Recommendations
0 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 |
|---|---|---|---|---|---|---|---|---|---|
| GOOGL | Alphabet Inc Cl A | 357.75 | SELL 30% | 357.75 | 348.52 | 356.86 (-0.2%) | n/a | 347.13 (-3.0%) | Alphabet is being de-risked, not written off. The signal engine has printed NO TRADE on two consecutive settled closes - it cannot read the name - and the 12-week view has turned negative at 347.13 against a 357.75 close, so the model sees no upside to pay for the uncertainty. The news explains why: Jeff Dean and Sanjay Ghemawat are leaving DeepMind to found their own company, which knocked the stock 4% on 05-Aug despite a Q2 that beat, and Reuters reports Alphabet seeking roughly $80bn in offerings on top of an already-raised capex guide. Valuation adds nothing to the case in either direction - at a 145 DCF anchor against a 347 price the intrinsic model has no view worth acting on - so this is a signal-originated trim, taking a third of the line off and holding the rest. Ops: 2,422 of; fires ADR-0012 ticks (c) dead 12wk + (d) 2x NO TRADE; tier held at 30% not the mechanical 50% because Unreal_PnL is -0.60% so the profit-take precondition is not truly met (see Systemic finding 1) and because the 08-05 card pre-committed to exactly; G2 suppresses the valuation-amplify path; trail on the remainder held at 348.52; cross-book: the us book fired the identical trim on tick (d) tonight, separate share lot, no L7 conflict. · news |
| NYT | New York Times Co Cl A | 64.77 | SELL 50% | 64.77 | 56.44 | 63.70 (-1.7%) | n/a | 51.95 (-19.8%) | The New York Times has a structural problem, not a bad quarter. Q2 added about 280,000 digital subscribers against 295,300 expected and 310,000 the quarter before - and it did that in a heavy news cycle, which is when this business is supposed to shine. Management pointed at falling Google search and referral traffic, and guided Q3 subscription revenue growth below consensus. The chart agrees: the stock fell more than 13% on the print, sits 10% below its 200-day average with RSI at 32, and the engine now prints STRONG SELL with a 12-week target of 51.95, nearly 20% under the current price. Half the position comes off here on top of the 30% already sold, leaving a smaller stake to close out if the selling label holds another session. Ops: 1,592 of at ~64.77, realizing roughly -on a -12.30% position; L2 satisfied on both limbs (stop break + negative 12wk); 50% rather than the engine's 88.52% because this is the first STRONG SELL print in the trail and MoS -1.81 is short of the -3 that would bump the tier; stop on the remainder raised to 56.44; 12wk target sits below the 68.16 DCF anchor so no over-projection flag. · news |
| GOOG | Alphabet Inc Cl C | 356.62 | SELL 30% | 356.62 | 347.25 | 355.86 (-0.2%) | n/a | 347.57 (-2.5%) | The C-share line is trimmed alongside the A-shares for the same reason and at the same time: two consecutive NO TRADE prints from the signal engine and a 12-week target of 347.57 that sits below today's 356.62 close, against a news backdrop of senior AI departures from DeepMind and a reported ~$80bn raise. Holding one Alphabet share class while cutting the other would leave the book with two contradictory views of one company. Ops: 161 of; ADR-0012 ticks (c)+(d), tier held at 30% per the 08-05 pre-commitment; G2 suppresses valuation-amplify (DCF 145.29); trail on the remainder held at 347.25. · news |
| VRSN | VeriSign Inc | 292.53 | NO ACTION | 292.53 | 276.38 | n/a | n/a | 324.88 (+11.1%) | VeriSign is a genuine winner (+14.08%) whose profit-take technically triggered tonight, but the trigger is the engine losing the signal rather than the business weakening - and the business is visibly not weakening. Q2 revenue rose 6% to, EPS came in at, and the company booked a record 12.7 million new domain registrations, its largest quarterly gain ever, with full-year guidance intact. The 12-week view still points 11% higher. The stake is held and the stop is raised instead. Ops: armed via Gain_ATR 3.40, fired on tick (d) only (2x NO TRADE); slot-crowded out of the <=3 dispose cap at of dollar bleed; mandatory candidate next EOD if a second, deterioration-based tick appears; trail held at 276.38; ex-div on 2026-08-19. · news |
