US — End-of-Day Verdict
NO ACQUIRE (OKTA adj 64 -- ONE point short of the floor; the cleanest, non-extended, quality-PASSING PullbackBuy in a month, killed by a maxed valuation haircut. DDOG/BAC/SNOW quality-blocked and/or sub-floor. US-20 6th straight session). NO ANTICIPATE (label absent from all 4 snapshots). DISPOSE: GLW 150/150 FULL EXIT @~162.41 -- take the +6.1% melt-up bounce on an exit outstanding five sessions; the signal escalated to an active STRONG SELL-REDUCE at Sell_Rank 1 on a -18.6% loser trading below EMA10/EMA100. NVDA 30/100 @~207.29 -- profit_take_1, at target with a flat forward on every horizon; keep 70, trail 200.50. Holds, all with raised trails: KLAC 205 (breach resolved, hold through 07-23 Q4), GOOGL 345 unchanged (hold through tonight's Q2 print), MU 880 (now risk-free after +12.2%), NBIS 190 (after +18.8%, forward +49%), LLY 1120, META 620, SPCX muted 135. The session's shape: cut the one broken name into a melt-up, bank a third of a winner whose forward has gone flat at target, hold every genuine trend position through the 07-22/23 earnings pair on raised stops, and carry of into the event-dense 07--31 window. Net +raised.
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 |
|---|---|---|---|---|---|---|---|---|---|
| GLW | Corning Incorporated | 162.41 | SELL ALL | 162.41 | — | 163.98 (+1.0%) | 160.33 (-1.3%) | 173.41 (+6.8%) | Corning has now printed a bearish signal for six straight sessions and today it escalated to an outright STRONG SELL - REDUCE at the settled close, ranked the single worst name in the book to hold, with conviction rising through the session. The stock is trading below both its short- and medium-term trend lines, is our only large loser at roughly -19%, and rallied noticeably less than the rest of the book on a broadly explosive day - relative weakness that confirms the problem is company-specific rather than market-wide. Valuation offers no cushion either, so we take the full exit into today's bounce rather than hold a broken position for a sixth session. OPERATIONAL: Sell_Rank 1, 88.52 amplified to full by MoS_FV -11.34 (<= -3 tier bump). Stable family across J20C->J21C (cv 52/62/62/69), close 162.41 below EMA10 172.51 and EMA100 169.41. US-18 second-print bar met many times over; G-gates inapplicable (broken downtrend, CLAUDE.md 9 'valid exits still fire'); US-23 - the stale-bullish Est_12W +6.8% must not veto a signal/stop exit, and Est_4W is already negative. Card outstanding since 07-16, repriced +6.1% higher than the 07-21 ref of 153.10. Proceeds realized ~-Cross-book: executes ONCE here (L7). |
| NVDA | NVIDIA Corporation | 207.29 | SELL 30% | 207.29 | 200.50 | 207.79 (+0.2%) | 207.69 (+0.2%) | 208.18 (+0.4%) | Nvidia has essentially reached its twelve-week objective, and the model's own forward view now sees the shares going nowhere from here - roughly flat at one week, four weeks and twelve weeks alike. That is a dead risk/reward on a position sitting on a large gain, so we bank a third and let the remaining seventy shares ride the AI-hardware trend with a raised protective stop. This is a partial de-risk, not a call against the story: the late-July cloud-capex updates are a reason to stay invested in the majority. OPERATIONAL: ADR-0012 profit_take_1 (30%), 1 tick - PROFIT_ARMED (T12_Progress 99.6%) + tick (c) dead forward (Est_12W 208.18 < 1.02x207.29 = 211.44). Ticks (a)/(b)/(d) do NOT fire (no 2nd warning code; conviction ROSE; label HOLD not NO TRADE). G1 n/a (not a fresh breakout >=80). US-19: SAME first fire re-verified at the settled close and repriced +2.0% from the 07-21 ref 203.28 - size deliberately NOT escalated, 4th reprice. Valuation MoS -2.33 > -3 so no amplify (G2 would suppress anyway). Keep 70, trail-raise.50 (EMA100 200.64 cluster; scan chandelier 193.29 looser). Proceeds Cross-book: executes ONCE here (L7); NVDA shared with ussemicon. |
