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 |
|---|---|---|---|---|---|---|---|---|---|
| — | No acquisition | — | NO ACTION | — | — | n/a | n/a | n/a | No acquisition tonight. Not one of the twenty-nine names in this book produced a buy signal of any kind on the close — the universe is entirely holds, cautions and sells, for the sixth session running, so the right answer is to buy nothing and let cash build. OPERATIONAL: zero STRONG BUY / BUY / BUY - ADD / BUY - ANTICIPATE rows in the 09-01 snapshot (9 NO TRADE, 6 HOLD, 4 AVOID, 8 STRONG SELL - REDUCE, plus unheld LPX/JEF both AVOID). No ANTICIPATE slot either — no name printed BUY - ANTICIPATE in >=2 of the last 3 snapshots. Highest held conviction is KHC cv85 (HOLD, and a fired profit-take tonight), then AAPL cv83 (HOLD - trail), OXY 75, VRSN 69 — none carries a buy label. (49.9% of NAV) untouched; cash rises to = 65.4% of NAV after tonight's sells. See systemic finding 6. |
| AXP | American Express Co | 324.19 | SELL 88.52% | 324.19 | 313.27 | n/a | n/a | 295.05 (-9.0%) | Selling most of American Express. It is the book's largest position and it has now broken down on every measure that matters: the shares closed below our risk line, below their long-term trend line, and the model's three-month view points nearly 9% lower. The stock is down about 12% this year and has just printed its first outright sell signal after four sessions of caution, so we are stepping out of the great majority of it and keeping a small residual. Valuation offers no rescue — it screens as expensive against its own returns on capital. OPERATIONAL: exit is originated by the trend/risk trio, NOT by the label — close vs recorded stop = -1.93% / -1.17 ATR (decisive), 12wk = -8.99%, and close is below EMA200. That is the CLAUDE.md section 9 trend-based exit, explicitly unaffected by the trim-discipline gates. The STRONG SELL - REDUCE cv63 label sits on RallySell_BearTrend, tonight's -8 bias cell (1 hit in 12, n=12 ACTIONABLE), so the label alone gives adj 55 and would FAIL the 65 floor — stated explicitly per the strongly-negative-cell rule. Valuation MoS -0.454 is nowhere near the -3 tier bump so portion stays at the scan's 88.52%; size x1.0. RSI 35.7, position -7.89% on cost (- unrealised). Leaves with a re-based stop at the scan's. No earnings inside the 10-day Event Horizon window — no catalyst-aware hold. WORK THE PRINT: is ~10% of NAV; use VWAP / participation, not one market order. · news |
| CB | Chubb Ltd | 338.74 | SELL 88.52% | 338.74 | 327.02 | n/a | n/a | 317.58 (-6.2%) | Selling most of Chubb — a decision first taken on 27 August that has still not been executed, so it is carried and repriced rather than re-litigated. The shares have spent five straight sessions below the risk line that triggered the exit and the model has had nothing positive to say about them for four sessions running, while the three-month view is about 6% lower. The insurer itself is fine — the business is profitable and analysts are constructive — but the position has stopped working and the discipline is to respect the line. OPERATIONAL: L2/L3 carried. The pre-cut risk line still governs because the cut never happened (executor blocker, 12 sessions); close is / -0.65 ATR below it, and 12wk = -6.25%. Against the post-cut re-based there is no break — the basis is the un-repaired pre-cut line, stated openly. 4 consecutive NoTrade cv0. MoS -0.006 (essentially at fair value) -> val_adj 0, no tier bump, portion stays 88.52%, size x1.0. Ex-dividend on Sep 11 is 7 trading days out, outside the 5-session catalyst window. CANCEL if CB opens above. Residual keeps the stop. · news |
