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 |
|---|---|---|---|---|---|---|---|---|---|
| CVX | Chevron Corp | 205.76 | WATCH — queued order lapsed | — | — | n/a | n/a | n/a | Queued buy @207.18 lapsed — closed 205.76 below the level; awaiting reprice at the next end-of-day run. Chevron is the strongest buy signal this book has produced in a month: a fresh MomentumContinuation_Up breakout at conviction 93, Buy_Rank 1, with the model's buy-stop sitting just above the close rather than chasing an extended move. The fundamentals corroborate — record Q2 earnings of 1bn and 4bn of free cash flow helped by Hess synergies, a new 600-metre hydrocarbon discovery offshore Angola, and a 20-year 2.67GW power agreement with Microsoft that reframes part of the story as AI infrastructure. Our own valuation marks it near fair value, so we are paying up only modestly for momentum. The main caution is that Barclays just trimmed its target to essentially where we are buying, so the four-week objective needs oil to cooperate. OPS: buy-stop @ (band -), = 99.33% x 1,104 held, size_mod 1.0x (MoS -0.77 above the -1 band), against. val_adj -4; MomentumContinuation_Up|buy cell is report-only (n=2, 0/2 at 1wk and 4wk) so no bias applied - flagged, not adjusted. Position is armed (Gain_ATR 9.21) with RSI_DIV, but G1 softening applies (fresh momentum print at cv93 >= 80) so the Tier-1 warning is a trail-raise, not a trim: stop raised ->. L8: executes at the 08-20 open; reprice if it gaps >1.5%. · news |
| AXP | American Express Co | 339.90 | SELL 50% | 339.90 | 334.75 | 339.90 (+0.0%) | n/a | 332.25 (-2.3%) | We are cutting half of American Express because the model's twelve-week view keeps getting worse, not better: it now projects 2.25% below the current price, widening for a third straight session, after two sessions of an outright bearish breakdown label. The franchise itself is fine - second-quarter EPS of against, full-year revenue growth guidance lifted to 10%, a 16% dividend increase - and that is precisely why we are selling half rather than all of it. This is the book's single largest position and its largest source of drag, so trimming it is a risk-line decision about concentration and a deteriorating forward view, not a verdict on the company. OPS: 4TH ISSUANCE - the 08-17/08-18 cards never filled (executor blocked). Action price = the 08-19 close (NoTrade row prints no Entry_Price), repriced from (+0.41%); band -. = 50% x 34,659 ->, #1 dollar bleed at 11.7% of NAV. Label lapsed AVOID cv51 -> NO TRADE cv0, so the governing cell is now NoTrade|sell (n=35, hit_4w 0.286 - the book's weakest actionable); basis rests on the widening negative 12wk + 5 sessions of non-actionable prints + L3. MoS -0.46 is nowhere near the -3 tier-bump line, so valuation up-ranks only. Stop held on the retained. · news |
| CB | Chubb Ltd | 340.92 | SELL 30% | 340.92 | 341.96⚠ | 343.44 (+0.7%) | n/a | 330.76 (-3.0%) | Chubb has closed back below our risk line while the model's twelve-week view deteriorated to nearly 3% below the current price, widening from 2% a session ago - the second time in three sessions the stop has been broken. The business is sound (second-quarter net income of 85bn on a beat, net premiums written up 3.6%, a fresh $5bn buyback) and our own valuation actually marks the shares as fairly to slightly undervalued, which is why this is a 30% trim held deliberately at the smallest tier rather than a larger cut. We are protecting against a broken trend in an insurer we still want to own most of. OPS: 4TH ISSUANCE (08-17/08-18 blocked). Action price = the 08-19 close, repriced from (-1.27%); band -. = 30% x 16,377 -> L2 re-fires (close vs the stop, -0.30%) but the break is only 0.17 ATR - NOT decisive on its own; carried by the widening negative 12wk plus the FailedBreakout_Down_20D AVOID cv57 one session back. Sec.9(b): UNDERVALUED argues the tier DOWN and 30% is already the floor, so it is held there and explicitly not escalated. Stop held on the retained. · news |