| NVR | NVR Inc | 6304.81 | NO ACTION | 6304.81 | 5954.11 | n/a | n/a | 5701.20 (-9.6%) | NVR is the clearest sell case in the book and the hardest one to act on, because it is only three shares. The engine has printed STRONG SELL for four consecutive sessions at the book's top sell rank, the stock trades 7.5% below its 200-day average, and the 12-week view is 9.6% lower. The fundamentals match: Q2 earnings fell 22.6% year on year and missed by 11.5%, homebuilding revenue came in under consensus, and the wider homebuilder complex is being squeezed by mortgage rates - Lennar, also held here, has cut its full-year delivery target. The exit stands; it simply keeps losing its slot to positions forty times its size. Ops: standing 100% close-out carried since 2026-07-31, 6th session deferred; of dollar bleed cannot outrank an card under the loss-prevention ordering - see Systemic finding 2; stop 5,954.11. · news |
| STZ | Constellation Brands Cl A | 132.52 | NO ACTION | 132.52 | 126.89 | n/a | n/a | 105.93 (-20.1%) | Constellation Brands prints STRONG SELL at the third sell rank, trades 9.4% below its 200-day average and carries a 12-week view 20% lower - a clean exit case on every limb except size. At and roughly it is rounding error in a book. Ops: adjusted conviction 64 sits one point under the 65 floor, so this is admitted by L2 (stop break + negative 12wk) rather than by score, which is why it is deferred rather than promoted into a slot; standing 100% close-out carried since 2026-07-31; stop raised to 126.89. |
| JEF | Jefferies Financial Group | 56.32 | WATCH | 56.32 | 55.25 | n/a | 60.61 (+7.6%) | 58.44 (+3.8%) | Jefferies was the only buy signal the entire book generated tonight, and it does not pass muster. The stock screens as overvalued against its own book economics and fails the quality screen on profitability and cash-flow consistency - and a failed quality screen blocks brand-new positions outright, regardless of how good the chart looks. The setup is also the least reliable buy pattern this book trades. No position is opened. Ops: BUY cv74 Buy_Rank 1, adjusted 72 clears the 65 floor but is hard-blocked twice - Quality_Pass=False on a new entry (, 13F seed rounded to zero shares), and US-10 requires raw cv >= 75 for a FailedBreakdown_Up_20D buy; not queued. |
| KHC | Kraft Heinz Co | 24.96 | HOLD | 24.96 | 24.15 | n/a | 27.04 (+8.3%) | 27.43 (+9.9%) | Kraft Heinz slipped just below its stop, and we are holding through it. The breach is barely 1% and the case underneath has improved, not deteriorated: Q2 beat on both earnings and revenue, management lifted the full-year outlook, the planned split into two companies remains on track, and another of marketing spend is committed for the second half. It is also the cheapest name in the book on intrinsic value, and the 12-week view points nearly 10% higher. The stop moves down to just under the long-term trend line so the position is not shaken out on noise. Ops: monitor breach -1.15% (24.96 vs 25.25) on /; L2 does NOT fire (12wk +9.9%, price above EMA200 24.41); stop re-based 25.25 -> 24.15, set just under EMA200 so the line is a trend break rather than noise; news overrides toward hold. · news |
| SIRI | Sirius XM Holdings Inc | 29.88 | HOLD | 29.88 | 28.60 | n/a | 32.21 (+7.8%) | 32.95 (+10.3%) | Sirius XM dipped fractionally through its stop on a day when the operating news was the best it has been in years: the company returned to subscriber growth in Q2 with 22,000 net additions, posted the lowest customer churn in its history at about 1.4%, and raised full-year guidance across revenue, earnings and cash flow. The 12-week view is 10% higher and the stock sits comfortably above its long-term trend. There is also a mechanical wrinkle worth naming: the shares go ex-dividend 27 cents on 10-Aug, which by itself knocks nearly a percent off the price - the old stop would have been triggered by arithmetic rather than by anything going wrong. The stop is reset below that gap. Ops: breach -0.86% (29.88 vs 30.14) on / inside one ATR (1.16); L2 does NOT fire; close 15.6% above EMA200; stop re-based 30.14 -> 28.60, deliberately below the 08-10 ex-div gap. · news |
