Recommendations
0 to acquire · 3 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 | 227.98 | SELL 30% | 227.98 | 211.56 | 243.72 (+6.9%) | 280.45 (+23.0%) | 241.32 (+5.9%) | Selling 76 of our 253 Nvidia shares - about 30% - into the eight-percent jump the stock made after reporting a blowout quarter, and keeping the other 177 running. This is not a change of mind about Nvidia: revenue of beat the expected, the current quarter is guided to -110.1B, and management pointed to roughly 70% revenue growth next fiscal year against the 45% Wall Street had penciled in. It is a discipline trade. One session ago this holding was 7.6% underwater; the post-earnings gap has carried it back to break-even, and our exit framework has flagged a momentum divergence - the price made a new high that the underlying momentum reading did not confirm. We wrote down in advance, before the print, that this exact warning would trigger the sale without appeal, and we are honouring that rather than finding a fresh reason to hold on. Taking roughly a third off the table ahead of Friday's Jackson Hole keynote leaves the majority of the position exposed to the upside we still believe in, with a materially smaller loss if the gap fills. || OPERATIONAL: ADR-0012 armed (PROFIT_ARMED, T12_Progress 94.47) + tick (a) Tier-1 RSI_DIV = profit_take_1 30% tier; ticks (b) conv = +38 (no), (c) T12 241.32 vs 232.54 band = +5.85% alive (no), (d) HOLD (no). G1 softening does NOT apply - signal is AnticipationUp_VCP, not BreakoutUp_*/MomentumContinuation_*. Fires the 08-26 pre-commitment clause (iii) 'any Tier-1/2/3 code joins PROFIT_ARMED', stated without appeal. No valuation amplify: MoS -2.28 is not <= -3 and G2 suppresses anyway (DCF 163.98 = 1.39x price). No scorecard cell for AnticipationUp_VCP|sell (n=0) -> bias 0. Counterfactual optional_trim n=14 median -5.14%/1wk, -15.33%/4wk verdict 'cost' supports firing. Tension recorded: USC-5's counterpoint / US-3 (trail-don't-trim a constructive name) argues against, and realized on the slice is only + - see systemic finding 13. Trail RAISED 207.19 -> 211.56 (printed chandelier stop, 1.91x ATR clearance) on the retained. · news |
| ASML | ASML Holding NV | 1735.01 | SELL ALL | 1735.01 | 1736.47⚠ | 1759.73 (+1.4%) | 1817.41 (+4.7%) | 1719.53 (-0.9%) | Closing the ASML position entirely at about taking a 5% loss, because the reason we were holding it has run out. The stock closed below the stop level we set for it, it has now slipped under its fifty-day average for the first time in this holding period, its trend rating has been downgraded, and our twelve-week projection has crossed from mildly positive to slightly negative - so the model no longer expects the position to make money over the horizon we bought it for. ASML's business is not the problem: the company raised its 2026 sales outlook to roughly -51.5B, well ahead of the analysts expected, its EUV order book stretches into 2028, and Intel is already running High-NA machines in volume production. But a strong business with a broken price trend and no projected upside is not a position worth carrying, and the stock is down 4.7% over the past month despite all of that good news. We would rather redeploy the roughly than keep slicing a shrinking holding one or two shares at a time. || OPERATIONAL: basis is L2 (close through the recorded stop + negative 12wk = signal-based exit basis) reinforced by the first close below EMA50 (1,736.41) and the Trend pillar downgrading Bullish -> Neutral_Bullish_Context (score, momentum and the raw label falling to WATCH. NOT taken as a decisive stop break on its own - the breach is = 0.084% = 0.035x ATR, exactly the noise-level cross finding 9 predicted from a trail raised with only 0.21x ATR clearance. ADR-0012 tick (c) also fires for a 4th session (T12 1,719.53 vs 1,769.71 band = -0.89%, decayed from +3.55% -> +0.82% -> +0.09% -> -0.89%), but per finding 10 the PROFIT_ARMED arm is the broken T12_Progress leg on an underwater position and is not the governing basis. Supersedes the unfilled 08-25 2-share card rather than stacking on it (finding 11); the terminal-tranche rule (finding 5/7) says exit-or-hold on a 4-share remainder, not slice. Executor lag is now 3 sessions - continued reaffirmation is the ADR-0020 'never got cut at all' failure mode. G2 suppresses the MoS -4.48 amplify (DCF 841.33 = 2.04x price); val_adj 0. Cell FailedBreakdown_Up_20D|sell n=10 bias +2; adj 42+0+2 = 44. · news |
