Recommendations
1 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 |
|---|---|---|---|---|---|---|---|---|---|
| BAC | Bank of America Corp | 63.86 | ADD — QUEUED @61.95 | 61.95 | 61.19 | 62.64 (+1.1%) | 64.25 (+3.7%) | 75.94 (+22.6%) | Bank of America is the one name in the book the model still wants to own: it printed STRONG BUY - ADD at conviction 88 on a pullback-to-VWAP setup, closed at a fresh high 17% above its 200-day average, and carries the book's best forward view at +22.6% over twelve weeks. We are not chasing it — the close at is 1.9 ATR past the model's entry, so this goes in as a good-till-cancelled limit that only fills on a pullback to roughly where analysts peg fair value (consensus). Valuation is the one reservation: on a residual-income basis the shares screen rich, which trims a couple of points off conviction, but that gate blocks new positions, not additions to one we already own. OPERATIONAL: 3rd consecutive queue attempt (08-04 @61.32 and 08-05 @61.52 both expired unfilled, 08-06 withdrawn at cv58); entry repriced to tonight's live print per the provenance rule, not carried. PullbackBuy_VWAP|buy is this book's only actionable buy cell and it is NEGATIVE (-1, n=7, hit_4w.00) - L1/US-10 inherited from `us` say prefer this cell, this book's own scorecard says otherwise; the -1 is applied, the L1 preference is not. Quality_Pass=False does NOT gate an ADD to a held name. Trail deliberately HELD at rather than ratcheted to the chandelier - a stop below our own live bid is incoherent; re-raise once the queue fills or expires. Withdraw at the open if the signal family flips off PullbackBuy or conviction drops <65 (US-8). · news |
| AAPL | Apple Inc | 308.26 | SELL 30% | 308.26 | 300.00 | 308.74 (+0.2%) | n/a | 314.03 (+1.9%) | We are cutting a third of the book's largest holding because the model has now gone three straight sessions unable to read Apple at all, and its twelve-week outlook has collapsed from +3.6% to +1.9% in two sessions - close enough to flat that the forward case has effectively died. Monday supplied the reason: Jefferies downgraded the stock and cut its target from to after supply-chain checks showed Apple has cancelled the 20th-anniversary all-glass iPhone on poor production yields, with rising memory costs and a slow AI rollout on top. Six firms now carry sell-equivalent ratings, the most since 2012, and the shares fell about 2% on the news. The position is essentially flat on cost, so this is risk reduction rather than profit-taking, and we keep the other 70%. OPERATIONAL: pre-committed in the 08-07 entry ('a 3rd NO TRADE with the 12wk turning negative makes AAPL a mandatory dispose'). 12wk is +1.87%, technically positive, but INSIDE the tick-(c) dead-view band (314.03 < 1.02 x 308.26 = 314.43), so the condition is met on the framework's own definition - stated as an assumption. Admitted under CLAUDE.md 9 sustained-NoTrade, NOT ADR-0012 profit-take: Gain_ATR is -0.04 and the PROFIT_ARMED flag is the inverted T12_Progress limb (Systemic finding 1). Tier held at the 30% minimum for that reason and because price sits 9.9% above EMA200. Stop HELD at the event level (2.0xATR chandelier computes 290.88 - lower; trails ratchet up only). Remaining. Ex-div went ex 08-10. · news |
