AI · Berkshire Mirror — Trades

Chevron is our best position, up over a quarter on cost and stretched ten ATR above it, and it has just printed an RSI divergence into a record high. We bank the minimum tier - 30% - and stay long the other 70%, because the case for the name is intact: Piper Sandler lifted its target to a Street-high analysts rate it Strong Buy, and management is guiding to a tight oil market. Taking a slice of a ten-ATR gain is risk discipline, not a change of view. ADR-0012: PROFIT_ARMED (Gain_ATR 10.43, +26.51% on cost) + fire tick (a) RSI_DIV on the same row = 1 tick -> profit_take_1 = 30%. G1 softening does NOT apply (signal is FailedBreakdown_Up_20D, not BreakoutUp_*/MomentumContinuation_*, and cv69 < 80). Tick (b) not fired - per the 09-10 ruling a flip to/from NoTrade cv0 is governed by (d), not (b); (c) not fired (T12 = 1.182x close); (d) not fired. News classified OVERRIDES and applied as a DEMOTE: it caps the trim at the minimum tier and blocks any valuation bump - it does not cancel a mandatory profit-take. Valuation FAIR_VALUE MoS -0.868 (not <= -3) so no bump anyway; 12wk < DCF so no DCF-anchor warning. Balance, trail stays. Estimates use f=0.30 (calib_1w = -0.086, degenerate, L5, 12th consecutive session). · news

decided @ 214.06 · stop 205.84 · t1w 216.64 · t4w 222.66 · t12w 253.10

Kraft Heinz closed below its risk line for a fourth straight session - wider than when we sold half - and its twelve-week outlook has slipped to ~8% lower, so we exit the remaining position as the prior card planned. L2: close vs stop = -2.41% / -0.91 ATR (widened from -1.09%), 2nd close below EMA200, Target_12Week = -8.24%; position -4.34% on cost. Engine AVOID cv52 RallySell_BearTrend (-8 -> adj 44), so origination is L2 not signal. FAIR_VALUE MoS -0.618 - no tier change. Escalation ladder completes (09-11: half now, balance on next break). First NYSE session - mechanical. 1wk (f=0.30), 4wk (Target_Price), 12wk. · news

decided @ 24.27 · stop 24.87 · t1w 24.23 · t4w 24.15 · t12w 22.27

Liberty Live has traded below its risk line for five straight sessions with a twelve-week view around 14% lower and no valuation anchor to lean on - the signal has qualified for an exit every session and has been passed over four times for want of a slot. We stop deferring and cut half. L2 basis: close vs stop (-1.35%, -0.51 ATR), 5th consecutive break, Target_12Week = -14.19%, NoTrade cv0 x4. Position FIFTH deferral closed per L3/US-1 and the 09-10 promise ('carded exit on the next EOD'). Half rather than the original 09-03 89% card because the break NARROWED this session (-1.86% -> -1.35%) and the close reclaimed EMA200 on a +0.53% day - take the escalating ladder, not the whole line. Balance keeps the stop; re-fire the remainder on any further close below. Valuation is NA (negative, no FV/sh) so no amplify either way. No company news on either approved source. 1wk/4wk n/a - NoTrade row prints no Target_Price; next-session model band -.

decided @ 97.36 · stop 98.69 · t12w 83.54

American Express keeps grinding lower: the shares closed at, below every major moving average including the 200-day at, and the model's 12-week view of implies a further 13% downside. The valuation work marks it OVERVALUED on a residual-income basis, and while the business is fine - Q2 revenue grew 19% and management still guides to 10% full-year growth - the chart has been broken for months and credit-quality questions are building. We are cutting half the position and keeping the rest, rather than defending a name the market has marked down for a quarter. Trend-exit basis (close below EMA200 + negative 12wk), NOT the conviction floor: tonight's engine label flipped to a bullish ReversalUp_Oversold at cv57, which does not clear 65 and whose |sell cell has n=1. Portion held at 50% because the reversal setup argues against a full cut. 7th consecutive ask (L3/US-1); the six prior cards never executed - executor blocked since 2026-08-18. Stop on the 17,330-sh remainder held at (never lower a trail). Engine printed Target_Price ABOVE the close on this row - a long-side target, not the exit path; 4wk time-scaled from the 12wk instead. · news

