Recommendations
2 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 | 187.27 | SELL 50% | 187.27 | 184.00 | 189.10 (+1.0%) | 184.00 (-1.7%) | 179.06 (-4.4%) | KLA has closed decisively through its stop at 187.27 - 3.16% below the 193.38 line - on a second consecutive bearish print, and the model's own twelve-week projection of 179.06 now sits BELOW today's price. When the risk line breaks and the forward view turns negative at the same time, that is an exit, not a dip. We take half off and keep half, because the business case is intact: the slide is China export policy rather than fundamentals, advanced-packaging revenue is still guided to grow over 70% this year, and the analyst average sits at against tonight's OPERATIONAL: escalation trigger pre-declared in the 08-18 card (decisive settled close below 193.38) has fired - tier escalates 30% -> 50%. Three independent bases: L2 (stop break + negative 12wk), ADR-0020 Trail Erosion session 16 (peak stop 272.00 vs 193.38 = -28.90%, never once cut - this cut discharges the clock), and ADR-0012 profit-take (armed on T12_Progress 104.59 + tick (c)). Valuation amplify to 100% SUPPRESSED by G2: T12 179.06 vs DCF 61.81 = 2.90x overshoot. Retained re-based to a 184.00 stop (below the 185.38 session low); a settled close below 184.00 exits the remainder. 08-18's 300-share card at 194.79 was never executed - the delay cost on. · news |
| NET | Cloudflare, Inc. | 291.88 | SELL 30% | 291.88 | 283.72 | 292.93 (+0.4%) | 296.04 (+1.4%) | 308.84 (+5.8%) | We wrote down a stop-loss trigger for Cloudflare a session ago and it has now fired on both limbs: two consecutive closes below 301.75, and a decisive close below the prior session's low. So we honour it and take a third off. The business is not the problem - Q2 revenue grew 36% to, full-year guidance was raised, and roughly 2 million developers joined as AI agents drove adoption - which is exactly why stay on and the twelve-week view of remains positive. This is stop discipline, not a verdict on the company. OPERATIONAL: pre-committed trigger from the 08-18 card, both limbs satisfied (301.36 then 291.88; 291.88 is 1.9% below the 297.63 low of 08-18). L3 - honour the trigger written when the evidence was cooler; signal actually IMPROVED intraday (NO TRADE cv0 -> WATCH cv29 -> WATCH cv54) which is why the tier stays at 30%. Valuation amplify SUPPRESSED by G2: T12 308.84 vs DCF 29.51 = 10.5x, the book's largest overshoot. Retained re-based to the model's 283.72 (below the 285.72 session low) - the size cut earns the re-base. Note P/S ~42.5x vs industry ~1.8x. Duplicate NET row in usportfolio.json, 17th session - de-duplicate. · news |
| NBIS | Nebius Group, N.V. | 223.90 | SELL 30% | 223.90 | 213.00 | 228.82 (+2.2%) | 256.92 (+14.7%) | 263.80 (+17.8%) | Nebius broke its stop by a full 5% today, and for the first time in this drawdown the cause is the company's own balance sheet rather than the market's mood: it announced a billion convertible note offering - its third in under a year and its largest - and the shares fell nearly 14% intraday as investors weighed the data-centre build-out against the dilution. We take a third off to respect the broken line. We keep two-thirds because the growth case is unchanged and the model still projects in twelve weeks, roughly 18% above tonight's price. OPERATIONAL: decisive settled stop break (223.90 vs 235.69 = -5.00%, low of day 213.75), 3rd consecutive NO TRADE. L2 NOT satisfied in full - the 12wk view is positive - so this is a 30% tier, not a full exit. Valuation amplify SUPPRESSED: DCF_per_share is BLANK, so there is no anchor against which G2 could be cleared; MoS -11.75 originates nothing. Retained re-based to 213.00 (just under the session low). 12wk band [133.37 - 650.16] is the honest statement of model uncertainty here. Overlay Meta-deal override intact and untouched. SHARED WITH ustech - execute once (L7). · news |
| LMT | Lockheed Martin Corporation | 589.15 | BUY — QUEUED @586.92 | 586.92 | 568.35 | 591.82 (+0.8%) | 599.79 (+2.2%) | 618.39 (+5.4%) | Lockheed is the one name we own whose earnings actually benefit from the macro that is dismantling the rest of this book. It just took one of the largest single contract awards in its history - a seven-year PAC-3 interceptor deal worth up to billion - closed the quarter with a record billion backlog on billion of new orders, and raised full-year guidance to -in earnings and -7.2 billion of free cash flow. The shares pulled back to their session low today, and we are bidding the pullback rather than chasing: a limit at, with analyst targets sitting -above it. OPERATIONAL: PullbackBuy_VWAP, the book's ONLY positive scorecard cell (+2, n=22, hit_1w.591). Entry_Distance_ATR -0.14 = inside the no-chase bar. Quality gate PASSES (F 8, Quality_Rank 19/52); DCF 692.84 sits ABOVE the 12wk 618.39 so the projection is inside its anchor - no DCF annotation. Size x0.75 (-3 < MoS -1.76 <= -1). REPRICED from the withdrawn 08-18 @583.11 per the provenance rule - not a carry. Closed at the session low (O 605.00 / L 589.04 / C 589.15); the limit sits 0.32% below that low so a fill needs one more down-tick. Auto-withdraw if conviction < 65 or the family flips off PullbackBuy_VWAP. Ex-dividend on 09-01. · news |
