This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.

US End-of-Day Verdict 7 Sept 2026, 16:00:00 GMT-4

Berkshire Mirror — End-of-Day Verdict

EOD 2026-09-08

Recommendations

1 to acquire · 3 to dispose

TickerNameCloseActionAction 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
AXPAmerican Express Co326.10SELL 88.5%326.10317.92322.52 (-1.1%)314.18 (-3.7%)290.35 (-11.0%)American Express is the book's largest single loser and the model has now stopped hedging about it: after a session as AVOID the label snapped back to STRONG SELL - REDUCE at conviction 69, the highest-ranked sell in the portfolio, with the price below every moving average it has (20/50/100/200-day) and below its volume-weighted average price. The 12-week view of sits 11% under tonight's close, and the 52-week low is only another 11% below that, so the downside is not yet exhausted. Fair value agrees from a different direction: the shares are priced above their residual-income anchor and the quality screen fails. Cutting 89% is the trade. OPS: adj cv69 CLEARS the 65 floor on its own this session (09-04 it was admitted at cv63 on the trend-exit basis) - Sell_Rank #1, dollar bleed, ~10.1% of NAV. 5th consecutive ask; the 09-04 card at went unexecuted, costing - the smallest slippage of the blocker's run, but only because AXP went sideways. Scorecard cell FailedBreakout_Down_20D|sell is +0 (n=52, hit.52) - a coin-flip cell, so this card rests on structure and size, not on cell edge. Size modifier 1.0x (MoS -0.43 is not <= -1). Remainder keeps the stop, not breached. 4wk estimate is time-scaled from the 12wk view - the engine prints no Target_Price on this row. WORK THE ORDER: ~10% of NAV. · news
CBChubb Ltd336.64SELL 30%336.64327.97334.48 (-0.6%)329.45 (-2.1%)315.07 (-6.4%)Chubb has stopped grazing its risk line and gone through it: the stop that held all through August was breached by 0.11% on Friday and by 1.56% tonight - a real break, not a wick - while the model's 12-week view of sits 6.4% below the close. A decisive stop break plus a negative forward view is a sell basis on its own, and it does not need the engine to print a sell label, which it has not done in three sessions of NO TRADE on consensus conflict. The insurer's own numbers say the same thing more quietly: Q2 revenue beat but book value per share missed, and the shares have drifted down since. This is a size cut, not an exit - Chubb still trades above its 200-day and screens at fair value, so the position stays. OPS: L2 admission (stop break + negative 12wk) on a NO TRADE row; US-1 says act the FIRST session the trigger fires rather than bleed for four. Portion set at 30%, not the engine's 89%: valuation reads UNDERVALUED (MoS +0.005) which argues a trim DOWN one tier per the amplify-only rule, and the break is one session old on a name still above EMA200. of a position. Stop for the 11,464-sh remainder RE-BASED -> (close - 1.5 ATR) - re-basing is earned by the cut, not granted instead of it. Ex-dividend on 09-11; immaterial at 0.3%. Escalates to a full-size REDUCE if the 12wk stays negative and the name loses EMA200. · news
DALDelta Air Lines Inc78.96SELL ALL78.9681.3677.40 (-2.0%)73.76 (-6.6%)63.37 (-19.7%)Delta has now spent five straight sessions below its risk line and the break is deepening again - 1.46% on Friday, 2.95% tonight - while the model's 12-week view of sits nearly 20% below the current price, still the worst forward view in the book by a wide margin. The airline's fundamentals remain genuinely good: September-quarter EPS guided to -2.50, full-year -7.50 affirmed, mid-teens revenue growth and a 15% dividend raise. That is exactly why this is a full exit rather than a trim - a stock the market keeps marking lower