decided @ 320.71 · stop 317.92 · t1w 316.44 · t4w 306.47 · t12w 278.03

We are exiting Delta completely. The stock has closed below our risk line for seven straight sessions, the position is 15.6% underwater, and the model's 12-week view of implies another 21% of downside. The reason is fuel: management flagged a $4bn increase in 2026 fuel costs and second-quarter fuel expense was up 77% year-on-year, the highest quarterly fuel bill in the airline's history. Delta has reaffirmed full-year guidance and raised its dividend 15%, but a stock the market marks lower through good guidance and into a crude-oil shock is telling a different story than its income statement. Basis L2 (decisive stop-break + negative 12wk) on four consecutive NO TRADE prints; `NoTrade|sell` bias +0 (n=57). 7th consecutive ask - the six prior full-exit cards never filled (executor blocked). Action price is the anchor close; the row prints no Entry_Price. Drop the stop from the monitor after the fill. · news

decided @ 78.24 · t1w 76.63 · t4w 72.87 · t12w 62.12

American Express keeps grinding lower: the shares closed at, below every major moving average including the 200-day at, and the model's 12-week view of implies a further 13% downside. The valuation work marks it OVERVALUED on a residual-income basis, and while the business is fine - Q2 revenue grew 19% and management still guides to 10% full-year growth - the chart has been broken for months and credit-quality questions are building. We are cutting half the position and keeping the rest, rather than defending a name the market has marked down for a quarter. Trend-exit basis (close below EMA200 + negative 12wk), NOT the conviction floor: tonight's engine label flipped to a bullish ReversalUp_Oversold at cv57, which does not clear 65 and whose |sell cell has n=1. Portion held at 50% because the reversal setup argues against a full cut. 7th consecutive ask (L3/US-1); the six prior cards never executed - executor blocked since 2026-08-18. Stop on the 17,330-sh remainder held at (never lower a trail). Engine printed Target_Price ABOVE the close on this row - a long-side target, not the exit path; 4wk time-scaled from the 12wk instead. · news

decided @ 320.71 · stop 317.92 · t1w 316.44 · t4w 306.47 · t12w 278.03

Moody's has broken down and the news now agrees with the chart. The stock closed at, 5.4% below our risk line and for the first time decisively under its 200-day average, and the 12-week view has flipped negative. Peer S&P Global's weak 2026 outlook knocked the whole ratings-and-analytics group lower, Moody's itself trimmed 2026 operating-margin guidance to 44-45%, and it has just put the business-development-company sector on negative outlook - a direct read-through to the private-credit issuance that drives its own revenue. It is a high-quality business (3rd best quality rank in this book) but marked OVERVALUED, so we halve rather than exit. Escalated exactly as promised on the 09-09 card: close stayed below and the 12wk stayed negative (-1.26% -> -4.35%). Basis is L2 + EMA200 break, not the conviction floor - `RallySell_BearTrend|sell` is the book's worst cell at -8 (n=15, hit.077), which haircuts cv52 to 44 and is why the portion is 50%, not the engine's 88.5%. Stop on the 2,502-sh remainder RE-BASED -> (close - 1.5 ATR), earned by the cut and conditional on the fill. Engine printed Target_Price ABOVE the close on an AVOID row again - 4wk time-scaled from the 12wk. · news

decided @ 467.36 · stop 450.30 · t1w 465.33 · t4w 460.58 · t12w 447.04

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.

decided @ 104.77 · stop 100.05 · t1w 109.01 · t4w 118.92 · t12w 112.90

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.

decided @ 108.68 · stop 103.90 · t1w 113.98 · t4w 123.01 · t12w 117.69

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.

decided @ 104.77 · stop 100.05 · t1w 109.01 · t4w 118.92 · t12w 112.90

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.

decided @ 108.68 · stop 103.90 · t1w 113.98 · t4w 123.01 · t12w 117.69

Carried from the 08-12 verdict, still unfilled at the executor. AAPL closed, +1.00% ABOVE the card price -- the lag has helped. Still 4x NO TRADE with a negative 12wk. Not re-issued tonight; no stacking.

decided @ 302.25 · stop 300.00 · t12w 295.85

Carried from 08-12, still unfilled. BAC closed, -1.11% BELOW the card price -- the lag cost on the slice. Conviction continued to collapse 45 -> NO TRADE cv0. Not re-issued; no stacking.

