Recommendations
0 to acquire · 7 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 |
|---|---|---|---|---|---|---|---|---|---|
| AXP | American Express Co | 337.33 | SELL 50% | 337.33 | 330.57 | n/a | n/a | 321.88 (-4.6%) | American Express has printed a stable STRONG SELL - REDUCE for a second straight session on FailedBreakout_Down_20D, the most reliable sell pattern this book has (44 graded trades, 53% four-week hit rate), with the model's twelve-week view now 4.6% below the current price for a fifth consecutive session. The card is deliberately sized at half the line, not the model's 88.5%, because the shares have just recovered back above both their 200-day average and our stop, and because the business itself is doing well - management guides to 10% annual revenue growth and mid-teens earnings growth, has raised the dividend 16%, and an intrinsic-value screen still reads the shares as cheap after a 120% five-year run. Trimming half respects a bearish chart on the book's single largest holding (11.6% of net asset value) while keeping half invested in a franchise that is not broken. OPERATIONAL: 6th issuance, blocked five times by the executor CASH-drift blocker (now six sessions old). L2 no longer fires - the EMA200 break and stop break that carried the 08-20 card have both reversed; the stable 2-print REDUCE label carries this alone. Model 88.52 REFUSED as pro-cyclical into a recovery - deviation from model sizing is deliberate and on the record. Valuation up-ranks only (MoS -0.455, far from the -3 tier-bump line); Quality_Pass False gates new entries only. Stop RAISED 324.60 -> 330.57 (close - 1.0 ATR, 2.00% cushion); peak-stop watermark re-bases with the 50% cut so ADR-0020 does not fire. REDUCE row prints no Target_Price -> 4wk n/a, 1wk reported as the model band 330.58-344.08. · news |
| CB | Chubb Ltd | 345.83 | SELL 30% | 345.83 | 341.96 | n/a | n/a | 333.95 (-3.4%) | Chubb has broken a three-session lull with a fresh STRONG SELL - REDUCE signal on the book's most reliable sell pattern, and the model's twelve-week view has been negative and widening for four sessions running, now 3.4% below the current price. We are taking only the minimum 30%, for two independent reasons: our own valuation work reads the shares as slightly undervalued, and Wall Street disagrees with the chart outright - Citi raised its target to in July, the 26-analyst consensus is a Moderate Buy with 6% average upside, and full-year earnings are seen growing nearly 11%. Selling cheap means selling small: we trim a third of a 5.6%-of-net-asset-value position and keep the rest, with a dividend going ex on 11 September. OPERATIONAL: this is a fresh origination, NOT a re-run of the 08-19 stop-breach card that lapsed and was correctly stood down on 08-20. Close is still above the stop so L2 does not fire - the label carries this alone. Model 88.52 stepped down to the 30% floor per S6(b) UNDERVALUED 'trim the minimum', reinforced by the dissenting news. Quality_Pass False gates new entries only. Stop HELD - a tight 1.12% / 0.68 ATR cushion, stated as such; the scan's would loosen it and was rejected. REDUCE row prints no Target_Price -> 4wk n/a, 1wk reported as the model band 338.91-352.75. · news |
| GOOGL | Alphabet Inc Cl A | 348.06 | SELL 50% | 348.06 | 333.68 | n/a | n/a | 312.32 (-10.3%) | Alphabet has broken a four-session drift with a high-conviction sell signal, and this time the news explains it: the company lifted 2026 capital spending guidance to billion and flagged still-higher spending in 2027, knocking the shares down almost 5% after hours, while its AI division lost key leadership including Jeff Dean. The model's twelve-week view is now 10.3% below the current price - the weakest forward read of tonight's four sell candidates. We take half, not the whole line, because the underlying business is genuinely strong: second-quarter revenue rose 24% to billion and Google Cloud grew 82% with a billion backlog. This is a margin-and-leadership de-risk on a stretched valuation, not a verdict on the franchise. OPERATIONAL: converts a prose-only 'TRIM 50% (carried), direction none' that has sat in the ledger SEVEN sessions - ADR-0012 bans exactly that. Trim-discipline gates