US — End-of-Day Verdict
1 ACQUIRE (QUEUED AAPL gtc_stop @341.50, adj 73) | DISPOSE fired: KLAC -500 (50%, keep 500), MU -50 (50%, keep 50), META exit 60 | DEFERRED by cap: NBIS -150 (50%), NVDA -50 (50%) | WITHDRAWN: GOOGL full exit (6 sessions refuted - US-17/US-22, stale overlay) + AAPL @316.32 limit (US-8 family flip) | HOLD LLY (trail raised 1155.71), OKTA (124), SPCX (muted 135), GOOGL (new stop 319.00)
Recommendations
1 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 |
|---|---|---|---|---|---|---|---|---|---|
| KLAC | KLA Corporation | 190.80 | SELL 50% | 190.80 | 171.00 | 193.67 (+1.5%) | 195.71 (+2.6%) | 200.38 (+5.0%) | Halving the book's largest position (28% of the portfolio) in KLA after the semiconductor-equipment maker's fiscal Q4 beat was sold hard: results and guidance topped estimates yet the shares fell 6.2% in the session and a further 7.5% after hours, which erases the entire gain on our cost. The chart confirms the de-rating - a second consecutive bearish signal print and a decisive break of the 198.40 stop, with the 20-, 50- and 100-day averages all lost. Demand is not the question here; the multiple is. We keep half for the trend and reset the stop to the 200-day average. Ops: escalated from the unexecuted 30% card per US-19 hard-tick carve-out (2nd decisive stop break is a new hard fact). Valuation amplify SUPPRESSED by G2 (DCF 93.26 unreliable vs Rev_Growth 33.5%). Scorecard cell RallySell_BearTrend|sell is -5 (n=5, hit_1w.25) and argues AGAINST - overridden on concentration + realised post-print de-rating, not on a bearish forward. Est_* computed pre-print, stale to the upside. REPRICE at the open (~176-177 expected, >1.5% gap); execute at market, do not cancel. Core 500, reset stop 171.00 (EMA200). Proceeds at close ref. · news |
| MU | Micron Technology, Inc. | 820.53 | SELL 50% | 820.53 | 770.00 | 826.51 (+0.7%) | 899.73 (+9.7%) | 962.99 (+17.4%) | Halving Micron because the memory thesis itself cracked, not just the chart. SK Hynix is slowing its high-bandwidth-memory expansion and China's ChangXin Memory completed its Shanghai IPO on the same day, reviving fears that Chinese supply undercuts pricing - Micron fell 8.9% (13.6% intraday) and the memory complex has now formally entered a bear market, giving back 30-50% of a 600% run. Our standing research condition for exiting this name was 'HBM pricing or demand cracks, or a decisive stop break'; both legs triggered at once. Because this is a sector-wide crack rather than a company-specific one, we halve and keep the core rather than exiting. Ops: overlay invalidation TRIGGERED, news-override force - this is the basis, NOT the NO TRADE label (US-17/US-29). L2 does NOT fire (12wk forward +17.4%, strongly positive); carried on US-23 (forward lags a regime shift) + the triggered override. Sizing 50% per US-12/US-22 sector-crack doctrine. Est_12W band [436-1836] is near-uninformative. Supersedes the 07-28 conjunctive deferral. Core 50, reset stop 770.00 (EMA100). Proceeds. · news |
| META | Meta Platforms | 593.41 | SELL ALL | 593.41 | 620.00⚠ | 592.40 (-0.2%) | 568.51 (-4.2%) | 592.24 (-0.2%) | Exiting Meta in full ahead of tonight's Q2 print. The position is 6.6% underwater, has spent five sessions below our 620 stop, trades below every major moving average, and the model's own forward view is negative at one, four and twelve weeks. The signal has now turned explicitly bearish after three quiet sessions. The options market implies a ~7% move on tonight's results, and the setup rhymes with last quarter, when Meta beat estimates and the stock still fell 9% the next day because capital-spending guidance kept rising - it now stands at -145bn for 2026. There is no modelled upside to justify holding a loser through a coin-flip event. Ops: L2 fires cleanly (decisive stop break + negative 12wk, both legs). 5th re-fire of an unexecuted card (L3). Signal cell FailedBreakout_Down_20D|sell is +2 (n=10, hit_1w.60) - the book's most reliable sell cell, supports. PROFIT_ARMED is the known loss-artifact (Gain_ATR -2.00), ignored. Promoted over NBIS on binary-event timing. Proceeds. · news |
| AAPL | Apple | 340.08 | BUY — QUEUED @341.50 | 341.50 | 317.61 | 342.87 (+0.4%) | 342.96 (+0.4%) | 357.15 (+4.6%) | Queuing a half-sized starter in Apple as the defensive-megacap destination for money leaving the chip complex. On Tuesday Apple briefly became only the second company ever worth trillion, touching a record 342.89 while semiconductors were routed, and it is the only name in our entire buy list that passes the quality screen (F-Score 9). The order is a stop above the market, so it fills only if Apple breaks to a fresh all-time high - never on weakness. Risk is deliberately capped: half-normal size, a 7% stop, roughly at risk. Ops: signal family MomentumContinuation_Up is the book's WORST buy cell (bias -10, n=6, hit_1w.0, hit_4w.0); admissible ONLY because raw conviction 93 clears the US-16 raw->=85 bar for that family. adj 73 = 93 -10 val -10 cell. eD +0.18 = at entry, RSI 68.9 < 75, no chase. Size x0.5 (MoS -3.16). HONEST CAVEATS: model 1wk/4wk points are essentially FLAT (+0.4%) against a -7.0% stop - reward/risk only works on the 12wk leg; AAPL reports FY26 Q3 07-30 AMC and FOMC lands 07-29, so a fill means holding through two binary events. 12wk 357.15 exceeds the 208.95 DCF anchor. Re-verify family + quality gate at any fill (US-8). expires 2026-08-03. · news |
