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 13 Aug 2026, 16:00:00 GMT-4

Berkshire Mirror — End-of-Day Verdict

EOD 2026-08-14

Recommendations

2 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 Co342.48SELL 50%342.48334.75342.48 (+0.0%)n/a342.48 (+0.0%)Trimming half of the book's largest position because the model now projects literally zero return on it over the next twelve weeks: the 12-week view sits exactly at the current price after four straight sessions with no tradeable signal. American Express itself is doing fine - Q2 revenue rose 10% to 6bn, EPS beat at, and full-year guidance was raised - but the shares are down 6% this year while Visa and Mastercard have held up, and only about half of Wall Street rates it a buy. We keep the other half rather than exiting: this is a valuation-neutral position with no modelled upside, not a broken business. OPS: ADR-0012 profit-take, armed + ticks (c) dead 12wk and (d) 4x NO TRADE -> profit_take_2 = 50%. The 08-12/08-13 escalation trigger ('12wk turns negative OR breaks') did NOT literally fire - 12wk is exactly flat at 0.00% and the stop holds 2.31% cushion; executing on the totality per L3 (4th consecutive night as #1 dollar bleed, 11.8% of NAV, counterfactual optional_trim median +2.81% cost at 4wk over n=64). Tier capped at 50% because NoTrade|sell grades 0.000 at 4wk over n=29. Valuation amplify-only, MoS -0.476 not <= -3, no tier bump. Stop HELD, never loosened. Estimates f=0.30 (calib_1w 0.006 degenerate, L5). · news
SIRISirius XM Holdings Inc28.46SELL 50%28.4628.6027.97 (-1.7%)n/a26.83 (-5.7%)Doubling the pending trim from 30% to 50% because the shares have now closed below our risk line four sessions in a row, and we said out loud last night that a fourth would escalate. The business is genuinely improving - the best second quarter in four years, record-low subscriber churn near 1.4%, and raised full-year guidance - but the market keeps refusing to pay for it, and a risk line that keeps getting broken while the position never shrinks is how losses compound. Half the holding stays, because a better business at a rejected price is a trim case, not an exit. OPS: ESCALATION of the 08-13 card (7,501 ->); repriced -> (+1.68% > 1.5% gap threshold). Closes below : 08-11 28.51, 08-12 28.44, 08-13 27.99, 08-14 28.46. ADR-0012 independently agrees: armed + (b) conviction + (c) dead 12wk -> profit_take_2 = 50%. Signal route deliberately NOT used - RallySell_BearTrend carries the book's worst bias at -8 (n=10, hit_1w 0.125). Stop HELD. Re-arm: 5th consecutive close below, or a close below, takes the remaining. Estimates f=0.30 (L5). · news
DVADaVita Inc180.06SELL 50%180.06174.64162.02 (-10.0%)n/a119.91 (-33.4%)Cutting the book's worst position in half. DaVita fell 17% on its Q2 report as investors looked past the earnings beat to weaker revenue per treatment and flat patient volumes, with an ACA subsidy headwind of roughly this year and next. Four sessions have now passed with no tradeable signal and a deteriorating twelve-week view, and the position sits 23% below cost. We keep half rather than exiting, because analysts still see a mean target about 27% above today's price and management reaffirmed its 2026 outlook - this is a de-risking, not a verdict that the business is broken. OPS: ADR-0012 armed + (c) dead 12wk + (d) 4x NO TRADE -> profit_take_2 = 50%. STATED DEVIATION from strict dollar-bleed ranking: GOOGL and KHC outrank on dollars and were declined - GOOGL was cut 30% at on 08-13 and closed unchanged at in the NoTrade|sell cell that grades 0.000 at 4wk (n=29); KHC's tick (c) clears by 0.008% while the name rises. Counter-argument acknowledged: RSI 26.1 is a washout print and the name rose into the anchor - deferred 3x on that basis, executing now because RSI has sat 22-26 for four sessions while the 12wk fell ->. Stop HELD. 1wk -10.0% is an f-interpolation artefact off a 12-week target, not a forecast; the -33% 12wk itself is trend-extrapolation and is flagged as a systemic finding. f=0.30 (L5). · news