cleared: not a profit-trim (position at a loss) so G1/G3 do not bite, and G3's own test passes twice (12wk negative AND signal bearish). G2 noted - DCF vs EPV on 27.4% revenue growth makes the DCF anchor unreliable - so no valuation amplification taken; none was available anyway at MoS -0.675. No catalyst-aware hold: only calendar item is the ex-dividend 09-04, 8 sessions out, and a dividend is not an index/earnings catalyst. Expired overlay's 'invalidated below 345' is no longer tripped and carries no force. Stop RAISED -> (scan level, 4.13% cushion, above EMA200). REDUCE row prints no Target_Price -> 4wk n/a, 1wk = model band 341.10-355.02. (!) 12wk exceeds the DCF anchor on a GROWTH_PRICED_IN name - projection assumes growth beyond priced-in. · news |
| NYT | New York Times Co Cl A | 67.49 | SELL 88.52% | 67.49 | 64.15 | n/a | n/a | 48.87 (-27.6%) | The New York Times carries the strongest sell signal in the entire portfolio - highest conviction, top sell rank, and a twelve-week view 27.6% below today's price - but the position is small enough that three larger, equally valid sells took the day's action slots ahead of it. OPERATIONAL: 9th consecutive deferral, purely on materiality (vs / /). Sell_Rank 1, cv76, on the book's best actionable cell (FailedBreakout_Down_20D n=44 hit_4w.529). Position -8.61%. Valuation up-ranks only (MoS -1.330, no bump). Stop RAISED -> (scan level, 4.95% cushion). See systemic finding 1 - dollar-bleed ranking is degenerate in a book whose lines span four orders of magnitude. |
| CVX | Chevron Corp | 203.09 | SELL 30% | 203.09 | 199.24 | n/a | n/a | 234.63 (+15.5%) | Chevron is sitting on a 20% gain and its buy signal has now gone quiet for two straight sessions, which under our profit-taking rules argues for banking roughly a third - but the amount involved is too small to displace tonight's larger trades. ADR-0012 PROFIT-TAKE FIRED: armed on Gain_ATR 9.33 (+20.03%, unrealised) with tick (d) = two consecutive NO TRADE prints (08-22, 08-24). G1 cannot soften it (NoTrade cv0 is neither BreakoutUp_*/MomentumContinuation_* nor >=80) and ticks (b)/(c)/(d) fire regardless of G1. Tick (c) clear - 12wk = +15.5%, alive. The 08-20 refusal rested on G1's intent protecting the book's Buy_Rank 1 name; that protection expired with the buy signal itself. 6th on materiality. Stop HELD (1.90% cushion). Refusal made AGAINST the counterfactual evidence (optional_trim n=116, median +2.81% 4-week cost) and recorded as such. |
| VRSN | VeriSign Inc | 290.79 | SELL 30% | 290.79 | 276.41 | n/a | n/a | 292.75 (+0.7%) | VeriSign is up 13% and the chart has recovered, but the model now sees essentially no further upside over the next twelve weeks, which triggers a partial profit-take; the position is too small to take one of tonight's three action slots. ADR-0012 PROFIT-TAKE FIRED: armed on Gain_ATR 4.12 (+13.40%) with tick (c) = dead forward view, 12wk < 1.02 x =. Label recovered to HOLD cv75 on FailedBreakdown_Up_20D - a recovering label does NOT un-fire a dead 12wk. G1 does not apply (wrong family, cv75 < 80) and tick (c) fires regardless. 7th on materiality. Valuation MoS -2.077, short of the -3 tier line. Stop HELD (4.94% cushion); the scan's would loosen it. |
| ALLY | Ally Financial Inc | 42.45 | SELL 30% | 42.45 | 41.56 | n/a | n/a | 37.66 (-11.3%) | Ally has broken decisively below our protective stop with a negative twelve-week outlook, which is our standard signal to reduce - the position is simply too small to claim one of tonight's three slots. Second-quarter results were solid (net interest income and net income both up year on year, a buyback completed), which is why we would trim rather than exit. L2 FIRES CLEANLY: close is 1.87% / 0.91 ATR below the recorded stop AND 12wk = -11.3%. The only decisive stop breach in the book. Also below EMA200, 5 consecutive NO TRADE, position -6.23%. Valuation UNDERVALUED (MoS +0.053) argues the tier down and 30% is already the floor. 5th on materiality. Stop RE-BASED -> (close - 1.0 ATR); peak stop recorded. · news |