decided @ 64.81 · stop 62.54 · t12w 76.60

Re-fires 30% (armed + tick (c), 12wk = -8.2%) after the 08-11 30% cut filled at. Label degraded HOLD cv52 -> AVOID cv50 on FailedBreakout_Down_20D. The overlay's standing 'invalidated when close below 345' override stays invalidated (close). Deferred on slot arithmetic only. Note the printed Stop_Price is a SHORT-side level on an AVOID row -- ignored; the long stop stays. · news

decided @ 343.54 · stop 319.70 · t12w 315.34

Apple has fallen about 10% since its 30-July earnings and the model has refused to give it a tradable signal for five straight sessions while its twelve-week outlook has slipped below the current price. Management's own flag of a '100-year flood' in memory pricing is a real margin drag, Jefferies moved to underperform with a target on 10-Aug, and the position is now underwater for us. We halve what is left rather than wait for a signal the engine is not producing. The counterweight is honest: the 46-analyst consensus target of still sits about 7% above the price and Wedbush is at so this is a de-risk into an unresolved argument, not a verdict that Apple is broken. OPS: ADR-0012 armed (T12_Progress 102.3) + tick (c) dead 12wk (, -2.2%) + tick (d) 5 consecutive NO TRADE -> 2 ticks -> profit_take_2 = 50%. HONEST CAVEAT: Unreal -2.07%, Gain_ATR -0.83 -- this is a LOSING position wearing a PROFIT_ARMED flag; the arm comes purely from the T12_Progress clause (Systemic finding 1, third night). Valuation MoS -2.83 -> val_adj -10 but no tier bump (-2.83 > -3). Scorecard NoTrade|sell flipped to +1 this run (n=19, hit_1w.538). No Target_Price on a NO TRADE row -> 4wk n/a, 1wk interpolated off the 12wk at f=0.30 (L5, calib_1w 0.006 degenerate). 12wk exceeds the DCF anchor -- projection assumes growth beyond priced-in. Cumulative cut reaches 82.5% of the 49,865-sh seed lot. Stop held on the residual. · news

decided @ 302.25 · stop 300.00 · t1w 300.23 · t12w 295.52

Moody's has gone four sessions without a tradable signal, its twelve-week outlook has now dipped below the current price, and the company itself cut 2026 operating-margin guidance while leaning harder on buybacks -- a company its share price rather than growing into it. On top of that, investors are actively re-rating Moody's exposure to deteriorating private-credit recoveries, and the most-followed fair-value narrative now sits below the market price. We take the second half of the position off and keep a third of the original stake. OPS: ADR-0012 armed (T12_Progress 100.5) + tick (c) dead 12wk (, -0.5%) + tick (d) 4x NO TRADE -> 2 ticks -> profit_take_2 = 50%. This card was ledgered optional:true on 08-11 and deferred once -- L3 says stop re-deferring, so it executes tonight. Same PROFIT_ARMED-on-a-loss caveat as AAPL (Unreal -2.20%). Valuation MoS -1.00 -> val_adj -5, no tier bump. Ex-div on 08-14: selling at the 08-13 open forgoes on the sold slice -- immaterial, stated. Stop held on the residual. · news

decided @ 477.84 · stop 461.27 · t1w 477.11 · t12w 475.42

A live SELL - REDUCE (Portion 49%, conviction 55, BreakdownDown_20D) with the most negative forward view in the book: 12wk = -33.3%. Already halved twice (08-07 and 08-12 fills). Deferred on size --. RSI 33.4 continues to argue against selling the low a fifth time. Stop raised ->.

decided @ 63.95 · stop 59.62 · t12w 42.66

DaVita fell 17% on results that were actually fine - adjusted EPS, revenue 55bn, treatment volumes ahead of plan and full-year guidance reaffirmed. What the market repriced is forward reimbursement: revenue per treatment slipped on a weaker commercial insurance mix, and CMS has proposed folding phosphate binders into the 2027 dialysis bundle at a rate update below the industry's cost trend. That is a structural de-rating rather than a bad quarter, so we halve the position to respect a risk line that was broken by 15% and keep the rest, because the rule is only proposed and the shares are deeply oversold. Exit basis: decisive stop-loss break (close 188.69 vs standing 222.80 = -15.31%) on 1.8x volume, closing near the 183.62 session low; anchor row NO TRADE cv0, Indeterminate consensus. Tier: 30% base bumped to 50% by MoS_FV -3.97 <= -3 (CLAUDE.md section 9b). G1/G2/G3 inapplicable - close is 17.4% below EMA20, this is loss prevention not a profit-trim. Stop RE-BASED 222.80 -> 158.04 (2.0xATR chandelier; the gap tripled ATR to 15.33). WATCH LINE: EMA200 - a decisive close below escalates the remaining to a full exit. 1wk/4wk n/a - Target_Price is blank on a NO TRADE row; engine band 132.08-205.26 tops out BELOW the 12wk 211.34. L2, L3 (execute promptly), L8. · news