| OKTA | Okta Inc. | 141.71 | NO ACTION | 140.27 | 129.63 | 143.95 (+2.6%) | 144.77 (+3.2%) | 150.56 (+7.3%) | Okta is the most attractive buy candidate the book has surfaced in a month - a steady pullback entry, not chasing, with a healthy balance sheet - but the shares already trade far above what the cash-flow model says the business is worth, and that valuation penalty pulls it just below our entry bar. We pass, and log it so we can measure what the discipline cost. OPERATIONAL: adj 64 vs the 65 floor - ONE point short, no fudging. cv71 PullbackBuy_VWAP Buy_Rank 2, BUY on 4-of-4 snapshots (64/57/57/71), eD -0.16 (not extended, inside the -0.25 no-chase bar), RSI 56.2, Quality_Pass TRUE (F=7) - the only candidate today that clears the gate. Killed by val_adj -10 (MoS_FV -5.30 maxes the haircut) against a PullbackBuy_VWAP|buy bias that has decayed +5->+4->+3 as it graded. Size modulator would have been 0.5x (MoS <= -3). Logged direction:none/optional:true so the counterfactual grader prices the veto (systemic #1). Re-arm: a pullback toward the 140.27 entry with the cell bias recovering, or a fresh valuation snapshot. |
| KLAC | KLA Corporation | 217.56 | HOLD | — | 205.00 | 219.11 | 219.11 | 226.47 | KLA has fully recovered the brief dip below our stop last week, closing up nearly 5% and back above every trend line - the decision to hold through a marginal breach was right. It reports quarterly results tomorrow, so we hold through the print with a raised protective stop rather than pre-trimming into a binary event. OPERATIONAL: breach RESOLVED (217.56 vs the 210.15 old trail and the 200.51 decisive line). Book's largest position, GENUINELY armed (T12_Progress 96.1%, +23.8%, Gain_ATR 2.68) with 4x NO TRADE cv0 -> tick (d) fires mechanically, but tick (c) does NOT (Est_12W 226.47 = +4.1% alive), US-17 forbids a bare NO-TRADE dispose (NoTrade|sell -2, hit.40), and Q4 earnings 07-23 (next session, +/-8%) is inside the <=5-session catalyst-hold window - no pre-trim absent a confirmed lower high or decisive stop break, and the name just made a HIGHER high. Armed-but-not-fired -> mandatory trail-raise to 205.00 (below the 207.60 swing low, above EMA100 197.34). Post-print: decisive close < 205.00 or confirmed lower high converts to a trim next EOD. Systemic #4: tick (d) again firing on a quiet winner with a live forward. |
| GOOGL | Alphabet, Inc. - Class A | 347.15 | HOLD | — | 345.00 | 349.54 | 361.59 | 379.25 | Alphabet reports quarterly results after the close tonight and was the book's only decliner on an otherwise explosive day, leaving it just above the level we have set as our line in the sand. We hold through the print - trimming into a binary event is how you get the direction right and the price wrong - and treat a settled close below that line afterwards as a full exit trigger. OPERATIONAL: close 347.15 (-1.4%), +0.6% above the 345 hard invalidation, sitting on EMA100 347.52, below EMA10 354.56, RSI 43.1. Q2 earnings 07-22 AMC, +/-7% / +/-1.88 ATR. NO TRADE cv0 x4 fires tick (d) on a PROFIT_ARMED row, but tick (c) does not (Est_12W 379.25 = +9.2% alive) and the catalyst rule mandates no pre-trim. Avg_Cost blank in usportfolio.json so the arm cannot be verified as a genuine gain (systemic #2). 345 stays hard THROUGH the print; a post-earnings settled close below it is a full exit, not a trim. |
| MU | Micron Technology, Inc. | 970.82 | HOLD | — | 880.00 | 971.83 | 1002.80 | 1025.79 | Micron surged more than 12% in a single session on the memory upcycle, and the thirty shares added at the start of the week are already up nearly 12%. The forward view stays positive across every horizon, so we hold the full position and lift the protective stop above our cost - the trade is now risk-free and we let the trend run. OPERATIONAL: close 970.82 (+12.2%), blended cost 859.49, +12.95%, Gain_ATR 1.32. The AVOID cv64 FailedBreakout_Down_20D print with Stop_Price 1139.79 sitting 17% ABOVE the close is the short-side structural artifact (systemic #3), NOT a sell instruction - 0.0, Sell_Rank blank, long-only means AVOID != short. Armed via T12 (94.6%) but tick (c) does NOT fire (Est_12W 1025.79 = +5.7%, Est_4W 1002.80 = +3.3%) and there is no second warning code -> no profit-take. Trail-raise.00 (1.07 ATR below on ATR 84.48; above the 859.49 cost). Far above EMA100 754.81 / EMA200 571.63. HBM-sold-out overlay unbroken; SK Hynix read-through 07-29. Shared with ussemicon - changed on 07-21, sub-books must not size off a stale snapshot. |