| SIRI | Sirius XM Holdings Inc | 28.45 | NO ACTION | 28.45 | 27.81 | n/a | n/a | 26.19 (-7.9%) | Sirius XM slipped a fraction below our stop, but not by enough to count as a real break - and the business news runs the other way: subscribers returned to growth for the first time in years at a record-low churn rate, and management raised full-year guidance. We hold, with a tightened stop. Breach is 0.52% / 0.23 ATR - NOT decisive, so L2 does not fire despite 12wk = -7.9%. 5 consecutive NO TRADE cv0, position -6.75%. Would rank 4th tonight on materiality and would have taken a slot ahead of ALLY and NYT had the breach been real. News DISSENTS: +22k self-pay net adds (best Q2 in four years), 1.4% record-low churn, FY26 guidance raised (revenue 525bn, adj EBITDA 625bn, FCF 375bn), buyback, dividend paid 08-26, one model reading 56.1% undervalued; offsets are a COO departure with no successor and a -5.8% reaction to the raise. Quality_Pass False. Stop RE-BASED -> (close - 1.0 ATR). · news |
| NUE | Nucor Corp | 244.64 | HOLD | 244.64 | 248.53⚠ | n/a | n/a | 251.15 (+2.7%) | The trim we flagged as owed on Nucor is stood down on merit: the tariff scare that triggered it has reversed, the shares are rebounding, analysts see roughly 20% upside, and the model's twelve-week view has turned positive again. We hold the full position. STAND-DOWN REVERSES the 08-19/08-20 written commitment, deliberately, on four grounds: (1) news flipped back - post-selloff rebound, mean PT (+20% vs), Street high Q2 beat (net sales +23% to 40bn, adj EPS, +7.2% on the print); (2) 12wk POSITIVE at = +2.66% so L2 does not fire; (3) breach is 0.32 ATR, not decisive; (4) ADR-0020 measures the RECORDED stop, which this book deliberately never eroded - is only -4.4% below the peak, far inside the 15% line. The 08-20 entry read the scan's as 'the live stop'; holding the line instead of chasing it down is precisely why the erosion never happened, and it is inconsistent to then trim on an erosion the discipline prevented. Not armed (Gain_ATR 1.98). Stop HELD, still 1.57% above market and non-functional until price recovers. RE-ARMS on a decisive close below (>0.5 ATR) or the 12wk turning negative. RECOMMEND enabling ADR-0020 for usbh at the us default 15% / 30%. · news |
| OXY | Occidental Petroleum Corp | 60.11 | HOLD | 60.11 | 57.90 | n/a | n/a | 68.89 (+14.6%) | Best position in the book at +22.9%; the profit-taking trigger has not fired, so we keep the full holding and rely on the stop. ARMED (Gain_ATR 7.88) but NOT FIRED. HOLD cv52 -> NO TRADE cv0 is ONE print, not two - tick (d) requires two consecutive, and a single flip to a cv0 no-signal row is NOT tick (b); reading it as a 52-point 'conviction collapse' would make tick (d)'s explicit two-print requirement dead letter. Tick (c) clear (12wk = +14.6%). Stop HELD (3.68% cushion); scan prints no stop on a NO TRADE row. A SECOND NO TRADE print fires a 30% trim. |
| MCO | Moody's Corp | 511.35 | HOLD | 511.35 | 486.35 | n/a | n/a | 549.99 (+7.6%) | Moody's continues to work: a fourth straight positive twelve-week reading and a solid gain, so the trim stays retired and we raise the stop instead. Retirement CONFIRMED. HOLD cv73, 12wk = +7.56%, position +4.66%. Not armed (Gain_ATR 2.14 < 3.0). Stop RAISED -> (scan's live level, 4.89% cushion). line. |
| BAC | Bank of America Corp | 62.33 | HOLD | 62.33 | 61.28 | n/a | 64.63 (+3.7%) | 68.65 (+10.1%) | Bank of America has recovered back above our stop and carries the best forward outlook of the book's large holdings - we hold. Breach RESOLVED (vs, +1.71%), vindicating the 08-20 refusal to chase the stop down. HOLD cv54, 12wk +10.1%, position +3.66%. Stop held (1.68% cushion). line, 3rd largest. |
| KHC | Kraft Heinz Co | 25.67 | HOLD | 25.67 | 24.15 | n/a | n/a | 26.23 (+2.2%) | Kraft Heinz has posted a fourth consecutive positive twelve-week reading, so the trim stays retired. Retirement CONFIRMED. 12wk = +2.18%, position +1.18%. Stop held (5.92% cushion); scan prints no stop. Quality_Pass False (gates new entries only). |
| AAPL | Apple Inc | 310.34 | HOLD | 310.34 | 306.15 | n/a | n/a | 302.73 (-2.5%) | Apple is quiet - marginally above cost, comfortably above its stop - so we leave it alone. Withdrawal CONFIRMED. 7th consecutive NO TRADE; +0.55% above cost, 1.37% above the stop, 12wk = -2.45%. NOT armed (Gain_ATR 0.27) - the row's PROFIT_ARMED flag is spurious, see systemic finding 2. Stop held. |