stop 158.04 · t12w 211.34

The New York Times has a structural problem, not a bad quarter. Q2 added about 280,000 digital subscribers against 295,300 expected and 310,000 the quarter before - and it did that in a heavy news cycle, which is when this business is supposed to shine. Management pointed at falling Google search and referral traffic, and guided Q3 subscription revenue growth below consensus. The chart agrees: the stock fell more than 13% on the print, sits 10% below its 200-day average with RSI at 32, and the engine now prints STRONG SELL with a 12-week target of 51.95, nearly 20% under the current price. Half the position comes off here on top of the 30% already sold, leaving a smaller stake to close out if the selling label holds another session. Ops: 1,592 of at ~64.77, realizing roughly -on a -12.30% position; L2 satisfied on both limbs (stop break + negative 12wk); 50% rather than the engine's 88.52% because this is the first STRONG SELL print in the trail and MoS -1.81 is short of the -3 that would bump the tier; stop on the remainder raised to 56.44; 12wk target sits below the 68.16 DCF anchor so no over-projection flag. · news

decided @ 64.77 · stop 56.44 · t1w 63.70 · t12w 51.95

Occidental has run 10% above our cost and the model's own 12-week view has now rolled over below the current price, with the shares slipping under both their 20- and 50-day averages. On an oil name priced well above its cash-flow anchor, that combination is the signal to bank half the gain and let the rest run on a wider stop. The three-session hold for the earnings date is over - Q2 was released after Wednesday's close. ADR-0012: armed (PROFIT_ARMED, Gain_ATR 2.74, T12_Progress 103.76%, +10.02%) + tick (c) dead forward view (12wk 51.86 < 1.02 x 53.81 = 54.89, -3.62%). One tick -> profit_take_1 30%, bumped to 50% by MoS_FV -3.17 <= -3 (section 9b). G1/G2/G3 inapplicable - not an uptrending name (below EMA20 54.94 and EMA50 54.81, RSI 46.2, -2.3% on the day). EXECUTION CAVEAT (L8): Q2 released post-close 08-05, call 08-06 1pm ET; consensus was adj EPS on 18bn rev, production 1,390-1,430 Mboe/d. If the 08-06 open gaps more than ~1.5%, REPRICE this card before executing. Stop RE-BASED 53.77 -> 50.23 (model) - the inherited trail sat 0.02xATR below spot, a guaranteed whipsaw. 12wk exceeds the DCF anchor. L3, L8. · news

decided @ 53.81 · stop 50.23 · t1w 54.88 · t4w 57.39 · t12w 51.86

The C-share line is trimmed alongside the A-shares for the same reason and at the same time: two consecutive NO TRADE prints from the signal engine and a 12-week target of 347.57 that sits below today's 356.62 close, against a news backdrop of senior AI departures from DeepMind and a reported ~$80bn equity raise. Holding one Alphabet share class while cutting the other would leave the book with two contradictory views of one company. Ops: 161 of; ADR-0012 ticks (c)+(d), tier held at 30% per the 08-05 pre-commitment; G2 suppresses valuation-amplify (DCF 145.29); trail on the remainder held at 347.25. · news

decided @ 356.62 · stop 347.25 · t1w 355.86 · t12w 347.57

Alphabet is being de-risked, not written off. The signal engine has printed NO TRADE on two consecutive settled closes - it cannot read the name - and the 12-week view has turned negative at 347.13 against a 357.75 close, so the model sees no upside to pay for the uncertainty. The news explains why: Jeff Dean and Sanjay Ghemawat are leaving DeepMind to found their own company, which knocked the stock 4% on 05-Aug despite a Q2 that beat, and Reuters reports Alphabet seeking roughly $80bn in equity offerings on top of an already-raised capex guide. Valuation adds nothing to the case in either direction - at a 145 DCF anchor against a 347 price the intrinsic model has no view worth acting on - so this is a signal-originated trim, taking a third of the line off and holding the rest. Ops: 2,422 of; fires ADR-0012 ticks (c) dead 12wk + (d) 2x NO TRADE; tier held at 30% not the mechanical 50% because Unreal_PnL is -0.60% so the profit-take precondition is not truly met (see Systemic finding 1) and because the 08-05 card pre-committed to exactly; G2 suppresses the valuation-amplify path; trail on the remainder held at 348.52; cross-book: the us book fired the identical trim on tick (d) tonight, separate share lot, no L7 conflict. · news