| MCO | Moody's Corp | 478.14 | SELL 30% | 478.14 | 461.27 | 478.32 (+0.0%) | n/a | 480.25 (+0.4%) | Moody's has printed two consecutive sessions the model cannot read, and its twelve-week projection has flatlined at +0.4% - there is no forward return left in the position at this price. The business is not the problem: Q2 revenue grew 15%, earnings per share rose 31%, and the buyback was lifted to $3bn. What changed is that management cut 2026 operating-margin guidance, and the shares at ~29.5x earnings are already priced above the peer group, which is exactly what a dead forward view looks like. The stock also sits barely 1% above its 200-day average, so the trend support underneath it is thin. We take a third off and keep the rest. OPERATIONAL: flagged in the 08-07 entry as 'the most likely card of the next two sessions after AAPL' - the second NO TRADE print arrived on schedule. Admitted under 9 sustained-NoTrade (same basis ALLY got on 08-07 at 2 prints); NOT armed (Unreal -2.14%, Gain_ATR -1.02) so ADR-0012 does not apply and the tier stays at 30%. G2 SUPPRESSES the valuation-amplify path (Target_12Week 480.25 > DCF anchor 392.22 = unreliable), so OVERVALUED_VS_RI contributes nothing to the decision - the signal originates alone. Ex-div on 08-14 (3 sessions out) is mechanical (0.22%), NOT an information catalyst, so the catalyst-aware hold does not apply. Stop HELD at (chandelier 457.68 is lower). Remaining. · news |
| NYT | New York Times Co Cl A | 64.80 | SELL 50% | 64.80 | 58.27 | 63.26 (-2.4%) | n/a | 46.27 (-28.6%) | The New York Times has re-armed the sell case we withdrew three sessions ago: it is back to STRONG SELL - REDUCE at conviction 68 and ranks first in the book on sell priority, trading 9.8% below its 200-day average with a twelve-week projection of against today's - a 29% downside view. The business is fine; the market has simply re-rated it. Q2 revenue rose 11%, digital subscription revenue 16% and digital advertising 21%, but the company added 280,000 digital subscribers against the 295,000 expected and guided to persistent cost pressure, and the shares fell 13% on the print and are down about 13% over four weeks. We halve the position rather than exit: at the stock is sitting almost exactly on our intrinsic estimate of and the momentum reading is already oversold, so this is de-risking a broken trend, not calling the business over. OPERATIONAL: the 08-07 card set the re-arm trigger explicitly ('a 2nd SELL-REDUCE print or a close below 56.44') and the 2nd print landed. Only sell candidate tonight clearing the 65 floor on its own merits (68 - 2 = 66). Cell is FailedBreakout_Down_20D|sell (-2, n=34), NOT the book's toxic RallySell_BearTrend|sell (-8) which governed the 08-07 withdrawal - that objection no longer applies. 50% not 88.52% (engine) and not 100%: RSI 34.3 oversold, DCF anchor 63.20 spot, and 30% was already banked at 64.75 on the 08-06 open. Stop RAISED 56.44 -> (2.0xATR chandelier = model Stop_Price). Remaining. · news |
| DVA | DaVita Inc | 183.69 | NO ACTION | 183.69 | 174.64 | 180.60 (-1.7%) | n/a | 146.61 (-20.2%) | Fourth straight session the model cannot read DaVita, and the twelve-week projection has collapsed from to in six sessions - a 20% downside view on a position already down 22%. Qualifies as a sell on sustained no-read plus a dead forward view. Held back this session because the position was already halved at the 08-07 open and the remaining half was explicitly tied to a break of the 200-day average at, which has not happened (close). OPERATIONAL: ranked 3rd by dollar bleed but displaced by NYT on the un-met EMA200 re-trigger the book wrote down itself - same discipline that vindicated the CVX and OXY deferrals. Mandatory next EOD if EMA200 breaks or a 5th NO TRADE prints. Stop. |
