Recommendations
1 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 |
|---|---|---|---|---|---|---|---|---|---|
| AMD | Advanced Micro Devices, Inc | 545.09 | BUY — QUEUED @532.61 | 532.61 | 434.64 | 591.39 (+11.0%) | 728.55 (+36.8%) | 600.69 (+12.8%) | AMD broke out to a 52-week high on its third straight bullish signal, with conviction climbing to 90, and it is the only name in tonight's buy pool that passes our financial-quality screen. But the whole chip sector rose names today with no company-specific news behind AMD's move, and AMD is the highest-beta name in that group, so we queue a limit order at the model's own entry price 2.3% below the close rather than chase the breakout. OPERATIONAL: Q4 satisfied ( consecutive bullish prints); Quality_Pass TRUE (sole pass); val_adj 0 per note B/P (DCF 171.74 = 31% of price, broken artifact) but size x0.5 retained (MoS <= -); no USC- overlay -- scoped to FailedBreakdown_Up_20D, this is BreakoutUp_52W with no scorecard cell (bias 0); adj 90. NO-CHASE BINDING: Entry_Distance_ATR -0.48, close at 73% of the 527.65-551.42 day range. PROVENANCE: the 09-@512.50 buy-stop is WITHDRAWN not carried -- AMD closed 545.09, 6.36% through it; repriced to the live Entry_Price. R4/S VEHICLE CONFLICT: Entry_Price 532.61 sits BELOW the close so a buy-stop is mechanically impossible (fills instantly = the chase); a limit is the only order expressing the printed entry, accepting R4's low-fill-probability risk. S BINDS: 0.9534 x 977,804.53 x 0.20 = 186,447.77, x0.50 = 93,223.88 ->, but the stop is 97.97/share (-18.4%, 3.76xATR) = risk = 1.75% of book (~22x this book's norm); capped at a stated 1.0%-of-book defined-risk budget ->, risk = 0.99%. V4 R:R 2.00:1 PASS; S7 PASS (12wk 600.69 > entry). R: the 19/20 cluster is ONE candidate -- AMD taken as its best-quality member (Buy_Rank 1, sole Quality_Pass). ESTIMATE CAVEAT: L1 records BreakoutUp_52W 4wk targets overshoot +55-75%; 12wk 600.69 < 4wk 728.55 (W8(c) inversion, th session); computed 1wk 591.39 EXCEEDS the engine's own Est_High 586.97. WITHDRAW LINE (Q2): auto-withdraw before 09-22 if conviction < 65, the label leaves the bullish family, or a settled close below EMA50 488.87. f=0.30 (L5, calib_1w -0.226 degenerate). Counterfactual no_chase_queue n=5 +2.88%/1wk [cost] cited advisory-only. · news |
| NVDA | NVIDIA Corporation | 219.34 | SELL 30% | 219.34 | 209.61 | n/a | n/a | 220.20 (+0.4%) | We cut a third of the NVIDIA position because the price has now closed below its trailing stop for six straight sessions, even though the stock rose today. This is a partial de-risk, not an exit -- three of the four reasons we had for selling were repaired in a single session, so we are selling only the portion our pre-committed rule still calls for and holding the rest with a raised stop. OPERATIONAL: Q2 honoured BOTH ways -- the 09-16 card's written line 'close >= EMA50 (~214.7) withdraws tier 2 ' FIRED (settled 219.34 > EMA50 214.84) so tier 2 is WITHDRAWN; close did NOT reach 221.79 so tier 1 stands. V1 HONESTY: basis decayed to ONE leg -- EMA50 reclaimed after closes below, NoTrade cv0 leg gone (now AVOID cv57), negative-12wk leg gone (220.20 now +0.39% ABOVE close); only theth sub-221.79 close survives, while the name sits above EVERY EMA with Trend Bullish. W2 AMBIGUOUS: +2.54% vs a +2.53% board median -- precisely with the tape, neither relative strength (would withdraw) nor failure-to-rise (would escalate). W5 re-validation done (>= re-fires, not-armed, trend-intact): USC-5 agrees with the written line -- de-risk to the technical's own tier, no further. U4: PROFIT_ARMED discarded (underwater -3.29%, Gain_ATR -1.28; arming leg is Close/Target_12Week, not gain). T4 bars tick (c) (standing level, not a transition); V bars 12wk origination (11/20 rows print T12 < Close = regime noise). STOP RAISED 205.91 -> 209.61 (EMA100); U2 clearance 1.67xATR PASS -- EMA20 (0.21 ATR) and EMA50 (0.77 ATR) rejected on the same floor. P2 does not block: witching 09-18 and the 09-21 rebalance are market-structure, not name-binary. peak_stop stays 221.79 until the cut fills; ADR-0020 erosion improves to -5.49%, no firing. TRIGGERS re-armed both ways: close >= 221.79 -> tier 1 WITHDRAWN, NVDA becomes a clean hold; close < 209.61 -> further 50%; close < 197.31 -> full exit. U8: no Target_Price on this row so 1wk/4wk are unstateable; band 212.21-232.91. Proceeds, realized - (-3.29%) vs cost 226.80. Scorecard FailedBreakout_Down_20D|sell n=2 report-only, cited not applied. The 'black sheep of the chip rally' thesis is dated 2026-07-06, ten weeks outside the bounded window -- background only, not scored (U). · news |