| NYT | New York Times Co Cl A | 65.96 | SELL 88.52% | 65.96 | 62.42 | 65.32 (-1.0%) | n/a | 45.17 (-31.5%) | The New York Times is the cleanest exit case in the book and we have deferred it for six sessions purely because the position is small. The signal is an unambiguous STRONG SELL at conviction 69, the shares trade below their 200-day average, and the model projects them roughly 31% lower over twelve weeks - the worst forward view of any holding. The news confirms rather than contradicts: second-quarter digital subscriber additions of about 280,000 missed expectations and slowed from the prior quarter, sending the stock down 13% on the print and 12.8% over four weeks, with print declines and elevated costs on top. Digital subscription revenue up 16% and digital advertising up 21% are real, but they have not stopped the de-rating. We keep a token stake and step out of the rest. OPS: action price = the anchor Entry_Price (= close); band -. = 88.52% x 1,592 -> retained. Takes slot 3 over DVA despite ranking below it on dollar bleed because DVA's news OVERRIDES (TD Cowen upgrade) and its governing cell is the book's weakest, while this card sits on FailedBreakout_Down_20D|sell - the best actionable cell in the book (n=43, hit_4w 0.562). MoS -1.33 is short of the -3 bump line, so valuation up-ranks only. Stop RAISED -> on the retained shares. · news |
| DVA | DaVita Inc | 177.27 | NO ACTION | 177.27 | 174.64 | n/a | n/a | 108.79 (-38.6%) | DaVita's planned trim is cut in half on a genuine change in the outside view. TD Cowen upgraded the shares from Hold to Buy and raised its target to about 24% above the current price, and management reaffirmed full-year earnings guidance of -after a second-quarter revenue and profit beat - roughly twelve times earnings. The bearish case is still real: the position is down about 25%, the signal has been unreadable for five straight sessions, and the model projects a further steep decline. But the shares are deeply oversold and the fundamental news flow has turned, so we halve the intended cut rather than press it. OPS: DEMOTED 50% -> 30% on the news override; no slot (bleed 4th). The MoS -3.35 that would have bumped 50% -> 100% is REFUSED - it comes from an EPV mark of against a DCF of on the same name, a 9x method spread that is systemic finding 1's artifact, not a valuation signal. Governing cell NoTrade|sell is the book's weakest actionable (n=35, hit_4w 0.286). 6th consecutive session deferred. Stop held (1.50% cushion); a decisive break re-escalates to 50% regardless of the upgrade. · news |
| MCO | Moody's Corp | 497.03 | NO ACTION | 497.03 | 461.27 | n/a | n/a | 509.00 (+2.4%) | Moody's is held back from the sell list for a second session. The signal reversed badly - from a firm HOLD to no readable signal at all - but the model's twelve-week view has now been positive twice in a row and improved to about 2.4% above the current price. A name with no signal and a positive forward view is not a sell, so the intended 30% trim stays on the shelf. OPS: label reversed HOLD cv68 -> AVOID cv0 NoSignal (a -68 collapse) but the 12wk confirmed positive for a 2nd print (+0.51% -> +2.41%). Would rank #3 on pure dollar bleed and is explicitly NOT slotted - NoSignal + positive 12wk is a thin sec.9 origination. Not armed for a profit-take either (Gain_ATR 0.67, far under 3.0). A 3rd positive-12wk print retires the card. Stop held. |
| OXY | Occidental Petroleum Corp | 60.09 | NO ACTION | 60.09 | 57.52 | 62.61 (+4.2%) | 68.49 (+14.0%) | 66.91 (+11.3%) | Occidental is sitting on a 23% gain and has started showing momentum divergence - the price is still rising but the underlying strength is not confirming it. That combination normally argues for banking part of the gain. It does not make tonight's three-name sell list because the amount at stake is small relative to the other candidates, so we record the intent and tighten the protective stop instead. OPS: ADR-0012 profit-take FIRED - armed on Gain_ATR 8.37, tick (a) RSI_DIV on the row. G1 softening does NOT apply (BreakoutUp_20D at cv40, nowhere near the cv>=80 bar), so the tick fires normally. 30% / @ = 7th on bleed - lost the slot on ranking, not merit. Ledgered optional so the counterfactual grader prices the omission. Stop RAISED -> (4.28% cushion). |