DALDelta Air Lines Inc89.35HOLD88.3990.14n/a93.79 (+6.1%)101.08 (+14.4%)Holding Delta without adding. It is the only name in the book flashing a fresh buy signal and the twelve-week view is a healthy +13%, but the shares are priced above our estimate of what they are worth, and this particular signal family has yet to reach a single four-week target in this book. The price also slipped below our trailing risk line for the first time. Record second-quarter revenue of 8bn and several raised price targets sit against roughly 3% of July flights cancelled amid a pilot dispute and the CEO selling of stock on 4 August - not a picture that justifies buying more today. OPS: BUY - ADD cv66, Buy_Rank 1 -> adj 60, five points under the 65 floor (val_adj -5 from MoS -1.017; PullbackBuy_VWAP|buy bias -1, n=7, 0.429 1wk / 0.000 4wk). NOT ACQUIRED. Stop-breach handling: close vs is 0.34 ATR = a noise break of a sub-ATR trail, and L2 does not fire (needs stop-break PLUS negative 12wk; 12wk is +13.1%). Stop HELD at, deliberately NOT loosened to the model's new structural. RE-ARM/ESCALATE: a 2nd consecutive close below trims 30% regardless of the 12wk; a reclaim of with the buy label intact re-tests the floor. · news
AAPLApple Inc305.93SELL 50%302.25300.00n/an/an/aCarried, not re-issued: the 50% trim decided on 08-12 is still unfilled because the executor has not run for two sessions. Apple has drifted up 1.2% above our decision price since, so the delay has helped, but the case is unchanged - four sessions with no tradeable signal and a twelve-week view that sits below today's price. OPS: gap +1.22% < 1.5% -> price HELD at. Armed + (c)+(d). Stop.
BACBank of America Corp64.49SELL 30%64.8162.57n/an/an/aCarried but flagged for withdrawal. The 30% trim decided on 08-12 is still unfilled, and in the meantime the signal has reversed: Bank of America now prints a hold with the strongest twelve-week view in the entire book at nearly +20%. Selling into that would be acting on a thesis that no longer exists. One more confirming session and this card is withdrawn. OPS: gap -0.49%, price held. cv0 -> cv61 FailedBreakdown_Up_20D, 12wk (+19.4%). Trail raised ->. Systemic finding 5: ADR-0012 queue-expiry auto-withdraws GTC orders below cv65 but does not reconcile pending market cards whose thesis reversed.
MCOMoody's Corp484.96SELL 50%488.42461.27n/an/an/aCarried, not re-issued: the 50% trim from 08-12 remains unfilled. Moody's has gone nowhere since, still shows no tradeable signal after four sessions, and the twelve-week view offers about 1% - not enough to hold a full position for. OPS: gap -0.71% < 1.5% -> price HELD at the 08-13 repriced. Armed + (c)+(d). Stop.
CBChubb Ltd343.64SELL 30%344.50341.96n/an/an/aCarried, not re-issued: the 30% trim from 08-13 is unfilled. The case has strengthened rather than faded - the twelve-week view has now tipped genuinely negative and the signal degraded again, so this remains the right trim at the same size. OPS: gap -0.25%, price HELD. New tick (b): conviction. 12wk = -0.79%. Stop (model prints a short-side level above the close; ignored).
NYTNew York Times Co Cl A64.79SELL 88.5%64.1360.57n/an/an/aCarried, not re-issued: the near-full exit decided on 08-13 is still unfilled. Nothing has improved - the sell signal is now in its fourth consecutive session and ranks first in the book, the shares remain well below their long-term trend, and the twelve-week view still points about a third lower after the subscriber-growth miss. OPS: gap +1.03% < 1.5% -> price HELD. STRONG SELL - REDUCE cv68, adj 67, Sell_Rank 1. Trail TIGHTENED -> (model's own level).
LLYVALiberty Live Group Ser A104.99ADD — QUEUED @104.77104.77100.05109.01 (+4.0%)118.92 (+13.5%)112.90 (+7.8%)Still the strongest signal in the book by a wide margin, and the queued add stands. Liberty Live keeps grinding to new highs with conviction at the top of the book, so we are repricing the limit up to where the model now wants to buy rather than leaving a stale order below the market that never fills. OPS: 2nd reprice ( ->, the printed Entry_Price; ran +1.87% > 1.5%). expires_on HELD at 2026-08-18 - a reprice does not buy a queue more time. Valuation layer blind on this line (NA_FV, negative book value) -> val_adj 0, quality gate off. Trail raised ->. Counterfactual no_chase_queue costs a median +3.00% at 1wk over n=16, which is exactly why the limit moves to the market rather than sitting under it. 1wk = f=0.30 off Target_Price.
LLYVKLiberty Live Group Ser C108.86ADD — QUEUED @108.68108.68103.90113.98 (+4.9%)123.01 (+13.2%)117.69 (+8.3%)Same underlying and same case as the Class A line - second-highest conviction in the book, still making new highs. The limit is repriced up to the model's live entry so the order can actually fill instead of trailing the market. OPS: 2nd reprice ( ->, printed Entry_Price; ran +1.80%). expires_on HELD 2026-08-18. NA_FV -> val_adj 0, quality gate off. Trail raised ->. Execute alongside LLYVA. 1wk = f=0.30 off Target_Price.