decided @ 357.75 · stop 348.52 · t1w 356.86 · t12w 347.13

Nucor has been the book's best trade - up 24.1% on cost, 55% year to date - and the steel case is still intact, with hot-rolled coil back near a short ton on mill price increases, lengthening lead times and tariff-compressed imports, on top of a record quarter of 7.1m tons shipped and adjusted EPS of. But the stock has now run past the entire sell side: the 16-analyst consensus target of sits 6% BELOW the close, and the shares are printing a momentum divergence at an RSI of 71.5. That is the moment to bank a third of a very large gain and let the rest run. OPERATIONAL: ADR-0012 armed (Gain_ATR 5.26, the book's largest) + tick (a) RSI_DIV alongside PROFIT_ARMED. One tick -> profit_take_1 = 30%. This is a DIFFERENT tick from the four prior sessions, which fired only tick (d) on a structural Conflicted NoTrade and were correctly cancelled; the 07-31/08-03 bright line at is NOT hit (close 12.9% above it). G1 does not shelter it - the softening needs a fresh breakout/momentum print at conviction >= 80 and this is cv63. MoS -2.40 > -3, no tier bump. Not the NBIS error: that was a full valuation-basis trim into a trending breakout; this is the 30% minimum with (70%) left fully exposed and a 12wk of +22.4%. Hold; trail RAISED 242.55 -> 253.79 and the bright line re-bases with it - a close below trims the remainder with no further news exemption. 12wk exceeds DCF anchor. · news

decided @ 274.04 · stop 253.79 · t1w 291.79 · t4w 333.22 · t12w 335.53

America's second-largest bank just posted record net interest income and 34% earnings growth, and management guides full-year NII to the upper end of 6-8% - a business that gets BETTER if the Fed holds or hikes, which is exactly the risk the 29 July FOMC carries. The chart has caught up to the story: the pullback that never came has resolved into a fresh momentum-continuation breakout, and we are buying strength on a stop order rather than chasing it. The one caution is price versus book - at 1.5x tangible book the shares screen mildly rich on our residual-income model, so we add a measured tranche, not a full one. OPS: GTC buy-STOP @62.55 (fills only on strength, self-limiting into the FOMC print); = 29.97% ADD tranche x 1.0 size mod =. Entry band 62.27-62.83. Supersedes and WITHDRAWS the unfilled 07-24 PullbackBuy_VWAP limit @60.06 - family flipped PullbackBuy->MomentumContinuation per US-8, so the old order is void and the action price is repriced to the live 07-27 Entry_Price. MomentumContinuation_Up|buy cell is n=1 (report-only, 0-of-1) - no bias applied, but flagged. Quality_Pass False does NOT gate an ADD to a held name. RSI 68.7 with RSI_DIV but <75, so not a chase; EDATR +0.37 = at/below entry. expires_on 07-30. · news

decided @ 62.55 · stop 60.29 · t1w 64.58 · t4w 69.33 · t12w 74.84

Moody's Q2 beat and raised full-year guidance (revenue +15%, adjusted EPS +31% to, adj op-margin +440bp to 55.3%, FY adj-EPS midpoint) — a clean post-earnings PullbackBuy_VWAP we add to on a disciplined pullback rather than chasing the +1.8% post-beat close. Queued GTC limit @480.93 (1.8% below close),, size x0.75 for the -1.33 MoS; extended -0.61 ATR so pullback-only, no gap-chase (L6). Adj clears the 65 floor. 12wk 556.59 exceeds the DCF anchor. expires 07-27. · news

decided @ 480.93 · stop 464.18 · t1w 490.99 · t4w 514.49 · t12w 556.59

Chubb printed a fresh STRONG BUY-ADD on the reliable pullback-to-VWAP setup (cv85, Buy_Rank 1) as its Q2 results landed 07-21; we queue a below-close buy-limit that fills only on a pullback so we ride the trend without chasing the earnings pop. Extended -0.84 ATR past the 348.24 entry -> QUEUE not market (no-chase). Model add portion 99.33pct x 8,216 = (, a near-doubling ADD; within 60pct max-position +; scalable at user discretion). Val neutral (OVERVALUED_VS_RI MoS -0.028, val_adj 0, size x1.0). 12wk 386.36 exceeds RI FV anchor 345.17 -> growth-beyond-priced-in. L6-safe pullback-only limit into the Q2 call 07-22 8:30 ET. expires_on 07-24; execute 07-22 open. · news

