US — End-of-Day Verdict
ACQUIRE NO ACTION (0 of <=3) - all 4 bullish candidates quality-gate-blocked, 13th consecutive session with no admissible new BUY (US-20): WDS adj 79 (F=4), DDOG adj 64 (F=5), AMKR adj 64 (F=4), SNOW adj 50 (F=5, extended RSI 74). ANTICIPATE 0 (zero BUY - ANTICIPATE rows). | DISPOSE 2 of <=3: META full exit 60 @ ~539.03 - BreakdownDown_52W cv85 Sell_Rank 1, L2 fires both legs, Q2 EPS miss + capex to -145bn + FCF, 7th re-fire; NVDA trim 50 @ ~195.04 - 3rd consecutive STRONG SELL - REDUCE + broken stop + dead forward, keep 50, stop reset 188.00, execute ONCE vs ussemicon. | The session was a violent AI-complex REVERSAL (MU +18.4%, NBIS +27.1%, SNDK +26%, CRWV +21.5%, LRCX +18%) on a hedge-fund unwind completing + a Samsung tightening-supply warning + strong Microsoft earnings - which withdrew the MU trim and vindicated the 07-30 KLAC/NBIS withdrawals. | : buys sells free ->.
Recommendations
0 to acquire · 2 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 |
|---|---|---|---|---|---|---|---|---|---|
| META | Meta Platforms | 539.03 | SELL ALL | 539.03 | 620.00⚠ | 536.77 (-0.4%) | 547.54 (+1.6%) | 548.12 (+1.7%) | Meta's AI build-out is now consuming the cash the business generates, and the market has repriced it. Q2 revenue beat at 8bn (+28% YoY) but EPS came in at against a -7.22 consensus, free cash flow collapsed to from 4bn the prior quarter, and 2026 capex guidance was lifted to -145bn with no visible AI monetisation to fund it. The stock fell ~9-10% on the print and ten brokers cut price targets, Wells Fargo from. The chart agrees with the fundamentals: a 52-week breakdown at the highest sell conviction in the book, trading below every major moving average and 13% under its stop, with the forward view flat at all three horizons. We exit the position in full. OPS: 7th re-fire of this exit (L3) - the position has bled across six non-executions. STRONG SELL - REDUCE cv85 Sell_Rank 1 Portion 88.52%, 3rd consecutive escalating bearish print (cv42 ->, so US-18 is satisfied twice over. L2 fires both legs: decisive stop break (-13.06%, 7th session below 620) + dead 12wk (Est_12W 548.12 < 1.02x close 549.81). -15.11% underwater on the 635.00 cost; below EMA20 606.93 / EMA50 609.05 / EMA100 617.29 / EMA200 638.18. PROFIT_ARMED is the known loss-artifact (Gain_ATR -3.38), ignored. Valuation FAIR_VALUE MoS -0.92 -> no amplify (outside the <=-3 tier). BreakdownDown_52W|sell has no graded cell, so no bias applied. Execute at market on the 07-31 open; of proceeds. Stop kept at the broken 620 line for monitor continuity until the exit fills. · news |
| NVDA | NVIDIA Corporation | 195.04 | SELL 50% | 195.04 | 188.00 | 194.18 (-0.4%) | 192.96 (-1.1%) | 195.11 (+0.0%) | Nvidia has become the black sheep of its own sector. It is up just 3.2% year-to-date while AMD has gained 171%, Micron 305% and Intel 278% - and on a session when the whole memory and semi-cap complex ripped double digits, Nvidia managed only +2.6%. That is a position carrying full AI-complex downside beta without the matching upside, and our own forward model agrees: flat-to-negative at one week, four weeks and twelve weeks. We halve the holding and keep the core, because the Vera Rubin ramp this autumn is a real catalyst and some analysts still see - this is reducing risk, not calling the story over. OPS: 2nd re-fire (carded 07-30, unexecuted; repriced 190..04, +). 