| NBIS | Nebius Group, N.V. | 169.69 | NO ACTION | 169.69 | 190.00⚠ | 176.68 (+4.1%) | 193.44 (+14.0%) | 257.70 (+51.9%) | A 50% de-risk in Nebius is warranted but deferred to the next session purely because three larger risks ranked ahead of it tonight. The stock fell another 9.7% to sit 10.7% below our 190 stop and 19.8% below our cost, the signal flipped to an outright sell, and the press now reports that Meta - its largest customer - is building its own compute business, the exact scenario our research flagged as thesis-breaking. A new Sell rating landed with a target against a street consensus. We size at half rather than exiting because this is a sector-wide neocloud unwind, and the twelve-week forward is still strongly positive. Ops: DEFERRED BY THE <=3 CAP, not optional advice - card #1 next EOD. Engine Sell_Rank 2, Portion 88.52% sized down to 50% per US-12/US-22. L2 does NOT fire (12wk +51.9%); basis is US-18's second branch (close below invalidation) + US-23. Overlay NBIS override adjudicated NOT-yet-triggered (no confirmed contract impairment). Cell bias -5 argues against. Q2 possibly 07-29 (unconfirmed). Stop stays 190.00 - broken, deliberately NOT lowered; core stop resets to 153.50 (EMA200) on execution. · news |
| NVDA | NVIDIA Corporation | 197.01 | NO ACTION | 197.01 | 201.00⚠ | 195.81 (-0.6%) | 194.64 (-1.2%) | 197.65 (+0.3%) | Nvidia is the engine's top-ranked sell and a 50% trim is due, but it is the smallest dollar risk of five firing candidates and gets deferred a third time by the three-card cap. The signal is an active strong sell, the stock sits below the broken 201 trail and below its 20-, 50- and 100-day averages, and the model's forward view is flat at every horizon - the position carries full AI-complex beta for no expected return. The $250bn OpenAI financing backstop keeps circular-financing worries alive, though Wedbush and Morningstar both call the fear overblown. Ops: DEFERRED BY THE <=3 CAP - card #2 next EOD, 3rd consecutive deferral (systemic finding #1). Armed (T12_Progress 99.68) + tick (c) dead forward (Est_12W 197.65 < 1.02x close) + tick (d) 3 prior NO TRADEs -> profit_take_2 = 50%, never 100% from a profit-take. Cell FailedBreakout_Down_20D|sell +2 supports. Shared with ussemicon -> execute ONCE (L7). Stop stays 201.00 - broken, deliberately NOT lowered. · news |
| LLY | Eli Lilly & Co | 1220.66 | HOLD | 1220.66 | 1155.71 | 1225.68 (+0.4%) | 1199.54 (-1.7%) | 1298.82 (+6.4%) | Holding Eli Lilly and raising the trailing stop to 1155.71. It is the one unambiguously healthy chart in the book - up 1.9% on a day the AI complex broke, trading above every major moving average, with conviction improving from and a twelve-week forward 6.4% above the current price. Ops: armed (T12_Progress 93.98) but NOT fired - conviction ROSE 14 pts (far from a -20 collapse) and tick (c) is alive. Armed-but-not-fired -> mandatory trail-raise, 1133..71 (model Stop_Price). Q2 08-05 BMO. |
| OKTA | Okta Inc. | 136.21 | HOLD | 136.21 | 124.00 | 136.87 (+0.5%) | 136.67 (+0.3%) | 143.41 (+5.3%) | Holding Okta on an unchanged 124 stop. The position is 2.6% underwater, the signal has gone quiet on a consensus conflict, and the forward view is modestly positive - nothing here justifies action either way. Ops: PROFIT_ARMED is the loss-artifact (Gain_ATR -0.50) - declined. Bare NO TRADE is never a dispose basis (US-17). Below EMA20 (138.51) but far above EMA50 (125.29). |
| SPCX | Space Exploration Tech Corp | — | HOLD | — | 135.00 | n/a | n/a | n/a | Holding SpaceX on the manual 135 stop, with two dates now close enough to matter: the maiden quarterly report on 4 August after the close, and the IPO lock-up expiry on 6 August, when roughly 911 million shares - some -123bn of stock - become eligible to trade, two sessions later. Ops: permanently muted on signals (private/unquoted, no public close) - news-assessed per the edge case. News read: CONCERN, calendar imminent. Index-eligibility mechanics open 08-07. A scan Close < 135 auto-unmutes and triggers a full-exit review. |
Outcome & track record
Accuracy at the time of this record.
99 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 →
-0.132 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.
23 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.