - REDUCE at conviction 69 and the board's top sell rank, and it is doing so at - the best price of the whole seven-session exit sequence and the top of a two-week -425 range, while its own twelve-week view points back to (-10.4%). The 07-22 high of has stood unbeaten for nine sessions, a confirmed lower high. TSMC's business is not the problem - a record quarter, guidance raised to 40%+ growth, and a ~28x multiple against a ~51x sector - which is exactly why this is a reduction and not an exit: a 47-share core stays on to carry the ~08-10 July revenue print. Sell 164 of at ~417.17, proceeds realised ~- (-4.76% vs 438.02 basis); retain behind a stop raised 375.56 -> 382.61. Portion is the engine's own printed 77.89%, size_mod 1.0 - G2 suppresses the MoS_FV -13.77 amplify (DCF is a broken artifact) so no tier bump, val_adj 0. Scorecard bias 0: the actionable NoTrade|sell -3 cell governed the six prior inferred cards and does NOT reach this one (FailedBreakout_Down_20D has no sell cell). PROFIT_ARMED is the 22nd consecutive false arm (Gain_ATR -1.21 on a -4.76% loss) and is NOT the basis; the stop is reclaimed by +9.03% so L2 is not claimed. Catalyst carve-out met via the confirmed lower high. Estimates: 1wk 404.17 at f=0.30 (L5, calib_1w -0.548) interpolated toward the 12wk because a REDUCE row prints no Target_Price; 4wk n/a, model band Est 375.00-425.51. Supersedes and reprices the unfilled 08-03 100% exit @406.11 (L8, L3/USC-6). · news |
| AMD | Advanced Micro Devices | 518.58 | HOLD | 518.58 | 475.00 | n/a | n/a | 547.27 (+5.5%) | Hold the full AMD position and cancel the pending trim. The case for cutting has gone: the stock closed +7.0% at, back above our cost, conviction turned back up from zero to 64, and the twelve-week view revived to (+5.5%) - so none of the deterioration tests that authorised last session's trim still fire. Q2 landed after this close and was a record: revenue (+50%), EPS, data centre (+107%), and Q3 guided -13.3B against a consensus - a raise. Shares fell ~6-8% after hours on an capex figure versus a expectation. That is a multiple compressing, not a thesis breaking, and selling a record quarter into a sell-the-news reaction is the mistake, not the discipline. The stop is where this gets decided. PROFIT_ARMED but ZERO ADR-0012 ticks fired -> mandatory trail-raise, no trim. Stop HELD at 475.00: the 2.0xATR chandelier off 518.58 is 432.10, i.e. BELOW the standing line, so the tighter existing stop is retained; not ratcheted up into a 477-489 indicated open. Committed escalation stands unchanged and un-relitigated: a decisive close below 475.00 re-fires the FULL exit. News overrides the anchor PRICE (L8 >1.5% gap) but per CLAUDE.md sec9 / Step 7 news demotes or cancels - it never ORIGINATES a sell; L2 also fails both legs (above stop, 12wk positive). USC-5 test: constructive fade -> trail, not broken technical -> de-risk (below EMA10/20, straddling EMA50 487.48, but +11.3% over EMA100 and +37% over EMA200). ADR-0020 WATCH, not fired: Peak Stop 526.88, live 475.00 = -9.85% with no size cut since the peak; a stop below 447.85 makes a 30% trim mandatory and veto-immune. Arm is LEGITIMATE this session (T12_Progress 94.76 on +0.49% gain) - first non-false arm in this book. val_adj 0 via G2 (DCF vs is a broken artifact). · news |
Outcome & track record
Accuracy at the time of this record.
17 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.