| DAL | Delta Air Lines Inc | 76.38 | SELL ALL | 76.38 | 81.36⚠ | n/a | n/a | 62.56 (-18.1%) | Closing Delta entirely. This exit was approved on 28 August at and never executed; the shares have fallen another 4.6% since, and the position is now 17.7% under cost. The stock is well below its stop, has gone five sessions without a single constructive reading, and the model's three-month view is 18% lower. The airline's own outlook is not the issue — cash flow guidance was affirmed and the dividend was raised — but the position has been bleeding for weeks and the rule is to cut it, not to keep writing it down. OPERATIONAL: third consecutive recommendation; L3 forbids a fourth. Decisive break — close vs recorded stop = -6.12% / -2.07 ATR, the deepest breach in the book. 12wk = -18.09%; RSI 29.8; 5x NoTrade cv0. Full exit, not a trim: the skill reserves 100% for exactly this (decisive stop break) and there is no bullish element left to retain. Valuation FAIR_VALUE MoS -0.846, Quality_Pass True, DCF sits 53% ABOVE the close — this is a risk exit against a fundamentally cheap-looking name; section 9 bars valuation from vetoing it. Trail-erosion check (ADR-0020): stop vs peak = 9.74%, under the 15% bar — Trail Erosion did NOT fire, this is purely signal-based. No earnings in the window. Position closes, stop retired. · news |
| NUE | Nucor Corp | 251.92 | NO ACTION | 251.92 | 248.53 | n/a | n/a | 254.77 (+1.1%) | Nucor has turned into an outright sell and we intend to step out of it, but it lost the last exit slot to three much larger positions. The steelmaker is 14% ahead of cost and the model now says the next three months are worth essentially nothing, so the profit is worth banking. OPERATIONAL: this is the 08-28 50% card RE-ISSUED AND UPGRADED to the scan's 88.52% — the case strengthened from a bare profit-take into a labelled STRONG SELL - REDUCE cv65 on FailedBreakout_Down_20D, whose cell is +0 bias (n=49, hit.51, ACTIONABLE), so adj 65 CLEARS the floor cleanly. ADR-0012 tick (c) also still fires: 12wk < 1.02 x close. Genuinely armed (Gain_ATR 4.06, +14.12%). MoS -2.278 amplifies but misses the -3 tier so no further bump. Slot-capped at of bleed against AXP / CB / DAL — ledgered optional so the counterfactual grader prices the deferral. First in the queue tomorrow. Stop held at (scan is lower). |
| MCO | Moody's Corp | 494.71 | NO ACTION | 494.71 | 493.86 | n/a | n/a | 531.94 (+7.5%) | A 30% trim in Moody's is warranted but lost the last slot. The model has gone silent on it for two sessions in a row while the position sits barely above cost and right on top of its stop — a small de-risk is prudent even though the three-month view is still the healthiest in the book at +7.5%. OPERATIONAL: ADR-0012 profit-take, ARMED via T12_Progress 93.0 on a thin +1.25% gain (Gain_ATR 0.56), FIRED on tick (d) — two consecutive NoTrade prints (08-31, 09-01). One tick -> profit_take_1 = 30%. The 08-28 tick (a) RSI_DIV has cleared; tonight PROFIT_ARMED is the only warning code. G2 DCF-reliability gate SUPPRESSES any valuation amplification: Target_12Week vastly exceeds DCF_per_share, so the MoS -1.148 reading cannot amplify. Close is only 0.17% above the stop — this becomes an L2 candidate the moment that breaks. Slot-capped at; ledgered optional. |
| SIRI | Sirius XM Holdings Inc | 27.62 | NO ACTION | 27.62 | 27.81⚠ | n/a | n/a | 24.44 (-11.5%) | Sirius XM has slipped below its stop and now carries an outright sell signal, with the model projecting another 11% lower over three months. The position is already 9.5% under cost and has not produced a single constructive reading in five sessions. We want out of most of it, but it lost the last slot to bigger, more decisively broken positions. OPERATIONAL: label is STRONG SELL - REDUCE cv63 on RallySell_BearTrend = the -8 ACTIONABLE cell (1 hit in 12), so adj 55 FAILS the 65 floor on the label alone. The independent L2 basis is real but SHALLOW — close vs recorded stop is only -0.68% / -0.30 ATR, a graze rather than the decisive break AXP (-1.17 ATR) and DAL (-2.07 ATR) print, and the close is still ABOVE EMA200. That combination is what kept it out of the top 3 despite of bleed ranking it 3rd by dollars. 12wk = -11.51%; RSI 36.8; Quality_Pass False. Ledgered optional; re-arms to a firm card on a decisive close below or a second sell print. |