| CVX | Chevron Corp | 189.23 | HOLD | 189.23 | 179.56 | n/a | 198.90 (+5.1%) | 195.41 (+3.3%) | Chevron is up nearly 12% and has reached the point where the book protects the gain rather than banks it - the forward view still points modestly higher, so the stop is raised and the position rides. Ops: ADR-0012 armed via Gain_ATR 4.14 / Unreal +11.84%; no tick fires - (c) misses narrowly (Target_12Week 195.41 vs 1.02x close = 193.01, +3.3% headroom), no second Exit_Warning code, conviction is well under the 20-point tick; mandatory trail-raise 178.72 -> 179.56 (2.0x ATR chandelier). |
| OXY | Occidental Petroleum Corp | 56.04 | HOLD | 56.04 | 51.76 | n/a | 60.32 (+7.6%) | 54.70 (-2.4%) | Occidental was already cut in half at this morning's open on exactly the concern that is still showing tonight, and cutting the same position again the very next session on an unchanged reason would be churn, not discipline - especially with the signal actually improving. The remaining stake is up 14.6%, holds above its long-term trend, and is protected with a tighter stop instead. Ops: armed via Gain_ATR 3.33 / Unreal +14.58%; tick (c) is live (Target_12Week 54.70 < 1.02x56.04 = 57.16) but does NOT re-fire - the 50% trim filled at 56.43 this morning on this same unchanged tick and conviction has since RISEN; trail raised 50.23 -> 51.76 (2.0x ATR); re-arms only on a NEW tick (second Exit_Warning code, 20-pt conviction drop, or 2x NO TRADE). |
| NUE | Nucor Corp | 272.00 | HOLD | 272.00 | 255.59 | n/a | n/a | 343.87 (+26.4%) | Nucor is the book's best position at +23% and the forward view still points 26% higher, so there is nothing to take here - the stop simply stays where it is. Ops: armed via Gain_ATR 5.90; tick (c) far from firing; tick (b) deliberately NOT read off the label flip - a NO TRADE row has no conviction to decay, which is precisely why tick (d) exists as a separate 2-print test, and this is print one; trail held at 255.59 (2.0x ATR would be 254.63, lower - trails ratchet up only); 30% already trimmed 08-04. |
| DVA | DaVita Inc | 180.67 | HOLD | 180.67 | 174.64 | n/a | n/a | 174.84 (-3.2%) | Half of DaVita was sold at this morning's open after the earnings de-rate. The remainder is held: the exit trigger the last decision named - a break of the 200-day average - has not been hit, the stock closed above that line, and momentum is washed out enough (RSI 21) that forcing the rest out here would be selling into capitulation. The stop is moved up to sit exactly on that line so the exit is mechanical from here rather than discretionary. Ops: remaining at -23.09%; EMA200 = 174.64 vs 180.67 close, trigger NOT hit; stop raised 158.04 -> 174.64, which now carries the full-exit instruction; NO TRADE cv0. |
| MCO | Moody's Corp | 473.13 | HOLD | 473.13 | 461.27 | n/a | n/a | 480.41 (+1.5%) | Moody's is the position to watch. It is the book's third-largest holding, the signal has weakened for four straight sessions, the pattern has turned to a failed breakout, and the price is now sitting exactly on its 200-day average with a forward view that has flattened to barely 1.5% of upside. Nothing is actionable tonight - the signal is too weak to trade in either direction - but a decisive close below that trend line with the forward view turning negative makes this the next name to reduce. Ops: conviction ->, label AVOID; adjusted 44 far under the 65 floor so no card; close 473.13 vs EMA200 472.48; NOT treated as a profit-take despite the PROFIT_ARMED flag - position is -3.16% so the arming is the inverted T12_Progress limb (Systemic finding 1); trail 461.27. |
Outcome & track record
Accuracy at the time of this record.
73 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 →
0.025 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.
46 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.