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 |
|---|---|---|---|---|---|---|---|---|---|
| SNDK | Sandisk Corporation | 1887.04 | BUY-STOP — QUEUED @1846.17 | 1846.17 | 1495.87 | 2056.34 (+11.4%) | 2546.75 (+37.9%) | 2359.85 (+27.8%) | SanDisk has now set up its early-entry pattern in two of the last three sessions at high conviction, and Rosenblatt started coverage at Buy with a target, citing long-term customer agreements that cover about 65% of fiscal-2028 output. It closed above the model's trigger after a 6.8% jump, so we place a small order at the trigger level rather than paying up at the open. OPERATIONAL: anticipate slot (4th slot, ADR-0012): BUY - ANTICIPATE 96 (09-19) -> BUY 68 (09-22) -> BUY - ANTICIPATE 88 (09-23) =; cv 88 + val_adj -8 (MoS -1.55 recomputed at the 1887.04 close, T4) + bias 0 (AnticipationUp_VCP|buy n=1 report-only) = adj 80 >= 55 slot floor; Quality_Pass True. Beats AMD (also, adj 65) for the single slot. No-chase: ED -0.35 (beyond the -0.25 bar) with the close 1887.04 within 0.19 ATR of the 1909.48 high, so a buy-stop at 1846.17 would fill at the open as a chase; ASSUMPTION: queued as a GTC LIMIT at the printed Entry_Price instead (provenance kept, fills only on a retest of the pivot). Volume: 50% x equity 1,050,946.68 x 20% = 105,094.67 x 0.75 (-3 < MoS <= -1) x 0.5 slot = 39,410.50 / 1846.17 = (, 3.7% of equity). Stop 1495.87 (scan, 3.0 ATR); risk U2 passes (12wk 2359.85 = 1.28x entry). Forward-curve inversion: 4wk 2546.75 > 12wk 2359.85. Counterfactual `anticipate` n=10 reads 'saved' (advisory). Expires 2026-09-25; withdraw earlier if conviction < 65 or the row stops printing a bullish label. · news |
| DELL | Dell Technologies Inc | 548.92 | HOLD | 515.65 | 517.26 | 529.61 (+2.7%) | 562.19 (+9.0%) | 782.21 (+51.7%) | Dell fell 4.6% the session after we banked 30%, so the trim filled at 569.80 and has already saved about; the model now wants to buy it back on a pullback, but after the valuation haircut the signal is under our bar, so we simply hold the remaining. OPERATIONAL: BUY - ADD 71 PullbackBuy_VWAP (limit 515.65, ED -1.06) - val_adj -10 (MoS -2.57) = adj 61 < 65 -> NO ADD. Also UT-10 caution: no pyramid back into a position trimmed one session earlier. Scan Held/shows 908 (stale; JSON governs, Y4). Not armed (Gain_ATR 2.27, no PROFIT_ARMED). Stop 517.26 held (chandelier 486.41 looser). Decisive close < 517.26 with Target_12Week < close -> exit per L2. · news |
| MU | Micron Technology, Inc. | 1096.16 | HOLD | 1096.16 | 998.56 | n/a | n/a | 1318.00 (+20.2%) | Micron rose another 5% to a record 1096 into its 09-30 results; the model flags it as overbought and says don't add, which we respect, but nothing says sell, so we hold and lift the stop to lock in most of the gain. OPERATIONAL: AVOID 65 (ReversalDown_Overbought) is a do-not-add label, not a sell-pool label; not armed (T12 83.17, Gain_ATR 2.20). Y5: stop 955.58 -> 998.56 (2.0xATR chandelier off the close; scan's 1239.35 is a short-side stop, ignored). FQ4 09-30 AMC: no adds (L6), catalyst-aware hold. A SELL - REDUCE/WARNING print or close < 998.56 with 12wk < close -> trim 30% (23). · news |