| KHC | Kraft Heinz Co | 25.87 | NO ACTION | 25.87 | 24.44 | 26.30 (+1.7%) | 27.30 (+5.5%) | 26.18 (+1.2%) | A 30% trim in Kraft Heinz is technically due — the model's three-month projection has gone dead flat against the current price — but this is a marginal case and it lost the slot. The position is only 2% ahead of cost, so there is very little profit actually being protected, and the shares carry the highest confidence reading in the book today. OPERATIONAL: ADR-0012 ARMED via T12_Progress 98.82 on a thin +1.97% gain (Gain_ATR 0.70 — economically trivial arming), FIRED on tick (c): 12wk < 1.02 x close. One tick -> 30%. Tick (d) does NOT fire — the NoTrade run was broken by tonight's HOLD cv85. G1 not engaged (FailedBreakdown_Up_20D is neither BreakoutUp_* nor MomentumContinuation_*), and tick (c) fires regardless of G1 in any case. MoS -0.665 misses the -3 tier; Quality_Pass False (blocks new entries only, never a held-name action). Slot-capped at Stop RAISED -> (scan chandelier). |
| LLYVK | Liberty Live Group Ser C | 101.08 | NO ACTION | 101.08 | 103.21⚠ | n/a | n/a | 100.02 (-1.0%) | Liberty Live C has dropped below its stop and the model has gone quiet on it after three constructive sessions. The three-month view is only marginally negative, so this is a watch-and-arm rather than a firm exit tonight. OPERATIONAL: L2 fires but only marginally — close vs stop = -2.06% / -0.86 ATR (decisive on ATR terms) BUT 12wk is just -1.05%, far short of the conviction the AXP (-8.99%) / DAL (-18.09%) cases carry. Conviction is a flip INTO NoTrade and is NOT treated as tick (b) (NoTrade rows print cv0 by construction; NoTrade deterioration is governed by tick (d) only — same reading applied since 08-26). Valuation NA (tracking stub) -> val_adj 0, quality gate off. Ledgered optional; escalates to a firm card on a second NoTrade print or a 12wk below -5%. |
| LLYVA | Liberty Live Group Ser A | 97.38 | NO ACTION | 97.38 | 99.29⚠ | n/a | n/a | 95.78 (-1.6%) | Liberty Live A mirrors the C-share read — below its stop, model gone quiet, three-month view only slightly negative. Watch-and-arm, not a firm exit. OPERATIONAL: close vs stop = -1.92% / -0.82 ATR; 12wk = -1.64%. Same NoTrade/tick-(b) interpretation as LLYVK. Valuation NA. Ledgered optional at; escalates on a second NoTrade print or a deeper 12wk. |
| M | Macy's Inc | 21.93 | NO ACTION | 21.93 | 22.49⚠ | n/a | n/a | 18.73 (-14.6%) | Macy's should be closed out — it is below its stop with the model projecting nearly 15% lower over three months — but at roughly it is one hundredth of one percent of the book and spending a scarce exit slot on it instead of on an eleven-million-dollar position would be poor allocation. OPERATIONAL: textbook L2 — close vs stop = -2.49% / -0.81 ATR, 12wk = -14.59%. Now prints NoTrade cv0 after two sessions of STRONG SELL - REDUCE cv61, so there is no live label to haircut; the basis is the stop break alone. Earnings Sep 10 — inside the 10-day Event Horizon window; clear this position BEFORE the print rather than holding a broken sub-scale name into a catalyst. Ledgered optional; carried forward as first-line housekeeping. |
| GOOGL | Alphabet Inc Cl A | 335.02 | HOLD - exit armed | 335.02 | 333.68 | n/a | n/a | 292.28 (-12.8%) | Alphabet has flipped to an outright sell signal and the model's three-month view is nearly 13% lower, which on a position is the largest unaddressed risk in the book after tonight's three exits. We are not acting yet: the shares are still above both their stop and their long-term trend line, and the sell signal comes from a pattern that has been wrong eleven times out of twelve here. A close below converts this into a firm exit. OPERATIONAL: adj conviction 55 FAILS the 65 floor after the -8 RallySell_BearTrend haircut, and no independent L2 basis — close is +0.40% above stop and +2.35% above EMA200. That margin is razor-thin: this is the book's primary re-arm trigger. NEWS is a genuine CONCERN and does not favour waiting long — a sweeping AI leadership overhaul (including Jeff Dean's departure) drove a sharp decline, and EU DMA / AI-content investigations remain live; note the cited article's price level does not reconcile with our close, so it is treated as directional colour, not a price fact. Ex-dividend Sep 04. Stop held (scan is lower). · news |