| LLY | Eli Lilly & Co | 1280.34 | ADD — QUEUED @1240.97 | 1240.97 | 1159.93 | 1291.78 (+4.1%) | 1328.32 (+7.0%) | 1421.04 (+14.5%) | Lilly is the receiving end of the exact rotation tearing through the rest of this portfolio - money leaving expensive technology for companies with real earnings growth - and today it made a fresh 52-week high, up 4.5%, on the back of Q2 revenue up 48% to billion, Mounjaro sales up 91%, raised guidance, and a market capitalisation through trillion, the first pharmaceutical company ever worth that much. We add a small tranche, funded directly by tonight's semiconductor disposals, and we bid a pullback at rather than paying It is expensive on intrinsic value, which is why the size is deliberately small. OPERATIONAL: BreakoutUp_52W, Buy_Rank 2/51, cv60 -> 78. Entry_Distance_ATR -0.86 = extended, therefore QUEUED not bought (L1). ORDER-TYPE CORRECTED: the model printed 'Buy Stop @ 1240.97' but that level is BELOW the 1280.34 close, so a stop would fill instantly at market - the same defect that gapped VIST's 08-17 anticipate through by 2.9%. Rendered as a GTC buy-LIMIT; 1240.97 is exactly today's session low (1240.54). Systemic #3. val_adj -10 CLAMPED from -14; size x0.75; quality gate passes at Quality_Rank 4/52. Model Stop_Price 1149.45 REFUSED as a US-29 loosening artifact - the standing 1159.93 trail applies to all. 12wk 1421.04 exceeds DCF anchor 507.33 (2.80x). Position cap not binding (headroom). PROFIT_ARMED without a fired tick -> trail-raise path only, no trim. · news |
| GOOGL | Alphabet, Inc.- Class A | 344.72 | NO ACTION | 344.72 | 332.73 | 346.98 (+0.7%) | 337.49 (-2.1%) | 376.25 (+9.1%) | Alphabet has now gone six straight sessions without producing any tradeable signal at all, while sitting on a gain of 180% over our cost. That combination normally means taking a third off the table. It does not happen tonight only because three positions actually broke their stops today and loss-prevention comes first - a stock that rose 0.2% waits its turn. This is queue position 4, and the first card next session. OPERATIONAL: ADR-0012 fired - armed (Gain_ATR 31.06, T12_Progress 91.62) + tick (d), six consecutive NO TRADE cv0. NoTrade_Reason verbatim: 'Consensus conflict + low conviction'. Ticks (a)/(b)/(c) decline - T12 376.25 = 1.091x Close, alive. One tick -> 30% = @ 344.72 = realised approx +vs the 123.00 cost. DISPLACED by Step 7's loss-prevention-first ranking, NOT withdrawn and NOT optional-trim prose - ledgered direction:none/optional:true purely so the counterfactual grader prices the delay. Session 7 owed. Stop 332.73 stands (3.5% below); ADR-0020 erosion -7.06%, below the 15% fire threshold. |
| NVDA | NVIDIA Corporation | 217.56 | HOLD | 217.56 | 217.07 | 218.46 (+0.4%) | 220.33 (+1.3%) | 233.85 (+7.5%) | Nvidia reports Q2 earnings on 26 August - now exactly five trading sessions away - and our rule is not to cut a position into its own catalyst. Yesterday that trim was on the table because earnings were six sessions out; one day later the calendar has closed the window, so we hold all through the print on a trail at. The stock traded below that line intraday but settled above it, and a settled close is what counts. Consensus looks for roughly billion of revenue, up 96% year on year. OPERATIONAL: ADR-0012 IS fired - PROFIT_ARMED via T12_Progress 93.03 + tick (d), five consecutive NO TRADE cv0 - but SUSPENDED by the catalyst-aware hold (earnings 08-26 = sessions 08-20/21/24/25/26 = exactly 5). Release conditions unmet: no confirmed lower-high break, and no decisive stop break (low 216.76 was below 217.07 but the settled close 217.56 is above - L4 governs). REVERSES the 08-18 card, which explicitly noted the window did not yet bind. IF 217.07 breaks on a settled close the 30% fires immediately regardless of the print. Position is 0.23% off its line - a coin-flip held deliberately. GF Securities target raised on Vera Rubin demand. Avg_Cost unrecoverable (07-04 adjust, price:null). SHARED WITH ustech/ussemicon - execute once (L7). · news |
Outcome & track record
Accuracy at the time of this record.
144 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 →
-0.148 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.
52 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.