| ALLY | Ally Financial Inc | 43.79 | NO ACTION | 43.79 | 42.63 | 43.58 (-0.5%) | n/a | 41.24 (-5.8%) | Third consecutive session with no readable signal and a twelve-week view that has deteriorated again to -5.8%, with the shares only 2.5% above their 200-day average. The sell case is intact and was already approved on 08-07 but never executed. Valuation argues the other way - Ally is one of only two names in the book screening undervalued - which caps this at the minimum 30% trim rather than anything larger. OPERATIONAL: this is the 08-07 card unexecuted, re-priced from 43.72 to tonight's 43.79 per the provenance rule. Ranked 5th by dollar bleed, crowded out by the 3-card cap. UNDERVALUED (MoS +0.05) argues the tier DOWN one step per 9(b) but 30% is already the floor. Mandatory next EOD. Stop. |
| OXY | Occidental Petroleum Corp | 58.65 | NO ACTION | 58.65 | 54.36 | 58.75 (+0.2%) | n/a | 59.89 (+2.1%) | A genuine profit-take has fired: the position is up 20% (4.5 ATR above cost) after a 4.9% jump on Monday's energy rally, and a fresh momentum-divergence warning appeared tonight for the first time - strength on deteriorating internals. On our intrinsic work the shares are the most stretched in the book, which pushes the trim from a third to a half. Deferred only because three larger positions took the day's slots. OPERATIONAL: armed on the REAL limb (Gain_ATR 4.54, +19.91%) + tick (a) RSI_DIV, which is NEW tonight - not the same unchanged tick the 08-07 entry declined to re-fire on. G1 does not soften (label is AVOID cv0, not a fresh cv>=80 breakout). MoS_FV -3.07 <= -3 bumps the tier per 9(b); G2 does NOT suppress (DCF anchor 101.91 sits well ABOVE the 59.89 12wk target, so the anchor is reliable). Mandatory next EOD. Trail RAISED 51.96 ->. |
| VRSN | VeriSign Inc | 293.34 | NO ACTION | 293.34 | 276.41 | 296.07 (+0.9%) | n/a | 326.10 (+11.2%) | Up 14% and armed for a profit-take, with a fourth consecutive session the model cannot read it. Deferred for a fifth time because the only thing wrong is that the engine has no view - every fundamental line still points up, and the twelve-week projection is the second-best in the book at +11%. OPERATIONAL: armed on the real limb (Gain_ATR 4.36) + tick (d) only. 5th consecutive deferral - the counterfactual series says deferred trims in this book cost a median +4.19% at 4wk (n=33), so this is a measured cost, not a free option (Systemic finding 2). Ex-div on 08-19. Trail RAISED 276.38 ->. |
| NVR | NVR Inc | 6292.28 | NO ACTION | 6292.28 | 6010.84 | 6245.76 (-0.7%) | n/a | 5734.06 (-8.9%) | Five consecutive sessions of STRONG SELL - REDUCE, trading 7.5% below its 200-day average with a -8.9% twelve-week view after a Q2 that missed badly. This is the cleanest sell signal in the book and has been carried unexecuted for eight sessions purely because the position is worth - a line can never outrank a one on dollar bleed. OPERATIONAL: adj 55 after the RallySell_BearTrend|sell -8 haircut, admitted independently by stable-REDUCE + L2 (below EMA200 + negative 12wk). 8th session pending - this is Systemic finding 2's materiality-floor case in its purest form. Stop. |
| STZ | Constellation Brands Cl A | 133.39 | NO ACTION | 133.39 | 126.89 | 131.55 (-1.4%) | n/a | 111.24 (-16.6%) | Back to STRONG SELL - REDUCE, 8.6% below its 200-day average with a -17% twelve-week view on falling revenue. A dust line carried as a standing close-out. OPERATIONAL: conviction 38, adj 36 - far below the 65 floor - but admitted by L2 (decisive sub-EMA200 + negative 12wk) as a carried close-out, not a fresh card. Stop. |