decided @ 348.24 · stop 332.71 · t1w 353.65 · t4w 366.26 · t12w 386.36

Profit-take on an armed winner (+10.7%, Gain_ATR 4.8): conviction collapsed and the 12wk view is dead (176.34 < spot 187.38) despite a fresh BreakoutUp label — the forward view and consensus disagree with the breakout, so bank 30%. Already halved 07-14; one new tick -> profit_take_1 (30%), keep 1,104 to ride. G2 suppresses valuation amplify (DCF 82 unreliable vs Target 176). Raise chandelier trail 174->179.80. Executes at the 07-20 open (L8); reprice if the open gaps >1.5%.

decided @ 187.38 · stop 179.80 · t1w 184.07 · t4w 208.46 · t12w 176.34

A +15.5pct armed winner (Gain_ATR 5.02, T12 105.7pct) whose signal deteriorated hard this session — conviction collapsed (a bullish add-candidate flipping to a neutral hold) on a dead 12wk — so we bank a measured 30pct and keep 70pct to ride the trail. Fire tick (b) conviction collapse; persistent dead-12wk (c) does not escalate to 50pct (optional_trim counterfactual: trims give back -0.6pct). Valuation MoS -3.087 would bump but G2 suppresses the amplify (DCF 33.47 << Target 53.44). Keep; raise trail 52.06->53.48 (2.0xATR chandelier). News neutral: Evercore Outperform PT is 07-08 stale/priced-in, supports keeping the residual, does not cancel a signal-based partial. 4wk n/a (65.11 target not conviction-supported at cv22). · news

decided @ 56.50 · stop 53.48 · t1w 55.58 · t12w 53.44

Loss-prevention: KR broke its stop (60.03, -2.37%) and flipped to STRONG SELL-REDUCE below its 200-day EMA with a sharply negative 12-week view (43.91, -25%) — an L2 stop-break + neg-12wk exit, and not-trimming it on 07-09 already cost 5.5%. But KR is UNDERVALUED with Quality_Pass True (F-Score 7), so valuation argues down one tier: take only 50% ('sell the cheap, quality name at the minimum') and keep on stop 54.95. Executes at the 07-17 open (L8).

decided @ 58.61 · stop 54.95 · t1w 54.20 · t12w 43.91

Lennar's spring selling season came in weak — UBS cut its target to (Neutral) and JPMorgan to (Underweight) on lowered FY26 guidance, and the stock is -13.6% YTD. Technically a stable 3-print STRONG SELL-REDUCE trading below its 200-day EMA with a dead 12-week view (70.20, ~-19%): a valid trend exit despite the -7 whipsaw cell, since the below-200DMA + negative-forward stack is structural. Exit the Lennar complex (paired with LEN-B); residual winds down on stop 80.26. Executes at the 07-17 open (L8). · news

decided @ 86.40 · stop 80.26 · t1w 81.54 · t12w 70.20

Class B of the Lennar exit — broke its stop (88.14, -3.86%, loss-prevention) and sits below its 200-day EMA with a dead 12-week view (70.18, -17%). Same downgrade-driven housing weakness as LEN (UBS/JPM target cuts, weak spring demand). Trim the complex together; residual winds down on stop 79.90. Executes at the 07-17 open (L8). · news

decided @ 84.74 · stop 79.90 · t1w 80.37 · t12w 70.18

Genuine fired profit-take: NUE reached its 12-week target (T12_Progress 100.2%, +6.8% unrealized, Gain_ATR 2.69) and then its signal collapsed from cv58 to NO TRADE cv0 ('Consensus conflict + low conviction') with a dead-flat forward view — bank half and protect the gain while the setup is spent. Armed + ticks (b) conviction collapse and (c) dead-12wk -> profit_take_2 50%. Book's most material sell candidate; keep on a raised trail Executes at the 07-17 open (L8).