3rd CONSECUTIVE settled STRONG SELL - REDUCE (cv68 ->, Sell_Rank 3, Portion 88.52% - US-18 amply satisfied, and this is a HARD basis (stable active SELL-REDUCE + broken stop) so the US-19 soft-tick escalation guard does not bind. -2.97% below the 201.00 trail; below EMA20 202.12 / EMA50 203.55 / EMA100 200.68, clinging to EMA200 193.46 (+0.8%). ADR-0012 also armed (T12_Progress 99.96) + tick (c) dead forward (Est_12W 195.11 < 1.02x close 198.94) -> profit_take_2 50%, never 100%. Cell FailedBreakout_Down_20D|sell +2 (n=11, hit_1w.60) - the book's better sell cell, supporting rather than fought. Valuation GROWTH_PRICED_IN MoS -2.07: outside the <=-3 amplify tier, and G2 would suppress regardless (DCF 163.98 unreliable vs Rev_Growth 94.3%). Est_12W exceeds the DCF anchor - annotated, not modified. SHARED WITH ussemicon - EXECUTE ONCE (L7). Keep 50, reset the retained core's stop to 188.00 (a decisive break of EMA200). of proceeds. · news |
| KLAC | KLA Corporation | 180.33 | HOLD | 180.33 | 168.00 | 184.99 (+2.6%) | 188.52 (+4.5%) | 196.49 (+9.0%) | KLA bounced 6% off its 200-day line and the sell case did not survive the session. The engine flipped to a sell print, but every hard test fails: the stop is 11% away, the position is back in profit, and the forward view is positive at all three horizons, +2.6% / +4.5% / +9.0%. The fundamentals point the same way - record Q4 revenue on accelerating AI infrastructure spend, advanced packaging growing ~70% year-on-year, and a Street consensus of Moderate Buy with an average target, 24% above the close. We hold the full position and tighten the trailing stop instead. OPS: STRONG SELL - REDUCE cv76 Sell_Rank 2 Portion 88.52% is a LONE flip out of HOLD / ReversalUp_Oversold cv46 - US-18 requires a 2nd consecutive settled print OR a close below the trail invalidation, and neither holds. L2 fails on both legs (no stop break, 12wk +8.96%). ADR-0012 armed (T12_Progress 91.78, Gain_ATR 0.29 - genuinely in profit this time, not the loss-artifact) but NOT FIRED: no second Exit_Warning code, no >=20pt conviction collapse, forward alive, no repeated NO TRADE -> mandatory trail-raise, not a trim. Counterfactual optional_trim n=18 median wh_4w -10.85% [SAVED] supports. This is the book's largest position (, ~27%) and the third consecutive night of declining to sell it - a deliberate, stated risk. TRAIL RAISED 162..00 (a decisive break of EMA200 171.16). RE-ARM: the 50% trim fires immediately on a 2nd consecutive bearish settled print, a settled close below 168.00, or Est_12W turning negative. Yesterday's identical withdrawal was worth Est_12W exceeds the DCF anchor 99.91 - annotated, not modified. · news |
| NBIS | Nebius Group, N.V. | 188.43 | HOLD | 188.43 | 153.00 | 197.37 (+4.7%) | 211.11 (+12.0%) | 269.19 (+42.9%) | Nebius rallied 27% - its biggest single day of the year - as the neocloud unwind reversed and Freedom Capital upgraded the stock to Buy with a target, arguing enough risk had been squeezed out after a 40% drawdown from the June peak. The position has recovered from -30% to -11% against cost in two sessions, reclaimed both its 200-day and 100-day lines, and carries a +42.9% twelve-week view. Nothing in the sell case survived; we hold and raise the stop. OPS: the 50% trim withdrawn on 07-30 stays withdrawn and is now strongly vindicated. NO TRADE cv0 - never a dispose basis (US-17). Still no confirmed Meta contract impairment, so the overlay override remains NOT triggered. Reclaimed EMA200 153.98 and EMA100 186.50. STOP RAISED 135..00 (just below the reclaimed EMA200) = the re-arm trigger. The 12wk band [111.68-591.00] is a 5.3x spread at Conf 80 - near-uninformative, cited with that caveat. Q2 due ~08-12 (some sources 08-06). · news |