| KO | Coca-Cola Co | 88.00 | HOLD - trail rebased | 88.00 | 85.90 | 89.26 (+1.4%) | 92.21 (+4.8%) | 96.05 (+9.1%) | Holding Coca-Cola and widening the trailing stop rather than selling. The shares dipped through our stop, but that stop was set far too tight — less than one percent below the price — on a name that had just made an all-time high, is up about 30% this year, and still carries the best three-month outlook in the book at +9%. The pullback is normal, not a breakdown. The error was the stop, not the position. OPERATIONAL: monitor flags a breach (close vs trail = -1.19% / -0.73 ATR) but L2 does NOT fire — it requires a negative 12wk and 12wk is +9.15%. Signal is PullbackBuy_VWAP, the book's most reliable buy cell, HOLD cv62 with conviction rising. ADR-0012: ARMED (Gain_ATR 2.64, T12_Progress 91.62, +4.59%) but ZERO ticks -> trail, no trim. JUDGMENT CALL RECORDED: the trail was set 0.67% below the then-close against a ATR — indefensibly tight. Deliberately RE-BASED to the scan's chandelier; peak_stop retained at, so ADR-0020 erosion is 3.55%, far under the 15% bar. Ex-dividend Sep 15. Conviction 62 is below the 65 floor so this is not an add either. · news |
| CVX | Chevron Corp | 211.05 | HOLD - trail raised | 211.05 | 201.58 | 218.32 (+3.4%) | 235.27 (+11.5%) | 242.28 (+14.8%) | Holding Chevron and raising the stop. It is one of the book's two best positions at nearly 25% above cost, momentum is intact and the three-month view is +15% — there is no case for banking it here. OPERATIONAL: ARMED (Gain_ATR 9.94) but NOT fired — conviction is a 2-pt drift, far under the 20-pt bar; 12wk is well clear of tick (c). Trail RAISED -> (scan chandelier is now higher). NOT an acquire: the signal family flipped PullbackBuy_VWAP -> MomentumContinuation_Up and conviction 63 is below the 65 floor; RSI 70.8 with Entry_Distance_ATR -0.25 also fails the no-chase bar. The 08-27 ADD queue stays withdrawn (US-8). |
| OXY | Occidental Petroleum Corp | 60.95 | HOLD - trail raised | 60.95 | 58.02 | 61.83 (+1.4%) | 63.88 (+4.8%) | 68.16 (+11.8%) | Holding Occidental and raising the stop. At nearly 25% above cost it is the book's other standout, the model's confidence is the second-highest here, and the three-month view is +12%. Nothing to bank. OPERATIONAL: ARMED (Gain_ATR 8.23) but NOT fired — conviction flat at 75, 12wk far clear of tick (c). Trail RAISED ->. Not an acquire: the label is HOLD, and FailedBreakdown_Up_20D|sell carries a -6 ACTIONABLE bias (n=5, hit.20) which underlines that this family is not a reliable trade trigger on this book in either direction. |
| VRSN | VeriSign Inc | 288.37 | HOLD - trail | 288.37 | 280.81 | 293.05 (+1.6%) | 303.96 (+5.4%) | 301.41 (+4.5%) | Holding VeriSign, 12.5% ahead, stop unchanged. The model has turned constructive on it again after two quiet sessions and the three-month view is positive; there is no reason to cut a winner. OPERATIONAL: ARMED (Gain_ATR 4.10) but NOT fired — conviction is recovering, 12wk clears tick (c). Scan Stop_Price is BELOW the recorded; trail never lowered — hold. |
| BAC | Bank of America Corp | 61.99 | HOLD | — | 61.28 | n/a | n/a | 65.26 | Holding Bank of America. The model has gone quiet after a run of cautious readings, but the position is 3% in profit, the three-month view is a constructive +5%, and the shares are above their stop. OPERATIONAL: conviction is a flip INTO NoTrade, NOT treated as tick (b). ARMED via T12_Progress 94.99 on +3.09% but zero ticks -> trail only. Close is just 1.15% above the trail — the book's second-tightest hold after MCO. NoTrade rows print no Stop_Price; trail held. |
| GOOG | Alphabet Inc Cl C | 332.03 | HOLD - exit armed | 332.03 | 320.20 | n/a | n/a | 286.31 (-13.8%) | Holding the small Alphabet C-share stake alongside the larger A-share position — same read, same trigger, no action tonight. OPERATIONAL: mirrors GOOGL exactly; adj 55 fails the floor after the -8 RallySell_BearTrend haircut, close is above both stop and EMA200. 12wk = -13.77%. Ex-dividend Sep 04. Stop held (scan is lower). |