through good news is telling a different story than its income statement, and the position is already 14.9% underwater. OPS: L2 basis (decisive stop break + negative 12wk), 5th consecutive ask - L3 says convert it rather than issue a 6th. Non-execution of the 09-04 card at cost this session. News is in-window this time but is reiterated June-quarter guidance, not a new catalyst, so it does not override a five-session technical breakdown; valuation is barred from vetoing a signal-based exit in any case. Below EMA20/50/100 and closing on EMA200. Drop the stop from the monitor after fill. · news
CVXChevron Corp209.80ADD — QUEUED @204.75204.75196.87207.49 (+1.3%)213.89 (+4.5%)244.58 (+19.5%)Chevron is the single best setup in this book and the only name in it with a genuinely bullish trend reading - a PullbackBuy at conviction 79, Buy_Rank #1, trading above every moving average with a 12-week view of about 17% above tonight's close, the strongest forward view in the portfolio. The business is delivering to match: Q2 EPS of beat the consensus on record US upstream output and 4bn of adjusted free cash flow, and the $7bn Venezuela expansion is targeted at doubling production there to 600,000 barrels a day. The catch is price - the stock is up 43% this year and sits 2.5% above the model's own entry, so this is queued at rather than bought at the market. OPS: L1 cell (PullbackBuy_VWAP, the preferred family; cell bias +0, n=8, hit.50 - actionable but no edge, so the card rests on the setup). Entry_Distance_ATR -1.28 = extended, no-chase rule applies -> GTC limit, expires 2026-09-11. val_adj -4 (MoS -0.85); quality gate F-Score/FCF FAILS but CVX is HELD so the gate does not block an ADD. Size 30% x 1,104 = = - immaterial at 0.07% of NAV, and that is a structural artifact worth naming: the ADD rule sizes off a position this book has already cut to a stub, so the best signal in a 50%-cash portfolio can only 0.07% of it. CONDITIONAL: the base trail sits at, only under the limit - if trades, the stop fires first and this order AUTO-WITHDRAWS; do not add to a name that just stopped out. · news
KOCoca-Cola Co88.36WATCH — queued order lapsedn/an/an/aQueued buy @89.22 lapsed — closed 88.36 below the level; awaiting reprice at the next end-of-day run. Coca-Cola prints the book's second buy signal of the night - a PullbackBuy at conviction 75, Buy_Rank #2, sitting on its 20-day average with the model's entry just above the close, so the order needs a touch of strength to fill rather than a chase. The business is behind it: Q2 revenue up 6.7%, a fifth consecutive EPS beat, and Trademark Coca-Cola volume up 5% globally, the strongest in 17 years outside the COVID recovery. What holds the size down is price - the shares are up 26% this year near an all-time high, the fair-value model reads GROWTH_PRICED_IN at a 2.1x margin-of-safety deficit, and at least one sell-side desk has downgraded on valuation alone. Buy it, but buy three-quarters of the normal clip. OPS: adj cv65 = EXACTLY at the floor - conviction 75 less the maximum -10 valuation haircut, cell bias +0. This is the thinnest card of the night and the first to drop if the open gaps. Size 30% x 96,154 x 0.75 modifier (MoS -2.13 falls in the -3 to -1 band) = =; max_position_pct (60%) and the are both non-binding. Limit (entry range -); close was so expect a fill at or below the limit. 12wk EXCEEDS the DCF anchor - projection assumes growth beyond what is priced in. Ex-dividend on 09-15. Trail on the existing 96,154-sh base RAISED -> (PROFIT_ARMED, Gain_ATR 3.50, no deterioration tick). · news