GOOGLAlphabet Inc Cl A345.90NO ACTION345.90319.70n/an/a313.58 (-9.3%)Declined for a second time. We already trimmed 30% of Alphabet on 13 August at and the shares have gone nowhere since - cutting the same line again days later, on the same stale reasoning, is churn rather than risk management. The trim tier did escalate on the model's own terms, so it is recorded for scoring, but not acted on. OPS: armed + (c) 12wk -9.3% + (d) 2x NO TRADE -> tier escalated 30% -> 50% (, #2 by dollar bleed). Declined: NoTrade|sell grades 0.000 at 4wk over n=29 and the 08-13 cut at is flat at. Overlay override 'invalidated below 345' stays re-armed, no delta. Stop held.
KHCKraft Heinz Co25.51NO ACTION25.5124.15n/an/a26.02 (+2.0%)Held one more session. The trigger for trimming Kraft Heinz clears by two-thousandths of a percent - effectively a rounding error - while the shares are actually rising and sit above their long-term trend. Acting on that margin would be false precision; one more session will settle it either way. OPS: armed + (c) but 12wk vs the 1.02x line of - clears by 0.008%. Name +3.5% over 4 sessions, above EMA200, RSI 43.8 -> 51.1. Label did degrade to AVOID cv62 on FailedBreakout_Down_20D (adj 61, under the 65 floor). Same 'three pennies' situation CB sat in before it resolved. Stop held.
COFCapital One Financial Corp227.34NO ACTION227.34218.60n/an/a271.40 (+19.4%)Not selling into strength. Capital One did flip to a sell label after touching overbought territory, but it is a single session's flip after three sessions of holding, the conviction behind it is weak, the position is up nearly 11%, and the twelve-week view is the second-best in the book at roughly +19%. That is the definition of cutting a winner. We raise the protective stop instead. OPS: STRONG SELL - REDUCE cv39, Sell_Rank 3, ReversalDown_Overbought at RSI 70.2 -> adj 39, 26 points under the 65 floor, single late flip. G3 blocks regardless (12wk +19.4% on a +10.8% winner). NOT armed-and-fired - no ADR-0012 tick. Trail RAISED ->.
ALLYAlly Financial Inc44.91NO ACTION44.9143.26n/an/a43.24 (-3.7%)Stays out of the sell queue. Last week's sell case on Ally has fully evaporated - the signal has improved two sessions running, the shares are rising and back above their long-term trend, and our own valuation work says the stock is cheap. The stop moves up instead. OPS: armed + (c) 12wk (-3.7%) fires the ADR-0012 route, but the label improved HOLD cv40 -> cv48 on FailedBreakdown_Up_20D, RSI 54.0 rising, above EMA200. Withdrawn 08-13; stays withdrawn. Trail RAISED ->.
GOOGAlphabet Inc Cl C343.54NO ACTION343.54320.20n/an/a311.42 (-9.3%)Same company, same decision as the Class A line - declined for the same reasons and to be executed alongside it whenever Alphabet is trimmed. OPS: armed + (c)+(d) -> 50% tier, Mirrors GOOGL. Stop held.
CVXChevron Corp200.00NO ACTION200.00189.84n/an/a219.37 (+9.7%)Riding the winner. Chevron closed at a new high with conviction still climbing and a twelve-week view about 10% above today's price, on a position already up 18%. A momentum warning did technically qualify this for a profit trim, but trimming a name making new highs on strengthening conviction is exactly the mistake this book has made before. We raise the stop and let it run. OPS: armed (Gain_ATR 7.09) + tick (a) RSI_DIV -> profit_take_1 30% DID fire. CORRECTION to 08-13 reasoning: ADR-0012 states G2/G3 govern the valuation-elevate path only and do NOT block a fired profit-take, so '08-13: G3 blocks' was the wrong basis. Declined tonight on dollar-bleed rank (8th) plus G1's spirit - Tier-1 code on a strengthening uptrend argues for the trail. cv65 -> cv68, new closing high. Trail RAISED ->.
NUENucor Corp268.91NO ACTION268.91259.99n/an/a335.79 (+24.9%)Leaving the book's best forward view alone. Nucor is up 22% and carries the strongest twelve-week outlook of any holding at roughly +25%. The trim that technically qualified would move - immaterial - and would cut precisely the position the book most wants to own. OPS: armed (Gain_ATR 7.60) + tick (b) cv67 -> % = Declined on materiality plus G3's spirit. Stop held (NoTrade row prints none).
