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 |
|---|---|---|---|---|---|---|---|---|---|
| ASML | ASML Holding NV | — | HOLD | — | 1611.58 | n/a | n/a | 1656.90 | Hold the. The stock has recovered above its short- and medium-term trend lines and the engine no longer flags it for sale, so the eight-session exit campaign ends here rather than being reissued a ninth time. The support is fundamental as well as technical: the three leading logic and memory manufacturers are all committing to ASML's newest lithography platform, and the company is expanding capacity to meet it. The position is still 3.4% below cost and the twelve-week projection remains 6.1% below today's price, so this is a hold-and-watch, not a re-entry: it stays on a defined risk line and the case is re-tested each session. Position @ cost 1826.98, close 1764.85, market value, unrealized - (-3.40%). No Target_Price on an AVOID row so 1wk/4wk are n/a rather than fabricated; model next-session band Est_Low 1638.90 / Est_High 1800.15. 12wk 1656.90 (-6.12%), still far above the 836.25 DCF anchor (projection assumes growth beyond priced-in). Quality_Pass False. Trend 69, Consolidation_Bullish, close above EMA10 1716.36 / EMA20 1727.39 / EMA50 1726.65 / EMA200 1495.90. ASML Q3 due 2026-10-14, outside horizon. · news |
| NVDA | NVIDIA Corporation | — | HOLD | — | 221.79 | 230.46 | 241.49 | 251.32 | Hold all but tighten the risk line. The model's confidence in this setup fell sharply this session - from, its largest one-session drop on record here - and NVDA was the only name in the book to fall while nineteen others rose. That is a real deterioration and it is being respected with a raised stop rather than ignored. It is not yet a reason to sell: the position sits fractionally below cost, so there is no profit to protect and no exit trigger; the longer-term trend is intact with the price above every major moving average and 4.6% off its 52-week high; and the fundamental picture is unchanged after an August quarter that beat with 2bn of revenue and roughly $108bn guided for the current quarter, at a reasonable ~25x forward earnings. The discipline here is to raise the stop and let the position prove itself, not to bank a small loss on a momentum wobble. Conviction collapse (-32); family rotated AnticipationUp_VCP -> FailedBreakdown_Up_20D; Signal_Rationale [STRONG BUY]->[WATCH]; Momentum_Score; close 225.73 (-2.01%). NOT armed (Gain_ATR -0.14, Unreal_PnL_Pct -0.47%, T12_Progress 89.82 just under the 90 threshold), so ADR-0012 never engages and the -32 drop cannot fire as tick (b). Exit_Warning blank. TRAIL RAISED 215.95 -> 221.79 (live Stop_Price, ratchet convention). WARNING: headroom is only 1.75% / 0.50xATR on a high_vol name (ATR 7.88), down from 1.89xATR - a routine session can take it out. PRE-COMMITMENT: a breach of 221.79 alone does NOT originate a sell; L2 requires stop-break PLUS a negative 12wk and the 12wk is +11.34%. Estimates f=0.30 (L5, calib_1w -0.226 degenerate). 12wk 251.32 exceeds the 163.98 DCF anchor. Market value = 82.7% of equity. Ex-div on 2026-09-10, one session out, clips ~0.11% against a 0.50xATR stop. · news |
| TSM | Taiwan Semiconductor Manufacturing | — | HOLD | — | 415.38 | 458.01 | 502.38 | 439.06 | Hold the and raise the stop. This is the strongest trend reading in the book - the price is above every moving average from the 10-day to the 200-day, and the business is delivering: July sales grew 44.7% year on year and the first seven months of 2026 are running 37% ahead of last year on AI demand. The engine flags a profit-take, but the position is only 0.2% above cost, so there is no profit to take; the flag is an artifact of the twelve-week projection drifting down onto today's price rather than the price reaching a target. With August monthly revenue due within a session, the right move is to hold through the print behind a tighter risk line rather than pre-trim into it. Conviction (drift, not collapse), Trend_Score 93 (highest in the book), close 439.00 above EMA10 422.82 / EMA20 420.40 / EMA50 418.34 / EMA200 377.65, RSI 61.7. PROFIT_ARMED DISCARDED for the 9th session: Gain_ATR 0.09, Unreal_PnL_Pct +0.22%, arming only off T12_Progress 99.99 because Target_12Week 439.06 has collapsed onto the 439.00 close (systemic finding 2). Fire tick (c) Target_12Week < 1.02 x Close is technically true but ticks require an ARMED position, so ADR-0012 does not engage - trail-raise, not trim. TRAIL RAISED 409.63 -> 415.38 (live Stop_Price), 5.38% / 2.17xATR headroom, fresh peak watermark. Trim remains WITHDRAWN (2nd session) per USC-5. Estimates f=0.30 (L5). G2 SUPPRESSES the valuation layer entirely: DCF 46.22 is the ADR-ratio artifact (1 ADR = 5 ordinary shares; correct anchor), so the -13.82 MoS_FV is unusable. Market value, unrealized +. Catalyst-aware hold: TSMC August revenue ~2026-09-10 (inside the <=5-session window), ex-div on 2026-09-16. · news |