decided @ 235.67 · stop 223.00 · t1w 235.51 · t12w 235.15

Genuine profit-take on the day-1 89%-trim residual (, avg cost 169.20) now +7.4%: Gain_ATR 3.33, T12_Progress 114.5% (12-week target exceeded). Fired by (c) dead/negative model 12wk (158.81 < close 181.76) and (d) two consecutive NO TRADE prints -> profit_take_2 50% = @ (, ~+ banked). CVX is -10% over the past month as oil pulled back from its May peak (guidance intact) — confirms the partial de-risk. Keep the retained and raise the trail to. Valuation OVERVALUED_VS_RI (MoS -0.03) negligible. L8: execute at the 07-15 open / closing auction. · news

decided @ 181.76 · stop 174.00 · t1w 174.88 · t12w 158.81

Immaterial residual that qualifies for an L2 exit — stop-break (53.96 < 58.14, -5.07%) + negative 12wk (51.68) — but is too small to earn a full card. Ledgered optional so the counterfactual grader prices the tail; winds down on its stop. FailedBreakout_Down|sell -7 whipsaw caution applies.

decided @ 53.96 · stop 58.14 · t12w 51.68

Residual clean-up exit on the L2 basis: a decisive close below the carried stop (75.41 < 77.96, -3.27%) paired with a negative 12-week projection (70.16 vs 75.41) is itself a signal-based exit (CLAUDE.md §9 / learnings L2), even though the engine still prints AVOID/NO-TRADE. Extreme overvaluation (MoS_FV -5.20) amplifies (§9c). Full exit of the 1,412-sh sliver @, realizing -4.9%. Cell caution noted: FailedBreakout_Down|sell grades -7 (whipsaws), but the stop-break + neg-12wk + MoS -5.2 stack is a strong combined basis and this is an immaterial residual. L8: execute at the 07-15 open.

decided @ 75.41 · stop 77.96 · t1w 73.84 · t12w 70.16

First GENUINE profit-take on this book: the day-1 89%-trim residual (, avg cost 48.91) has run to +11.6% and crossed a real arm — Gain_ATR 3.2, T12_Progress 116.6% (the 12-week target is exceeded). Deterioration ticks fire it: (c) the model's forward view is now negative (12wk 46.81 << close 54.57) and (d) two consecutive NO TRADE prints. Per ADR-0012 that is a signal-based reason to bank the winner: profit_take_2 = 50% = @ (, ~+ banked). Oil rolling off the May peak (WTI) confirms the de-risk; but a fresh Evercore upgrade (Outperform, PT) argues to keep the other half — hold and raise the chandelier trail to. Distinct from the persistent spurious seed-cost PROFIT_ARMED (this arm is a real positive Gain_ATR>=3). L8: execute at the 07-15 open / closing auction; reprice if it gaps. · news

decided @ 54.57 · stop 51.00 · t1w 52.24 · t12w 46.81

Pyramid add on a fresh MomentumContinuation_Up STRONG BUY-ADD (cv93, Buy_Rank 1, conviction climbing into a confirmed Q2 earnings beat (07-14 6:45am ET: EPS 13e +27% YoY, rev 62Be) — the market's post-print reaction, so no front-running. Held core so the failed quality gate does not block the ADD; mild overvaluation (OVERVALUED_VS_RI, MoS -0.30) trims adj conviction to 91, well clear of the floor. QUEUED as a GTC buy-stop @61.17 (fills only on continued strength, self-limiting) — sized down from the model's 99.33% (full double, +) to a measured ~35% pyramid of because L6/US-9 forbid chasing a post-earnings gap on a core mirror name. Stop 58.37; RSI 69.6 + RSI_DIV is a G1 caution not a block. Expires 07-17. L8/US-9: if the 07-15 open gaps >~1.5% past 61.17 (>), WITHDRAW and re-arm on a pullback to VWAP/EMA10 — do not chase. · news

decided @ 61.17 · stop 58.37 · t1w 63.69 · t4w 69.58 · t12w 72.27

Small pyramid add on the L1-favored PullbackBuy_VWAP family (Buy_Rank 2) with a proper pullback-limit entry below the close (not a chase). Marginal — adj conviction 66 clears the 65 floor by one; the PullbackBuy cell is only report-only here (n=2) so no positive bias applied. NA valuation so the quality gate is off and val_adj is 0. QUEUED GTC buy-limit @103.48 (range 102.77-104.19), + (, Portion 29.97% x 2,462) — kept small: thin/illiquid Liberty Live tracking stock and the add sits on a -3.6% position. Stop 97.81. Expires 07-17. L8: reprice if the open gaps >1.5%.