| TSM | Taiwan Semiconductor Manufacturing C | 427.30 | SELL 30% | 427.30 | 405.91 | 433.72 (+1.5%) | 448.69 (+5.0%) | 414.21 (-3.1%) | Selling one of our three TSMC shares at about a repricing of the same small de-risk we ordered on 25 August but which never filled. The trade stands for one reason: our twelve-week projection of sits about 3% BELOW today's price, so the model sees no return left over the holding horizon even though the stock just rose 2.3%. Everything else about TSMC improved - the trend and momentum readings both turned outright bullish, conviction jumped sharply, and the company has guided 2026 revenue growth above 40% with July sales up 45% year over year as Nvidia's next-generation chips ramp at its fabs. That is why this is a one-share trim and not an exit: we keep two shares in a business we like, and simply stop paying for upside the model does not see. || OPERATIONAL: REPRICED per L8 - the 08-25 card at 417.41 is 2.37% below the settled close, outside the ~1.5% threshold, so it is withdrawn and reissued at the anchor's own Entry_Price of 427.30 (action-price provenance rule). ADR-0012 armed (PROFIT_ARMED, T12_Progress 103.16) + tick (c) only, 10th consecutive session; ticks (a) bare PROFIT_ARMED, (b) conv = +29 IMPROVING, (d) HOLD - so this reaffirms the existing 30% tier and does not escalate. G2 suppresses the MoS -13.76 amplify (DCF 45.96 = 9.01x price, broken-anchor class); val_adj 0. Cell FailedBreakdown_Up_20D|sell n=10 bias +2; adj 86+0+2 = 88. Trail RAISED 404.87 -> 405.91 (printed stop, 2.00x ATR clearance) on the post-fill; EMA50 417.28 declined at 0.94x ATR under finding 9's proposed 1.0x floor. Terminal-tranche note (finding 5/7): at the NEXT slice is the position - future action on TSM is exit-or-hold, not another trim. · news |
| LITE | Lumentum Holdings | 956.14 | WATCH | 956.14 | 918.01 | 998.16 (+4.4%) | 1096.21 (+14.6%) | 1142.66 (+19.5%) | Not buying Lumentum despite it being the only outright buy signal on the board, because it fails our quality screen and the price has run too far, too fast. The business case is real - fiscal Q4 revenue of more than doubled year over year, gross margin hit 50.4%, twenty of twenty-five analysts rate it buy, and J.P. Morgan just lifted its target to - but the stock is up roughly 124% this year and 700% over twelve months, and it now trades about 10% above its own volume-weighted average price. Paying up here on a signal family that has lost money every single time this book has traded it is how you turn a good story into a bad entry. We would rather wait for a pullback toward the model's entry level than chase. || OPERATIONAL: Quality_Pass FALSE = hard block on a new entry. Independently fails the floor: MoS -32.51 -> val_adj -10 (floor); USC-3 mandates reading FailedBreakdown_Up_20D|buy at the -10 floor rather than the live -2; 76 - 10 - 10 = adj 56 vs the 65 bar (adj 64 even on the live -2 bias). VWAP_Distance +10.06%, the most extended print in the mandate. Re-arm unchanged (USC-3): raw >= 82 on a bullish-actionable print, or a PullbackBuy_VWAP confirmation. · news |
| MU | Micron Technology, Inc. | 935.39 | WATCH | 935.39 | 906.89 | 965.55 (+3.2%) | 1035.93 (+10.7%) | 989.20 (+5.8%) | Still not buying Micron. Last night it was the closest this book had come to an acquisition in twelve sessions; tonight the signal downgraded from buy to watch and the stock slipped 0.3% while almost everything around it rallied - it opened at and closed at giving back the whole day. It remains the best-quality candidate on the board and the only one that passes our financial screen, with the cleanest trend structure of any name we follow, so it stays on the shortlist. But a name that fades a intraday range on an up day for the sector is telling us to wait. || OPERATIONAL: label whipsaw continues BUY->WATCH->WATCH->WATCH->BUY->WATCH, failing the 'stable or fresh' test. adj 61 - 6 - 10 = 45 vs the 65 floor (val_adj -6 degree-scaled off MoS -1.107, the mildest in the mandate; note B does NOT suppress it - DCF 772.94 = 1.21x price, a sound anchor unlike LITE 17.7x / TSM 9.0x). Momentum pillar collapsed 26 and Volatility pillar 21 (Bearish) on the reversal day. Re-arm per USC-3 unchanged. |
| NVDA | NVIDIA Corporation | 227.98 | HOLD | 227.98 | 211.56 | 243.72 (+6.9%) | 280.45 (+23.0%) | 241.32 (+5.9%) | Keeping 177 of the 253 Nvidia shares - the core of the position - after trimming 30% into the post-earnings jump. The signal engine now rates Nvidia the strongest name we follow, with a trend score of 93 out of 100 and conviction up from in a single session, and it sees the stock reaching about over four weeks. The stop rises to, roughly 7% below the current price, so the retained shares carry about of defined risk. || OPERATIONAL: stay-decision recorded so the grader prices the retained 70%. Trail raised 207.19 -> 211.56 = the printed chandelier stop, 1.91x ATR clearance, legal (below the 227.98 close, above the standing stop). Pillars T93/M58/V38, Consolidation_Bullish, RSI 61.3, VWAP_Distance +4.94%, volume 297.2M - the heaviest print in the mandate two sessions running. · news |
Outcome & track record
Accuracy at the time of this record.
34 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.