| NVR | NVR Inc | 6275.82 | HOLD | 6275.82 | 6088.57 | n/a | n/a | 5814.55 (-7.3%) | Holding the three-share NVR stake. It has flipped to a sell reading, but on a pattern this book has been wrong on eleven times out of twelve, and the shares are still above their stop. OPERATIONAL: adj 55 fails the 65 floor; no independent basis (close is 3.07% above stop), though it is now BELOW EMA200 with a 12wk of -7.35% — half of an L2. Re-arms on a decisive close below. Quality_Pass False. Stop held (scan is lower). |
| LEN-B | Lennar Corp Cl B | 81.40 | HOLD | 81.40 | 81.78⚠ | n/a | n/a | 72.88 (-10.5%) | Holding the nine-share Lennar B stake — the highest-ranked sell in the book by the model, and worth Not worth a slot or a ticket. OPERATIONAL: STRONG SELL - REDUCE cv71 Sell_Rank 1, but adj 63 fails the floor after the -8 haircut. Stop grazed (close vs = -0.18 ATR — not decisive) with 12wk = -10.47% and close below EMA200; it is a genuine L2-shaped case that survives only on immateriality. Bundle with LEN as housekeeping when a slot is free. |
| LEN | Lennar Corp Cl A | 82.75 | HOLD | 82.75 | 82.64 | 83.09 (+0.4%) | 83.87 (+1.4%) | 73.44 (-11.3%) | Holding the 34-share Lennar A stake; same weak housebuilder read as the B shares, same immateriality. OPERATIONAL: AVOID cv53 -> adj 45, well below the floor. Close is 0.13% above the stop and below EMA200; 12wk = -11.25%. Scan Stop_Price prints ABOVE the close and is REJECTED (see systemic finding 3); trail held at. |
| NYT | New York Times Co Cl A | 67.73 | HOLD | 67.73 | 64.15 | n/a | n/a | 54.09 (-20.1%) | Holding the New York Times. A dormant sell case — five straight cautionary readings and the worst three-month projection in the book — but the shares are still comfortably above their stop and the position is small. OPERATIONAL: 5 consecutive AVOID, conviction 49 well below the floor. 12wk = -20.14%, worst in the book. Scan Stop_Price prints ABOVE the close and is REJECTED — 7th consecutive session. Trail held at; becomes an L2 candidate the moment that breaks. |
| DVA | DaVita Inc | 176.14 | HOLD | — | 169.18 | n/a | n/a | 107.73 | Holding DaVita on its stop. It is the book's worst position at 25% under cost and the three-month projection is severe, but the shares are still above the stop and there has been no sell signal since the one that appeared and evaporated on 27 August — we do not chase a signal that un-fired. OPERATIONAL: 5 consecutive NoTrade cv0. NOT armed despite the PROFIT_ARMED print — position is -25.02% / -12.30 ATR and T12_Progress 163.5 is the degenerate-ratio artifact in systemic finding 1. MoS -3.376 is the most extreme in the book and would carry a tier bump, but valuation cannot ORIGINATE a sell (CLAUDE.md section 9) and there is no Exit_Warning code to elevate. Close is only 0.41% above EMA200 — an L2 sets up fast if goes. Stop held. |
| ALLY | Ally Financial Inc | 41.85 | HOLD | — | 41.56 | n/a | n/a | 35.99 | Holding Ally Financial. It screens as the cheapest name in the book on intrinsic value, but the model has been silent on it for five sessions and it sits 7.6% under cost — cheap is not a buy signal without a trigger. Just above its stop, so we wait. OPERATIONAL: 5 consecutive NoTrade. UNDERVALUED MoS +0.075 but Quality_Pass FALSE, so a new entry would be blocked by the quality gate anyway (ADDs to held names are never blocked). NOT armed (spurious PROFIT_ARMED; -7.56%). 12wk = -14.00%; close is 0.70% above the trail and 1.91% below EMA200 — an L2 fires if the stop goes. Stop held. |
| KR | Kroger Co | 57.94 | HOLD | — | 54.87 | n/a | n/a | 50.86 | Holding the small Kroger stake, flat on cost and quiet from the model. No action. OPERATIONAL: 3 consecutive NoTrade cv0; UNDERVALUED MoS +0.033 with Quality_Pass True — the only name in the book that both screens cheap and passes quality, so it is the first place to look if a buy signal ever appears here. Close is 8.4% below EMA200 and 12wk = -12.21%, so the technical picture does not support acting on the valuation yet. Stop held. |
Outcome & track record
Accuracy at the time of this record.
149 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.