| NBIS | Nebius Group, N.V. | 216.92 | HOLD | — | 190.00 | 217.46 | 234.68 | 323.62 | Nebius jumped nearly 19% in one session, flipping the position from a double-digit loss to a gain and reclaiming its medium-term trend line. It carries the strongest forward outlook in the entire book, so we hold the full stake and raise the protective stop sharply to lock in the bulk of the two-day move ahead of its late-July results. OPERATIONAL: close 216.92 (+18.8%), cost 211.60, +2.51%. NOT armed (no Exit_Warning, T12_Progress 67.0) so the profit-take path is not open at all. Same degenerate-stop artifact as MU (Stop_Price printed EQUAL to Close 216.92 - meaningless; systemic #3), 0.0, no Sell_Rank -> the AVOID label is not a sell instruction. Est_4W 234.68 (+8.2%), Est_12W 323.62 (+49.2%) - strongest forward in the book. Now above EMA100 186.25, far above EMA200 151.19. Keep-core overlay live (Meta-deal invalidation NOT triggered). Trail-raise.00 (1.09 ATR below) ahead of Q2 earnings 07-29, +/-15% - the widest band in the book; revisit the line as the date approaches per the horizon's tighten-trail stance. |
| META | Meta Platforms | 643.81 | HOLD | — | 620.00 | 645.98 | 622.75 | 635.00 | Meta is a four-day-old position up a little over 1% with quarterly results due at the end of the month. There is nothing to bank and nothing broken - we simply hold on the existing stop and let the earnings print set the direction. OPERATIONAL: close 643.81 (-0.3%), cost 635.00, +1.39%, Gain_ATR 0.36. The PROFIT_ARMED flag is an ARTIFACT for the second consecutive EOD - it arms only via T12_Progress 101.4%, which reads 101% because Target_12Week (635.00) sits BELOW the close; a +1.4% four-day-old position with 0.36 ATR of gain is not a profit to take. Explicitly DECLINED (systemic #2). Intraday HOLD cv59 x2 faded to NO TRADE cv0 at the settled close - the familiar fade (L4/US-7/US-11). EMA100 621.17 / EMA200 639.63 straddle the 620 line; keep it. Q2 earnings 07-29 AMC, +/-8% / +/-1.74 ATR - a direct book event. |
| LLY | Eli Lilly & Co | 1175.41 | HOLD | — | 1120.00 | 1180.90 | 1195.49 | 1212.93 | Lilly added another 2.5% and continues to grind higher well above its longer-term trend lines, with the forward view still pointing modestly up. Nothing here calls for action beyond tightening the protective stop under the advancing price. OPERATIONAL: close 1175.41 (+2.5%), HOLD FailedBreakdown_Up_20D cv50 (trail 55/49/50/50 - a -5 conviction drift, nowhere near the -20 tick-(b) bar). Armed (T12_Progress 96.9%) but tick (c) does NOT fire: Est_12W 1212.93 vs 1.02x1175.41 = 1198.92 -> forward alive at +3.2%; Est_4W 1195.49 (+1.7%). Far above EMA100 1068.43 / EMA200 1006.16. Trail-raise.00 (1.44 ATR below; scan chandelier 1098.44 is looser - keep the higher line). Overlay hard invalidation 975 now academic; pharma-onshoring tariff step 07-31 (+/-3%). |
| SPCX | Space Exploration Tech Corp | — | HOLD | — | 135.00 | n/a | n/a | n/a | A private holding with no public quote, so there is no tradable signal - we hold on the manual stop. The one thing to watch is insider selling restrictions beginning to lift after the company's second-quarter results, which can weigh on secondary-market pricing over the coming weeks. OPERATIONAL: permanently absent from the scan -> MUTED on the verdict (no Action/conviction/entry/targets ever synthesized), but news-assessed every run per the edge-case rule. News read UNCHANGED: CONCERN - lock-up overhang persists (first ~20% frees after Q2 results late July, then ~7% tranches Aug-Oct); the horizon's bonus-tranche trigger observation window 07--03 (+/-6%, low confidence) is now LIVE. Classify: watch/concern. Manual stop 135 stands. AUTO-UNMUTE and resume full signal+fusion treatment the moment SPCX appears in the scan CSV with a non-null Close. |
Outcome & track record
Accuracy at the time of this record.
87 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 →
-0.171 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.