| GOOGL | Alphabet, Inc.- Class A | 333.66 | HOLD | 333.66 | 319.00 | 335.39 (+0.5%) | 338.86 (+1.6%) | 359.26 (+7.7%) | Alphabet gave back 0.9% in a session where the rest of the book ripped, snapping a four-day run, but the position is intact: comfortably above its 200-day line with a +7.7% twelve-week view and no active sell signal. We hold on the existing stop. OPS: exit stays WITHDRAWN, 8th session. NO TRADE cv0 for a 4th+ consecutive session - never a dispose basis (US-17; cell NoTrade|sell -1). ADR-0012 nominally armed (T12_Progress 92.87) and tick (d) repeated-NO-TRADE technically fires, but the fire is DECLINED on three grounds: (i) no in the portfolio JSON so Gain_ATR is unknown and PROFIT_ARMED is unverifiable - the known loss-artifact, systemic finding 4; (ii) tick (c) is alive, forward +7.7%; (iii) US-19 bars re-firing a soft-tick-only trim across consecutive EODs on a name whose forward keeps oscillating - GOOGL already fired-and-withdrew this exact trim on 07-09/10 and again on 07-30. Stop 319.00 (EMA200 320.14) held. Overlay '<345' line still triggered but healing - needs +3.4%, having slipped from +2.5% yesterday. Est_12W exceeds the DCF anchor 145.29 - annotated, not modified. RE-ARM on a settled close below 319.00, or failure to reclaim 345 within 3 more sessions WHILE the 12wk turns negative. |
| LLY | Eli Lilly & Co | 1154.97 | HOLD | 1154.97 | 1155.71⚠ | 1161.65 (+0.6%) | 1146.39 (-0.7%) | 1187.01 (+2.8%) | Lilly fell 4.6% as money rotated violently out of defensives and into the rebounding chip complex, closing a hair below our trailing stop - 0.06%, which is noise, not a break. Q2 earnings land on 5 August, three sessions away. Our rule for a held name with a catalyst inside a week is to hold and trail through the event rather than pre-trim into it, and the signal is still constructive with a positive twelve-week view. We hold. OPS: 6th consecutive FailedBreakdown_Up_20D, cv. Breach is MARGINAL (-0.06%), not decisive - the same test that correctly held KLAC at -0.47% on 07-30. Catalyst-aware hold governs (Q2 08-05 BMO). Lost EMA20 1179.23 today but holds EMA50 1137.69 / EMA100 1084.32 / EMA200 1018.87. ADR-0012 armed (T12_Progress 97.30) but NOT fired: cv drop is 6 versus the -20 tick, and tick (c) is alive though thin (Est_12W 1187.01 vs the 1178.07 threshold). Model Stop_Price printed 1083.03, BELOW the standing trail - a trail is never lowered, so 1155.71 STANDS as the stop of record and will keep flagging on the monitor, correctly. Est_12W exceeds the DCF anchor 387.04 - annotated, not modified. ACTION threshold is a DECISIVE close below 1120.00 (~-3%) or a post-earnings breakdown. |
| OKTA | Okta Inc. | 140.42 | HOLD | 140.42 | 130.00 | 141.25 (+0.6%) | 145.42 (+3.6%) | 158.51 (+12.9%) | Okta added 2.6% to close above our cost for the first time since the July add, on a second consecutive constructive print and a +12.9% twelve-week view. It sits well clear of its 50-day line. We hold and tighten the stop to lock in the recovery. OPS: 2nd consecutive FailedBreakdown_Up_20D cv51; +0.44% on the 139.80 cost. NOT armed (T12_Progress 88.59 < 90, Gain_ATR 0.10 well under 3.0) so no profit-take question arises. Model Stop_Price 135.88 prints ABOVE the standing 124.00 trail but is only 0.71 ATR below the close - too tight for this name's volatility. STOP RAISED 124..00 (~1.6x ATR below close, under EMA20 138.55). Est_12W exceeds the DCF anchor 107.59 - annotated, not modified. |