AAPLApple Inc316.22HOLD314.62n/an/a325.23Apple holds through its 09-Sep product event - the iPhone 18 Pro line plus a first foldable - which lands inside the catalyst window and argues for sitting still rather than pre-positioning. The signal itself is inert: three sessions of NO TRADE on consensus conflict, position 2.5% ahead of cost. The risk line is uncomfortably tight, against a close, so a 0.5% slip trips it. OPS: catalyst-aware hold; PROFIT_ARMED prints but Gain_ATR is 1.0, so the flag is the degenerate T12_Progress leg (see systemic finding 2), not real arming. Valuation GROWTH_PRICED_IN, MoS -2.97 - amplify-only, cannot originate. Re-assess on the post-event close.
BACBank of America Corp62.39HOLD61.28n/an/a65.37Bank of America holds at conviction 66 with the 12-week view, about 4.8% above the close, and the position 3.8% ahead of cost. Nothing to do. OPS: stop held at (engine's is lower). CPI 09-11 and the FOMC 09-16 are the live inputs for the book's financials sleeve.
OXYOccidental Petroleum Corp60.65HOLD58.18n/an/a67.44Occidental is 24% ahead of cost and gaining - conviction ticked and the 12-week view sits 11% higher - so the profit-arming here resolves as a trail instruction, not a trim. OPS: PROFIT_ARMED on Gain_ATR 8.53, ZERO deterioration ticks (no second warning code, conviction rising, 12wk alive, no repeated NO TRADE) -> mandatory trail-raise, not a cut. Stop held (engine's is lower; never lower a trail).
NUENucor Corp256.40HOLD248.53n/an/a269.56Nucor is 16% ahead of cost with the 12-week view still 5% higher, but the label dropped from three sessions of HOLD to NO TRADE tonight. That is one print, not two. OPS: PROFIT_ARMED on Gain_ATR 4.34; deterioration tick (d) needs TWO consecutive NO TRADE prints and this is the first - a second one fires a 30% profit-take. Trail held.
COFCapital One Financial Corp213.96HOLD210.24n/an/a224.08Capital One fell 2.6% on the session and dropped to NO TRADE, but the 12-week view is still positive at (+4.7%) and the stop has 1.7% of headroom. OPS: no longer PROFIT_ARMED - Gain_ATR has slipped 3.09 to 1.81, below the 3.0 arm. Financials sleeve is exposed to CPI 09-11 / FOMC 09-16.
MCOMoody's Corp481.25HOLD493.86n/an/a489.17Moody's broke its risk line properly tonight - 2.55% below the stop after grazing it by 0.06% on Friday - but the 12-week view is still POSITIVE at, 1.7% above the close, so the stop-break-plus-negative-12wk exit does not fire. OPS: stop deliberately NOT re-based. A broken line stays where it is until the position is actually cut - moving it away from an uncut position is the Trail Erosion failure ADR-0020 exists to catch. ESCALATES to DISPOSE slot #1 the moment the 12wk turns negative or the engine prints a sell label. position.
LLYVKLiberty Live Group Ser C97.68HOLD98.69n/an/a88.62The re-arm trigger set on 09-04 - a close back below - fired tonight at, with the 12-week view, 9.3% lower. The case is live and qualifies; it lost on slot capacity to three larger, older cards, not on merit. OPS: DISPOSE slot #4 tonight, PROMOTED to slot #1 on the next EOD if the break persists. position, 89% = of bleed deferred one session. Technical-only card - fair value returns NA.
SIRISirius XM Holdings Inc28.80HOLD27.96n/an/a27.14Sirius XM prints STRONG SELL - REDUCE at conviction 62 and Sell_Rank #4, but two independent layers say do not take it. The scorecard's RallySell_BearTrend sell cell is the worst in the book at 1-of-13 on 1-week targets, an -8 haircut that drops this to adj 54, below the 65 floor. And the news is squarely against the short case: Deutsche Bank upgraded to Buy on 02-Sep and lifted its target to from self-pay net additions turned positive for the first time in four years, churn hit a record low 1.4%, and the company becomes YouTube's exclusive digital-audio advertising partner this autumn. OPS: news OVERRIDES the sell label; the queue stays withdrawn and this is now a constructive hold. Stop.