| TSM | Taiwan Semiconductor Manufacturing C | 452.00 | HOLD | 446.74 | 431.52 | 465.17 (+4.1%) | 508.18 (+13.8%) | 477.85 (+7.0%) | TSMC added 1.5% and conviction stays high; the position is now far enough in profit to count as 'armed', so we tighten the stop rather than sell. OPERATIONAL: PROFIT_ARMED (T12 94.59) with Gain_ATR 1.37 >= X2 materiality 1.0 -> armed; ticks: no other Exit_Warning, conv (-5), 12wk 477.85 > 1.02x close 461.04, no NO TRADE -> NOT FIRED -> mandatory trail-raise: 425.52 -> 431.52 (2.0xATR chandelier). Any tick -> trim 30% (19). |
| NVDA | NVIDIA Corporation | 228.87 | HOLD | 228.87 | 218.10 | 232.10 (+1.4%) | 239.64 (+4.7%) | 243.78 (+6.5%) | Nvidia edged up again with conviction rising to 78, so we hold and tighten the stop. OPERATIONAL: PROFIT_ARMED on T12 93.88 but Gain_ATR 0.38 < X2 materiality 1.0 -> inert. Y5: 215.66 -> 218.10 (scan = chandelier, 2.0 ATR). Close < 218.10 with 12wk < close -> exit 48. |
| PLTR | Palantir Technologies Inc. | 184.99 | HOLD | 184.99 | 176.76 | 188.43 (+1.9%) | 196.47 (+6.2%) | 238.81 (+29.1%) | Palantir firmed 1% and conviction ticked back up to; we hold with the stop unchanged. OPERATIONAL: not armed (Gain_ATR 1.60, T12 ). Stop 176.76 held (scan 182.03 only 0.52 ATR below close, rejected per V5; chandelier 173.51 looser). Decisive close < 176.76 -> trim 116. |
| AAPL | Apple | 339.75 | HOLD | — | 326.25 | n/a | n/a | 379.38 | Apple's 30% trim filled at 340.33 and the stock closed nearly flat at 339.75; the model printed no signal today, which is a watch item rather than a sell, so we hold the remaining on a raised stop. OPERATIONAL: 1st NO TRADE ('Consensus conflict + low conviction'); not armed (T12 89.55 < 90, Gain_ATR 1.28, no PROFIT_ARMED print) so tick (d) is inert (Y6). Stop 325.59 -> 326.25 (chandelier). A 2nd NO TRADE while armed -> trim 30% (36). |
| CRM | SalesForce Inc | 233.28 | HOLD | 233.28 | 238.25⚠ | 238.71 (+2.3%) | 251.39 (+7.8%) | 317.87 (+36.3%) | Salesforce slipped another 1.3% and its signal strength has collapsed to 29, but it is still above our written hard-exit level and the model's 12-week view is positive, so we hold on a tight leash. OPERATIONAL: monitor breach 233.28 vs alert 238.25 (-2.09%, 0.55 ATR); intraday low 229.81 came within 0.83 of the 228.98 hard exit. Conviction (-23) = deteriorating signal, but ADR-0012 tick (b) needs an armed position (Gain_ATR -2.69; Y6/UT-9 inert). L2 needs both legs; 12wk 317.87 > close. Z3/T1 written trigger stands: close < 228.98 OR (close < 238.25 AND Target_12Week < close) -> SELL market. ADR-0020 clear (238.25 vs peak 251.77 = 5.4%, cut 09-). · news |
| LITE | Lumentum Holdings | 945.67 | NO ACTION | 945.67 | 914.19 | n/a | 1071.58 (+13.3%) | 1095.19 (+15.8%) | Lumentum is the book's #1-ranked BUY but fails the quality screen (thin free cash flow), so it stays out. OPERATIONAL: cv 78 - 10 = adj 68 would clear 65; Quality_Pass False blocks new entries. Ledgered as a gate-cost counterfactual. |
| NBIS | Nebius Group, N.V. | 236.12 | NO ACTION | 236.12 | 228.26 | n/a | 267.56 (+13.3%) | 254.29 (+7.7%) | Nebius ranks #2 on the buy list but fails the quality screen, so it stays out. OPERATIONAL: adj 66 would clear the floor; Quality_Pass False (hard gate on new entries). V4: MoS is a degenerate anchor. |
Outcome & track record
Accuracy at the time of this record.
43 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 →
0.138 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.
91 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.