NVRNVR Inc6307.93NO ACTION6307.936069.66n/an/a5803.85 (-8.0%)Qualifies for a trim but the position is three shares. Recorded so the scorecard prices the omission; the trade itself is one share and not worth the ticket. OPS: armed + (c) 12wk -8.0%; AVOID cv50 x4 on FailedBreakout_Down_20D, close below EMA200 Stop HELD (model's is a short-side level above the close).
KRKroger Co56.69NO ACTION56.6954.87n/an/a45.10 (-20.4%)The cleanest sell signal in the book tonight, and it still does not get executed - the entire position is worth about six thousandths of one percent of the portfolio. Kroger is genuinely weak on the charts and the twelve-week view points 20% lower, but our valuation work says the shares are cheap on a forward P/E of about 11 against an industry average near 15, so any trim would be the minimum size anyway. Recorded for scoring. OPS: STRONG SELL - REDUCE cv67, Sell_Rank 2, adj 66 - the ONLY floor-clearing signal-originated sell available (NYT already in flight). Excluded on dollar-bleed rank alone (0.006% of NAV) - structural, not a re-litigated judgement. Tier already down-ticked 88.52% -> 50% because valuation dissents (UNDERVALUED, MoS +0.062, FV > close). Ex-div on 08-14 = 0.69% of the session's drop. Stop held (model's is looser).
MMacy's Inc23.41NO ACTION23.4123.35n/an/a24.69 (+5.5%)Technically qualifies for a small trim on a collapse in signal conviction, but the trade is worth about and the twelve-week view is still positive. Recorded, not acted on. OPS: armed + (b) cv56 -> 0. 12wk is +5.5%, so tick (c) does NOT fire. Stop held (model's is looser).
STZConstellation Brands Cl A139.18NO ACTION139.18128.06n/an/a125.04 (-10.2%)Qualifies on a negative forward view and a weak chart, but the position is. Recorded for scoring only. OPS: armed + (c) 12wk -10.2%; AVOID/NoSignal, close below EMA200. Quality_Pass False (would block a new entry, never a held-name ADD or a sell). Stop held.
LENLennar Corp Cl A86.83NO ACTION86.8382.64n/an/a79.62 (-8.3%)Qualifies on a negative twelve-week view, but the position is and the current signal is a hold with rising conviction. Recorded, not traded. OPS: armed + (c) 12wk -8.3%; HOLD cv60 FailedBreakdown_Up_20D. Stop HELD (model's is looser).
LEN-BLennar Corp Cl B85.36NO ACTION85.3681.78n/an/a79.20 (-7.2%)Same case as the Class A line on a nine-share position. Recorded, not traded. OPS: armed + (c) 12wk -7.2%; HOLD cv67. Stop HELD (model's is looser).
KOCoca-Cola Co87.71HOLD85.21n/an/a95.52Holding. Coca-Cola is up 4% for us with a twelve-week view about 9% higher and a steady hold signal - nothing to do but raise the protective stop. OPS: HOLD cv45 FailedBreakdown_Up_20D; armed but NO tick fires (12wk > 1.02x close). Trail RAISED ->.
OXYOccidental Petroleum Corp58.36HOLD55.23n/an/a62.49Holding. Occidental is our second-best performer at +19% with a positive twelve-week view and a recovering hold signal. OPS: HOLD cv62 FailedBreakdown_Up_20D; armed (Gain_ATR 6.03) but NO tick (12wk = +7.1%). Note MoS -3.065 is past the -3 line, which would bump a trim tier - but valuation never originates a sell (CLAUDE.md section 9) and no signal-based reason exists. Trail RAISED ->.
VRSNVeriSign Inc284.24HOLD276.41n/an/a299.48Holding. VeriSign is up 11% and its twelve-week view is still positive; the pullback in signal strength is not enough to act on. OPS: the one clean armed-but-not-fired hold - Gain_ATR 3.54, but conviction fell only 13 pts (under the 20-pt threshold) and 12wk = +5.4%, alive. No tick, no trim. Stop HELD (model's is looser).
JEFJefferies Financial Group55.14NO ACTION55.1454.37n/a58.22 (+5.6%)55.30 (+0.3%)No action. Jefferies is on the watchlist but the signal is only a watch, conviction is weak, our valuation work says it is expensive, and the twelve-week view offers less than 1%. Nothing here worth a first position. OPS: WATCH cv42 -> adj 40, 25 points under the 65 floor., never seeded. Quality_Pass False would block a new entry anyway.
LPXLouisiana-Pacific Corp73.89NO ACTIONn/an/a69.52No action. No tradeable signal, a negative twelve-week view, and the most expensive valuation in the book's universe. OPS: NO TRADE cv0, 'Consensus conflict + low conviction'. MoS -5.253 -> val_adj -10 (the floor)., unheld. Quality_Pass False.

Outcome & track record

Accuracy at the time of this record.

109 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.

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.