DVADaVita Inc180.06HOLD169.18n/an/a123.20DaVita carries the worst forward view in the book - a 12-week estimate of against a close, 31.6% lower - and the position is 23.4% underwater. It is still not a sell card tonight: AVOID is not a sell label, the stop has 6.4% of headroom, and half the position was already cut on 05-Aug. OPS: near-miss, no origination class fires. ADR-0020 Trail Erosion does NOT fire - the raw 27.5% erosion from the July peak re-bases to 3.1% off the post-cut watermark. Valuation FAIR_VALUE, MoS -3.41 - would amplify a sell if one existed, cannot originate one.
MMacy's Inc22.45HOLD22.49n/an/a19.54Macy's is 0.18% under its stop - a graze, not a break - with earnings due around 10-Sep. The catalyst-aware rule says hold through the print unless there is a confirmed lower high or a decisive break, and 0.18% is neither. OPS: position, immaterial. Re-assess on the post-earnings close.
STZConstellation Brands Cl A120.98HOLD128.36n/an/a95.98Constellation is 5.75% below its stop with a 12-week view 20.7% lower and RSI at 28 - the clearest L2 trigger in the book on the merits. It is not carded because the position is, 0.003% of NAV; the commission would exceed the risk avoided. OPS: below materiality, not below the signal bar. Engine also reports the liquidity gate failed on this row.
LEN-BLennar Corp Cl B78.98HOLD81.78n/an/a67.43Four consecutive STRONG SELL - REDUCE prints and Sell_Rank #2, but nine shares - of exposure. OPS: fails on size, and separately would fail the floor anyway (adj 57 after the -8 RallySell_BearTrend haircut).
NVRNVR Inc6116.86HOLD6105.18n/an/a5542.78STRONG SELL - REDUCE at conviction 65 and Sell_Rank #3, three shares, OPS: adj 57 after the -8 RallySell_BearTrend haircut - below the floor on signal AND below materiality on size.
LENLennar Corp Cl A80.37HOLD82.64n/an/a67.96Below its stop with a negative 12-week view, but and OPS: below materiality.
GOOGLAlphabet Inc Cl A338.36HOLD333.68n/an/a295.03Alphabet is inert - four sessions of NO TRADE, 6% under cost, 12-week view 12.8% lower - and the one standing overlay override on this name (invalidated below) is long since void at. OPS: 16-Sep deadline for redaction motions on the sealed ad-tech remedies memorandum is the live catalyst. Stop, 1.4% of headroom - tight.
KHCKraft Heinz Co24.90HOLD24.87n/an/a24.39Kraft Heinz sits 0.12% above its stop with the 12-week view 2% lower. Nothing fires. OPS: watch the line.
ALLYAlly Financial Inc42.28HOLD41.63n/an/a37.65Ally fell 3.3% to but remains 1.6% above its stop; the label is AVOID, which is not a sell basis. OPS: the buy queue withdrawn on 09-04 stays withdrawn. Valuation reads UNDERVALUED but the quality gate fails, which halves any positive adjustment to nil.
KRKroger Co57.20HOLD54.87n/an/a51.42OPS: below materiality. Trail erosion 11.5% off the 06-Jul peak with no size cut since - the highest genuine reading in the book, still under the 15% ADR-0020 threshold. Watch it.
NYTNew York Times Co Cl A67.96HOLD64.15n/an/a56.00The re-arm set on 09-04 - a second consecutive sell print or a close below - did NOT trigger. The Times rose 1% to and prints AVOID, not a sell. OPS: the 12-week view is still 17.6% lower and the name sits below EMA50/100/200, so the case has not gone away; the trigger simply has not fired. Re-arm stands unchanged.
GOOGAlphabet Inc Cl C335.38HOLD320.20n/an/a289.48The C shares track the A shares - four sessions of NO TRADE, 12-week view 13.7% lower. OPS: position, stop with 4.7% of headroom.

Outcome & track record

Accuracy at the time of this record.

159 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.

This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.