| ASML | ASML Holding NV | 1629.67 | HOLD | 1629.67 | 1622.58 | n/a | n/a | 1348.57 (-17.2%) | We are holding ASML rather than selling it: the stock reclaimed the stop level that was the only reason to exit. The underlying picture is still weak -- it trades below three of its four moving averages and lagged today's sector rebound -- so the same stop stays armed and a close back below it exits the whole position. OPERATIONAL: exit WITHDRAWN tonight (see reconcile line). Stop 1,622.58 re-armed unchanged; clearance only 0.13xATR (U2 floor breach, flagged). 12wk -17.25% but V bars it from originating (11/20 rows print T12 < Close). U4: PROFIT_ARMED discarded, position -10.80% / Gain_ATR -3.71. U8: no Target_Price so 1wk/4wk unstateable; band 1,592.67-1,780.23. 12wk exceeds DCF anchor 827.05. · news |
| TSM | Taiwan Semiconductor Manufacturing C | 430.26 | HOLD | 430.26 | 415.38 | 436.57 (+1.5%) | 451.30 (+4.9%) | 422.78 (-1.7%) | Taiwan Semiconductor is our strongest holding tonight -- conviction jumped from and it outperformed a strong sector day -- so we hold all three shares and leave the stop where it is. OPERATIONAL: HOLD cv76, Trend Bullish, +3.00% vs a +2.53% board median (W2 passes on the up leg); close above EMA20 422.46 / EMA50 419.94 / EMA100 409.71 / EMA200 380.99. NO TRAIL RAISE (U2): EMA20 gives 0.74xATR and EMA50 0.98xATR clearance, both below the 1.0 ATR floor; the held 415.38 line carries 1.41xATR. The engine's printed Stop_Price 409.22 is looser than the line already held, so not adopted. T5: at any percentage tier is a stub -- hold-or-exit-whole, never sliced. U4: PROFIT_ARMED discarded (underwater -1.77%, Gain_ATR -0.74). 12wk 422.78 below close but V bars origination. TRIGGER re-armed unchanged: settled close < 415.38 -> full exit of all at the next open. 12wk exceeds DCF anchor 45.44. f=0.30 (L5). · news |
| — | ASE Technology Holding | — | NO ACTION | — | — | n/a | n/a | n/a | ASE Technology produced the second-strongest buy signal on the board, but it fails our financial-quality screen, which blocks new positions outright, and this was its first bullish print after three flat sessions. We do not buy it. OPERATIONAL: Quality_Pass FALSE = hard gate on new entries. Q4 FAIL: WATCH -> 56 -> STRONG BUY 86 is a single print. USC-: FailedBreakdown_Up_20D is this book's/ stop-first family, overlay -10 applied (the V USC-/scorer -2 disagreement is non-deciding here -- the gate blocks independently). R: part of tonight's 19/20 single-catalyst cluster, not an independent candidate. val_adj 0 (no DCF anchor). +5.38% on the sector print. |
| — | Marvell Technology | — | NO ACTION | — | — | n/a | n/a | n/a | Marvell held a buy signal for a second session, but it fails our financial-quality screen, so it cannot be a new position -- and today's move was part of a sector-wide rally rather than anything specific to Marvell. OPERATIONAL: Quality_Pass FALSE = hard gate. Q4 PASSES (BUY 84 -> BUY 76, two consecutive) but the gate blocks independently. USC- overlay -10 (/ stop-first family); V disagreement non-deciding. R: one of six names in tonight's single-catalyst cluster -- AMD taken as its best-quality member. val_adj 0 per note B/P (DCF 63.95 = 27% of price, broken artifact). +4.81% on the sector print; conviction actually FELL into the rally. |
| — | Tesla Inc | — | NO ACTION | — | — | n/a | n/a | n/a | Tesla flipped to a buy signal today, but its own twelve-week projection sits below the price we would pay, and it fails our financial-quality screen. A trade whose own forward view is negative is not a trade at any conviction. OPERATIONAL: THREE independent blocks -- (1) S7: Target_12Week 348.66 < entry 366.20 (-4.79%); (2) Quality_Pass FALSE = hard gate; () Q4 FAIL: NO TRADE x -> BUY is a single print. USC- overlay -10; adj 64 is below the 65 floor anyway. val_adj 0 per note B/P (DCF 34.25 = 9% of price, broken artifact). R cluster member. Also one of the 11/20 rows printing T12 < Close (V). |
| — | Qualcomm Incorporated | — | NO ACTION | — | — | n/a | n/a | n/a | Qualcomm turned to a buy signal and does pass our quality screen, but its conviction is below our threshold once we mark down what we would be paying versus what the business is worth, and the stock is already overbought. We pass. OPERATIONAL: adj 63 - = 57, below the 65 floor. val_adj - is APPLIED not suppressed -- P: DCF 146.90 is 78% of the 188.71 price, a USABLE anchor, unlike the broken artifacts on AMD/MRVL/TSLA. Quality_Pass TRUE (flipped False->True on the 09-17 regeneration and now Q5-verified, but moot). Q4 FAIL: WATCH -> NO TRADE 0 -> BUY 63 is a single print. RSI 67.90; Entry_Distance_ATR +0.51 -- price must RISE 2.29% to 193.04 to trigger. BreakoutUp_20D|buy has no scorecard cell (bias 0). Stop 160.95 is 3.76xATR wide -- S would bind hard if it ever cleared. |
Outcome & track record
Accuracy at the time of this record.
55 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.
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.