| COF | Capital One Financial Corp | 220.73 | HOLD | 215.05 | 218.60 | 218.22 (+1.5%) | 225.63 (+4.9%) | 256.86 (+19.4%) | Capital One flipped back to a buy signal and is the second-ranked buy candidate in the book, with the model projecting roughly 16% upside over twelve weeks. It misses our threshold by three points once valuation and our own track record on this signal type are applied, and the shares have already run more than an ATR past the level the model wants to buy - so chasing it here would break our no-chase rule. We pass, and note that its earlier sell case has now dissolved. OPS: adj 63 = 66 - 2 (val_adj, MoS -0.35) - 1 (PullbackBuy_VWAP|buy cell, n=7) vs a 65 floor. Entry_Distance_ATR -1.25 (extended past the buy-limit); NOT queued because a sub-floor name does not get a GTC. Separately, the 08-18 fired profit-take is CANCELLED - it fired on tick (d), 2 consecutive NO TRADE prints, and the anchor row is now BUY - ADD cv66, so the deterioration resolved. Stop held (the scan's would loosen it materially; rejected) - only 0.97% of cushion, deliberately tight. Re-arms on adj >= 65 at or below. |
| NUE | Nucor Corp | 248.74 | HOLD | 248.74 | 248.53 | n/a | n/a | 288.08 (+15.8%) | Nucor gave back its recent spike and closed 4.3% through our protective stop, the deepest breach in the book tonight. We are not selling: the model still projects about 16% upside over twelve weeks and the steel cycle news remains strongly supportive - hot-rolled coil has surged toward a short ton on mill price increases, longer lead times and tariff-suppressed imports, the shares are up 23% in three months, Morgan Stanley raised its target and the company is investing in its Indiana facility. Our exit rule requires a broken stop AND a negative forward view; only one of the two is present. We move the risk line to where the model actually puts it. OPS: L2 does NOT fire (12wk +15.8%). NoTrade_Reason verbatim: 'Liquidity gate failed'. Stop RE-BASED -> (the last live scan level, 08-19) because now sits 4.5% above market and had stopped functioning - the DAL pathology. PEAK STOP RECORDED; erosion -4.4% with no size cut. Inside ADR-0020's 15% line and ADR-0020 is off for this book, but a SECOND erosion from this watermark with size still uncut triggers the mandatory 30% trim. Cushion after the re-base is 0.08% - razor thin by construction. · news |
| VRSN | VeriSign Inc | 273.03 | NO ACTION | 273.03 | 276.41⚠ | n/a | n/a | 267.08 (-2.2%) | VeriSign closed below our protective stop and, for the first time, the model's twelve-week view turned negative - the two conditions that normally trigger a trim. The break is shallow though, well under half a day's typical range, and the position is still 6.5% ahead. We arm a 30% trim rather than execute it, and it fires on a second consecutive close below the line. OPS: L2 fires for the 1st time on this name (close vs the stop, -1.22% / 0.41 ATR; 12wk -2.18%). Break is NOT decisive, label went HOLD cv51 -> NO TRADE (1st print), Gain_ATR 2.02 so not profit-armed. Bleed ranks 8th - no slot regardless. Stop held. |
| KR | Kroger Co | 56.27 | NO ACTION | 56.27 | 54.87 | 56.27 (+0.0%) | n/a | 44.78 (-20.4%) | Kroger carries the single strongest sell signal in the book - a STRONG SELL at conviction 69, top-ranked on the sell side, with the model projecting about 20% downside over twelve weeks. We are not acting on it because the position is a rounding error: worth roughly Recording it so the signal is not lost. OPS: excluded on materiality for a 6TH session (bleed). Sell_Rank 1. Sec.9(b): UNDERVALUED (MoS +0.07) argues the tier DOWN, already at the printed portion. Systemic finding 3 - the >=3 dispose cap plus pure dollar-bleed ranking keeps burying the strongest signals on small lines. Stop held (the scan's would loosen it). |
| NVR | NVR Inc | 6418.86 | NO ACTION | 6418.86 | 6088.57 | 6418.86 (+0.0%) | n/a | 5927.40 (-7.7%) | NVR flipped from a bearish watch to an outright STRONG SELL, trading below its 200-day average with the model projecting about 8% downside over twelve weeks. The holding is three shares, so there is nothing meaningful to act on. Recorded for the signal history. OPS: Sell_Rank 3. 88.52% x 3 =, Stop RAISED ->. |
| GOOGL | Alphabet Inc Cl A | 344.72 | NO ACTION | 344.72 | 319.70 | n/a | n/a | 308.32 (-10.6%) | Alphabet has been unreadable to the model for five straight sessions and its twelve-week projection sits about 11% below the current price. We already cut 30% of the position last week and it has gone sideways since, so the remaining trim stays queued behind three larger, clearer cases rather than being forced through tonight. OPS: would rank #3 on dollar bleed and is passed over on case quality - NoTrade|sell is the book's weakest actionable cell. Cut 30% @ on 08-13. The expired 07-05 overlay's 'invalidated below 345' line is tripped again at with no delta attached. Stop held. |