| AXP | American Express Co | 338.86 | HOLD | 338.86 | 334.75 | n/a | n/a | 339.37 (+0.2%) | First STRONG SELL - REDUCE print at conviction 38 on the book's second-largest line, but a single low-conviction flip with the shares still above their 200-day average, a flat twelve-week view and no negative news - Q2 revenue +10%, 2026 guidance raised to 10% growth, dividend up 16%. Not actionable yet; a second print makes it the dominant card. OPERATIONAL: adj 36 after the -2 cell haircut, far below floor. Close sits only above the trail - a breach would trigger L2 on its own. |
| CB | Chubb Ltd | 348.30 | HOLD | 348.30 | 341.96 | n/a | n/a | 360.71 (+3.6%) | Range-bound HOLD, +3.6% 12wk view, 7.9% above EMA200. Trail (chandelier 335.24 is lower). |
| COF | Capital One Financial Corp | 218.97 | HOLD | 218.97 | 209.77 | n/a | n/a | 255.07 (+16.5%) | Recovered NO TRADE -> HOLD cv54, +16.5% 12wk view, +6.75% unrealized. Trail (chandelier 209.49 is lower). |
| CVX | Chevron Corp | 194.91 | HOLD | 194.91 | 184.28 | n/a | n/a | 202.47 (+3.9%) | The 08-07 profit-take deferral is VINDICATED: +4.5% Monday, conviction, and tick (c) has cleared (12wk 202.47 now above the 1.02x threshold of 198.81). No longer fires. Armed (Gain_ATR 4.84, +15.2%) but not fired -> mandatory trail-raise 179.56 ->. Ex-div on 08-19. |
| DAL | Delta Air Lines Inc | 89.21 | HOLD | 89.21 | 86.70 | n/a | n/a | 103.82 (+16.4%) | Third NO TRADE but the forward view is the book's third-best at +16.4% and price is 21% above EMA200 - G3 blocks any trim. Trail. |
| KHC | Kraft Heinz Co | 24.93 | HOLD | 24.93 | 24.15 | n/a | n/a | 26.30 (+5.5%) | HOLD cv64 -> 52, +5.5% 12wk view. Stop held (chandelier 23.09 lower). |
| KO | Coca-Cola Co | 86.87 | HOLD | 86.87 | 83.81 | n/a | n/a | 95.21 (+9.6%) | One NO TRADE print, +9.6% 12wk view, 11.5% above EMA200. Trail raised 83.78 ->. |
| KR | Kroger Co | 56.48 | HOLD | 56.48 | 54.87 | n/a | n/a | 44.52 (-21.2%) | AVOID cv61, -21.2% 12wk view, 12.3% below EMA200 - a dust close-out awaiting a sell label. Stop. |
| LEN | Lennar Corp Cl A | 85.60 | HOLD | 85.60 | 82.64 | n/a | n/a | 76.49 (-10.6%) | dust; -10.6% 12wk, 12.6% below EMA200. Stop. |
| LEN-B | Lennar Corp Cl B | 83.82 | HOLD | 83.82 | 81.78 | n/a | n/a | 75.74 (-9.6%) | dust; -9.6% 12wk, 10.9% below EMA200. Stop. |
| LLYVA | Liberty Live Group Ser A | 100.42 | HOLD | 100.42 | 94.88 | n/a | n/a | 101.45 (+1.0%) | Recovered cv0 -> 75; valuation permanently NA (negative book anchor) so fusion is technical-only. Trail raised 93.90 ->. |
| LLYVK | Liberty Live Group Ser C | 104.46 | HOLD | 104.46 | 99.04 | n/a | n/a | 106.83 (+2.3%) | Recovered cv0 -> 75; valuation NA. Trail raised 98.01 ->. |
| M | Macy's Inc | 24.90 | HOLD | 24.90 | 24.57 | n/a | n/a | 29.44 (+18.2%) | One NO TRADE print, +18.2% 12wk view, +7.1% unrealized. Trail (chandelier 23.02 lower). |
| NUE | Nucor Corp | 274.61 | HOLD | 274.61 | 259.19 | n/a | n/a | 351.06 (+27.8%) | The book's best position (+24.4%, Gain_ATR 6.99) with a +27.8% 12wk view - armed but NO tick fired, so trail-raise not trim. 255.59 -> 256.22 ->. |
| SIRI | Sirius XM Holdings Inc | 28.78 | HOLD | 28.78 | 28.60 | n/a | n/a | 30.09 (+4.6%) | AVOID cv52 - not an actionable sell label. Close is only above the stop, but of the 3.8% two-day fall is the 08-10 ex-dividend, so the real decline is ~2.9%. +4.6% 12wk view. Watch for a breach. |
Outcome & track record
Accuracy at the time of this record.
79 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 →
0.006 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.