| — | Micron Technology, Inc. | — | NO ACTION | — | — | n/a | n/a | n/a | The best-shaped trade on the board and still not taken. Micron offers the strongest reward-to-risk in the book - roughly four dollars of upside for every dollar risked - with a constructive trend and a twelve-week projection 13% above today's price. It is declined on track record, not on shape: this book has taken five trades from this particular setup pattern and none reached its four-week target, with the earlier three all stopping out within a week. Until the pattern earns back its credibility, or the entry re-appears as a genuine pullback rather than a bounce, the capital stays uncommitted. Buy_Rank 1. Entry 1000.26, stop 977.81, target 1090.06 -> R:R 4.00:1 (best on the board). 12wk 1133.87 (+13.4%), one of only four positive forward views. Consolidation_Bullish, Trend 78, close above EMA20/50/200. BLOCKED: 68 - 6 (val_adj, MoS -1.16 x 5) - 10 (USC-3 mandated floor on FailedBreakdown_Up_20D|buy; cell now n=5 ACTIONABLE at hit_1w 0.40 / hit_4w 0.00; scorer computes -2, USC-3 overrides to -10) = 52, thirteen short of the 65 floor. Conviction also FELL while price fell -1.61%. Fourth consecutive NO ACTION. RE-ARM: a PullbackBuy_VWAP print, or a quality-gate-passing confirmation of the FailedBreakdown entry. MU FQ4 due 2026-09-30. |
| — | Sandisk Corporation | — | NO ACTION | — | — | n/a | n/a | n/a | Same setup as Micron, same rejection, and now a fading one. Sandisk shows attractive reward-to-risk on paper, but the buy signal is weakening rather than building - the model's confidence dropped while the price went nowhere, one week after an 11.9% single-session jump. That is the classic pattern of a spike that fails to hold, which this book has been caught by before. Declined. Buy_Rank 2. Entry 1737.99, stop 1685.68, target 1947.23 -> R:R 4.00:1; 12wk 1969.65 (+13.3%). BLOCKED: 68 - 6 (val_adj, MoS -1.11 x 5) - 10 (USC-3 FailedBreakdown floor) = 52, thirteen short of 65. Label decaying cv76 -> 68 with price flat (-0.12%) after last week's +11.90% gap - the US-1 fade pattern. L6 gap-chase caution also applies. RE-ARM: a PullbackBuy_VWAP print or a pullback toward EMA20 with the family intact. |
| — | Lumentum Holdings | — | NO ACTION | — | — | n/a | n/a | n/a | The strongest technical signal this book has ever printed, and it is still declined. Lumentum jumped 11% on renewed enthusiasm for AI data-centre optics, and the model's entry level sits above Tuesday's close - meaning the book would only buy on further strength, which is normally the disciplined way in. Two things stop it. The company fails the quality screen on cash generation and balance-sheet strength, which is a hard block on any new position here. And the price has now run 521% in a year to about 104x earnings, with the freshest bullish broker upgrade carrying a target that sits BELOW the level the book would have to pay - so even the most optimistic published view offers no headroom from the entry, while analysts flag that large cloud customers can commoditize this supply. Paying above the bull case on a name that fails the quality screen is the exact chase this book has been burned by. Watching, not buying. BLOCKED on two independent hard grounds: (i) Quality_Pass == False - F-Score/FCF screen failed; the quality gate is a hard block on new entries and USC-3 explicitly forbids overriding it to admit a chase; (ii) the ADR-0012 anticipate slot requires BUY - ANTICIPATE in >=2 of the last 3 snapshots and LITE printed WATCH 42 -> WATCH 57 -> BUY-ANTIC 100 =, a single-session flip after +11.04%. val_adj SUPPRESSED to 0 per learnings note B (DCF 53.03 vs a 978.53 price is a broken anchor, not a valuation, so the -34.07 MoS must not originate a -10 haircut). Scorecard AnticipationUp_VCP|buy n=1, hit_1w 0.00, report-only - stated, not applied. Entry would be a GTC buy-stop 1006.38 (+2.85%, Entry_Distance_ATR +0.42, self-limiting), stop 805.30, target 1408.53, 12wk 1136.71 (+16.2%). RE-ARM LIVE: a second BUY - ANTICIPATE print on the 2026-09-09 close satisfies 2-of-3; the quality gate would then be the only remaining blocker. |
| — | ASE Technology Holding | — | NO ACTION | — | — | n/a | n/a | n/a | A fresh buy signal that clears the confidence bar but fails on quality and on its own forward view. ASE Technology fails the balance-sheet and cash-generation screen, which blocks new positions outright, and the model's own twelve-week projection sits fractionally below today's price - so the setup is asking the book to buy strength into a flat forward view. The signal is also one session old after a 6% move, and this is one of the three names that stopped this book out on the same pattern earlier in the year. Declined. Adjusted conviction 83 - 10 (val_adj, MoS -9.83) = 73, clears the 65 floor. BLOCKED on: (i) Quality_Pass == False (hard gate on new entries); (ii) 12wk 39.70 sits BELOW the 39.79 close; (iii) anticipate slot precondition (NT 0 -> AVOID 64 -> AVOID 53 -> BUY-ANTIC 83) after +6.08%. Historical: ASX is one of this book's three realized FailedBreakdown stop-outs (2026-07-10, -8.65%). Entry would be a GTC buy-stop 40.47, stop 35.68, target 50.05. |
Outcome & track record
Accuracy at the time of this record.
45 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 →
-0.226 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.