| GOOG | Alphabet Inc Cl C | 341.70 | NO ACTION | 341.70 | 320.20 | n/a | n/a | 303.87 (-11.1%) | The C-share line mirrors the Alphabet A-share decision and executes with it. Same read: five sessions without a readable signal and a twelve-week projection about 11% below the current price, deferred behind larger cases. OPS: bleed Executes with GOOGL. Stop held. |
| KHC | Kraft Heinz Co | 25.68 | NO ACTION | 25.68 | 24.15 | n/a | n/a | 25.87 (+0.7%) | Kraft Heinz printed a bearish breakdown label again, but the model's twelve-week view is now positive for a second consecutive session and the shares keep grinding higher. That is a contradiction we resolve in favour of holding: the intended trim stays on the shelf and a third positive reading retires it entirely. OPS: weakest of the carried set. Bleed (would be #4-5). 12wk +0.74%, positive 2nd print. NOTE the scan's stop prints ABOVE the close - a short-side stop on an AVOID row - and is rejected; state stop held. |
| SIRI | Sirius XM Holdings Inc | 28.67 | NO ACTION | 28.67 | 28.60 | n/a | n/a | 26.89 (-6.2%) | Sirius XM is clinging to our stop by a quarter of a percent after four sessions without a readable signal, with the model projecting about 6% downside. It has not broken, so we hold and let the line do the work. OPS: close vs the stop, +0.24%. A decisive close below re-escalates the carried trim 30% -> 50%. Bleed Stop held. |
| ALLY | Ally Financial Inc | 43.11 | NO ACTION | 43.11 | 43.26⚠ | n/a | n/a | 39.93 (-7.4%) | Ally slipped a fraction below our stop - well inside daily noise - with a twelve-week projection about 7% lower. Our own valuation marks it as undervalued, which argues for the smallest possible cut, and we are already at the smallest tier. Not actioned tonight. OPS: breach -0.35% / 0.18 ATR, sub-noise. Sec.9(b): UNDERVALUED argues the tier DOWN; already at the 30% floor. Bleed Stop held - likely the next genuine breach. |
| DAL | Delta Air Lines Inc | 83.29 | HOLD | 83.29 | 81.36 | n/a | n/a | 85.49 (+2.6%) | Delta's trim stays withdrawn. Berkshire raised its stake to 8.7% from 6.1% and Delta is the only airline it owns, which in a book built to mirror that portfolio is close to decisive on a marginal cut. The model's twelve-week view is positive again at about 2.6% above the current price, and the shares have recovered back above the re-set risk line. No action. OPS: 12wk +2.64%, close now 2.37% ABOVE the re-based stop. NO second stop erosion this session, so ADR-0020's conditional trigger stays unarmed. PEAK STOP carried (-9.7% erosion, size uncut). Stop held. · news |
| KO | Coca-Cola Co | 90.35 | HOLD | 90.95 | 87.88 | 93.22 (+2.5%) | 100.51 (+10.5%) | 100.37 (+10.4%) | Coca-Cola broke out to a 52-week high with conviction improving and the model projecting about 11% more over twelve weeks. The position is 7.4% ahead and profit-armed, but nothing has deteriorated, so the correct move is to raise the trailing stop and stay long rather than take money off the table. OPS: armed on Gain_ATR 4.04, ZERO fire ticks - no second warning code, conviction ROSE, 12wk alive at +11.1%, not NoTrade. ADR-0012 trail-raise, not a trim: stop RAISED -> (the scan's live level, 2.73% cushion). Largest single line in the book at. |
| BAC | Bank of America Corp | 63.17 | HOLD | 63.11 | 62.57 | 63.87 (+1.2%) | 65.64 (+4.0%) | 73.47 (+16.4%) | Bank of America holds for a fourth straight session with the best forward view in the book - the model projects roughly 16% upside over twelve weeks - and the position is 5.1% ahead. Nothing to do but keep the stop tight underneath it. OPS: DISARMED - Gain_ATR fell 4.12 -> 2.79, back under the 3.0 line, so the profit-take path is closed. Stop held (the scan's would loosen it; rejected) - 0.95% cushion, tight by intent. Sell case withdrawn 08-17 remains withdrawn, confirmed a 4th time. |
Outcome & track record
Accuracy at the time of this record.
120 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 →
-0.010 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.