| GOOG | Alphabet Inc Cl C | 344.59 | NO ACTION | 344.59 | 320.20 | n/a | n/a | 306.52 (-11.0%) | The C shares of Alphabet face exactly the same capex and AI-leadership news as the A shares we are trimming tonight, but the model has not produced a sell signal on this line, so it is carried unchanged. CARRIED ON AN ABSENT LABEL, and it should not have been. Same underlying as GOOGL, same news, same 12wk direction (-11.1% vs -10.3%) - yet GOOG prints NO TRADE cv0 while GOOGL prints STRONG SELL - REDUCE cv75. Only of exposure. Flagged as systemic finding 3 (dual-class divergence; LLYVA/LLYVK show the same pattern at cv62 vs cv69). Stop held. · news |
| COF | Capital One Financial Corp | 216.27 | HOLD | 216.27 | 208.23 | n/a | n/a | 239.52 (+10.8%) | Capital One has recovered well clear of its stop with a strong forward view - hold. Breach RESOLVED (vs, +3.86%). 12wk +10.7%, position +5.44%. Stop held. |
| DAL | Delta Air Lines Inc | 82.48 | HOLD | 82.48 | 81.36 | n/a | n/a | 77.15 (-6.5%) | Delta has climbed back above its stop but its forward outlook remains the weakest of the recovered names - held on a short leash. Breach RESOLVED (vs, +1.38%). 4 consecutive NO TRADE, 12wk = -6.5%, position -11.07%. Stop held (1.36% cushion) - re-arms on a decisive break. |
| LLYVK | Liberty Live Group Ser C | 106.13 | HOLD | 106.13 | 101.22 | n/a | 111.04 (+4.6%) | 111.29 (+4.9%) | Liberty Live's C shares hold a constructive signal with modest upside to the model's targets - hold and raise the stop. HOLD cv69, 12wk +4.86%. Valuation NA (negative-EPV row) -> val_adj 0. Stop RAISED -> (scan level). |
| LLYVA | Liberty Live Group Ser A | 102.10 | HOLD | 102.10 | 97.57 | n/a | 106.95 (+4.8%) | 106.66 (+4.5%) | The A shares mirror the C shares' constructive read - hold. HOLD cv62, 12wk +4.47%. Valuation NA -> val_adj 0. Stop HELD (tighter than the scan's). Note the cv62 vs cv69 divergence against LLYVK on identical underlying - systemic finding 3. |
| KR | Kroger Co | 59.08 | HOLD | 59.08 | 54.87 | n/a | n/a | 50.38 (-14.7%) | A residual position - too small to trade. Immaterial. AVOID cv57. Scan stop prints ABOVE the close - a short-side stop on a long-only book - REJECTED for the 5th consecutive session (systemic finding 6). Recorded stop held. |
| NVR | NVR Inc | 6397.59 | HOLD | 6397.59 | 6088.57 | n/a | n/a | 5962.81 (-6.8%) | Three shares - too small to trade. Immaterial. AVOID cv57. Scan stop prints ABOVE the close - short-side stop, REJECTED (systemic finding 6). Recorded stop held. |
| M | Macy's Inc | 23.18 | HOLD | 23.18 | 22.49 | n/a | n/a | 21.82 (-5.9%) | A position - too small to trade; stop tightened. Immaterial. Breach 0.73% / 0.24 ATR - not decisive. Stop RE-BASED -> (close - 1.0 ATR). |
| LEN | Lennar Corp Cl A | 88.16 | HOLD | 88.16 | 82.64 | n/a | n/a | 82.53 (-6.4%) | A position - too small to trade. Immaterial. Stop held. |
| LEN-B | Lennar Corp Cl B | 86.77 | HOLD | 86.77 | 81.78 | n/a | n/a | 82.28 (-5.2%) | A position - too small to trade. Immaterial. Prints RSI_DIV on an AVOID cv0 row - noted, not actionable at this size. Stop held. |
| STZ | Constellation Brands Cl A | 137.60 | HOLD | 137.60 | 128.06 | n/a | n/a | 124.97 (-9.2%) | A position - too small to trade. Immaterial. Quality_Pass False. Stop held. |
| — | ACQUIRE slot | — | NO ACTION | — | — | n/a | n/a | n/a | No purchases today: for the first time in this book's history the model produced not a single buy signal anywhere in its 29-name universe. NO ACTION - empty buy pool, not a near-miss list. The 08-24 snapshot prints 14 NO TRADE, 5 AVOID, 6 HOLD, 4 STRONG SELL - REDUCE across 29 rows; zero BUY / BUY - ADD / STRONG BUY / BUY - ANTICIPATE. The 08-22 snapshot was identical. The book's only recent candidate CVX went BUY - ADD cv79 -> NO TRADE cv0 -> NO TRADE cv0 and now fires a profit-take instead. ANTICIPATE slot EMPTY for a 24th session. Highest non-sell labels for the record: VRSN HOLD cv75 (fires a profit-take), MCO cv73, LLYVK cv69, KO cv68, LLYVA cv62, BAC cv54 - HOLD is not a buy label. |
Outcome & track record
Accuracy at the time of this record.
129 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 →
-0.086 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.