| SPCX | Space Exploration Tech Corp | — | HOLD | — | 135.00 | n/a | n/a | n/a | SpaceX has no public quote so it stays muted on signals, but its calendar is now the most loaded in the book. The maiden quarterly report as a public company lands 4 August, with 8bn of revenue and a per-share loss expected, and two days later roughly 911.5 million shares worth ~$116bn come off lock-up - with insiders able to sell as much as 44% of the company by early September, expanding the float around ninefold. Morgan Stanley argues the fundamentals are largely unchanged and the pressure creates an entry point. We are not adding into that, and we are not exiting a position we cannot price: we hold on the manual stop through both events. OPS: permanently absent from the scan (private/unquoted) - muted per the edge case, news-assessed every run. Classification CONCERN, both catalysts INSIDE 5 sessions. Horizon bands: earnings +/-18%, lock-up -25%/+10%. Further ~7% tranches ~08-21 and ~09-10; a 10% tranche unlocks if the stock trades 30% above the IPO price, above; Musk and core insiders stay locked to mid-2027. Index-eligibility mechanics open 08-07. Manual stop 135.00; a scan Close < 135 AUTO-UNMUTES and becomes a full-exit trigger. · news |
| WDS | Woodside Energy Goup Ltd | 23.38 | NO ACTION | 23.38 | 21.19 | 23.43 (+0.2%) | 24.35 (+4.1%) | 26.38 (+12.8%) | Highest-conviction buy candidate in the book and the cleanest entry in weeks - a third consecutive STRONG BUY on a 20-day breakout, sitting exactly at the model entry with no chase - but it fails the quality screen and cannot be opened as a new position. OPS: STRONG BUY cv90 Buy_Rank 1, escalating -> 90, entry 23.38 = close (eD 0.00, RSI 68.04), above every EMA. val_adj -4 (MoS -0.80), cell BreakoutUp_20D|buy -7 -> adj 79, comfortably clear of the 65 floor. BLOCKED: Quality_Pass False (F-Score 4, GROWTH_PRICED_IN, Rev_Growth -1.5%). 13th consecutive session with no admissible new BUY (US-20). Do NOT loosen the gate to force a fill. |
| DDOG | Datadog Inc. | 268.56 | NO ACTION | 268.56 | 262.08 | 269.80 (+0.5%) | 278.04 (+3.5%) | 283.45 (+5.5%) | Datadog is building strength - a second consecutive BUY with conviction rising - but it is priced far beyond any defensible valuation anchor and fails the quality screen, so it stays on the watchlist rather than in the book. OPS: BUY cv81 Buy_Rank 2, at entry, RSI 61.9. Two independent bars: val_adj -10 (MoS -18.97, the most extreme in the pool) plus cell FailedBreakdown_Up_20D|buy -7 -> adj 64, BELOW the 65 floor; and Quality_Pass False (F-Score 5). L1/US-16 also discount this momentum-chase family. |
| AMKR | Amkor Technology | 48.26 | NO ACTION | 48.26 | 38.27 | 51.23 (+6.2%) | 56.66 (+17.4%) | 60.52 (+25.4%) | Amkor flipped from a hard breakdown straight to an oversold-reversal buy on the sector bounce, but that is a single-session reversal on a name that fell 30% in three days, and it fails the quality screen. No action. OPS: BUY cv70 Buy_Rank 3, fresh flip from AVOID / BreakdownDown_20D cv83, RSI 33.48, still below EMA20 62.49 / EMA50 67.42 / EMA100 64.52. val_adj -6 (MoS -1.11); the ReversalUp_Oversold|buy cell is NOT yet actionable (report-only, no bias applied) - flagged, because this family generated most of tonight's signal churn and has zero graded history. adj 64, BELOW the 65 floor. Also Quality_Pass False (F-Score 4). |
| SNOW | Snowflake Inc. | 298.10 | NO ACTION | 293.87 | 244.12 | 299.12 (+1.8%) | 318.18 (+8.3%) | 332.19 (+13.0%) | Snowflake kept running to a new high but conviction fell as it went - the classic extension pattern - and it now sits overbought, above its own entry trigger, and fails the quality screen. No chase. OPS: BUY cv65 Buy_Rank 4, conviction DEGRADED while the price rose 5.4%. RSI 74.07 and close 298.10 is above the 293.87 buy-stop (eD -0.26, at the no-chase boundary). val_adj -8 (MoS -1.66), cell BreakoutUp_20D|buy -7 -> adj 50, far below the floor. Quality_Pass False (F-Score 5). Three independent bars. |
Outcome & track record
Accuracy at the time of this record.
105 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.