decided @ 103.48 · stop 97.81 · t1w 105.45 · t4w 110.06 · t12w 115.07

Complete the KR residual de-risk: stable STRONG SELL - REDUCE (FailedBreakout_Down_20D) across 3 snapshots (conv ->65, weakening but still STRONG) with a deeply negative 12wk 41.95 (-29.3% vs 59.32 close) -- a signal-based trend exit (CLAUDE.md §9 / L2). News neutral (Giant Eagle acquisition, new CPO, retail-media/AI push vs margin pressure; analyst consensus above price but no override of an active trend-sell). Reduce 89% = @ ~59.32 (Market, next open); residual on raised chandelier stop 55.72. Valuation NA (no amplify). Trim-discipline G1-G3 N/A (bearish downtrend signal). L8 pending_next_open, execute 07-09, reprice if open gaps >1.5%. · news

decided @ 59.32 · stop 55.72 · t12w 41.95

STRONG SELL - REDUCE on a fresh RallySell_BearTrend flip with strengthening conviction and a negative 12-week view (75.71 vs 86.74 close, -12.7%; 4wk target 85.80, -1.1%) -- a signal-based trend exit (CLAUDE.md §9 / L2). Homebuilder rolling over into a hawkish-rate week reinforces the sell but is not required for it. Valuation GROWTH_PRICED_IN (MoS_FV -0.78, Quality_Pass F) amplifies only, no tier bump. Reduce 89% = @ ~86.74 (Market, next open); residual on reset chandelier stop 80.90. L8 pending_next_open, execute 07-08, reprice if open gaps >1.5%. · news

decided @ 86.74 · stop 80.90 · t4w 85.80 · t12w 75.71

STRONG SELL - REDUCE (FailedBreakout_Down_20D, conv 69) flipped down from 07-06, carrying the deepest structural downside in the pool (12wk 100.36 vs 131.76 close, -23.8%) -- a signal-based trend exit (CLAUDE.md §9 / L2). Small dollar size (204-sh full position) but a legitimate de-risk of a rolling-over staple. Valuation GROWTH_PRICED_IN (MoS_FV -0.92, Quality_Pass T) amplifies only, no tier bump. Reduce 89% = @ ~131.76 (Market, next open); residual on reset chandelier stop 120.50. L8 pending_next_open, execute 07-08, reprice if open gaps >1.5%. · news

decided @ 131.76 · stop 120.50 · t12w 100.36

Stable STRONG SELL - REDUCE (FailedBreakout_Down_20D, conv 68 across all three sessions) with a deeply negative 12-week projection (218.81 vs 266.78 close, -18%) -- a valid signal-based trend exit per CLAUDE.md §9 / learnings L2, extending the 07-06 de-risk to the next tier. News confirms: VRSN fell after Berkshire Hathaway CUT its stake, with an AI-chatbot navigation shift and a core-contract-renewal overhang weighing on the name. Valuation GROWTH_PRICED_IN (MoS_FV -1.93, deepest full position -> up-ranked) amplifies only, no tier bump (not <=-3). Reduce 89% = @ ~266.78 (Market, next open); residual on reset chandelier stop 251.00. Trim-discipline G1-G3 N/A (bearish-signal REDUCE, not a profit-trim on an uptrend). L8 pending_next_open, execute 07-08, reprice if open gaps >1.5%. · news

decided @ 266.78 · stop 251.00 · t12w 218.81

Fresh multi-snapshot add on the bankable PullbackBuy_VWAP cell (learnings L1) with conviction rising ->66 across the last three sessions and a strong 12wk view (49.76, +9.9%). Queued as a GTC buy-limit at 45.28 (below the 45.40 close) so it fills only on a pullback -- self-limiting into an event-gated week. Size halved (995 = 0.5x the 1,990-sh model ADD) for the FOMC-minutes 07-08 no-new-leverage posture + de-risk-into-CPI-07-14 bias and the marginal conviction / OVERVALUED_VS_RI, Quality_Pass F read (gate off -- ADD to a held name). val_adj 0. Cost if filled; stop 42.84. L8 pending_next_open, execute at the 07-08 open, reprice if the open gaps >1.5%; auto-withdraw if unfilled by 07-10 or if the family flips off the PullbackBuy cell (US-8). · news

decided @ 45.28 · stop 42.84 · t1w 46.13 · t4w 48.12 · t12w 49.76

Stable STRONG SELL - REDUCE (FailedBreakout_Down_20D) across both 07-06 close and 07-07 pre-open snapshots; negative 12-week projection (123.67 vs 169.20 close) confirms a rolling-over trend — a valid signal-based exit per CLAUDE.md §9 / learnings L2. Book is at cost (seeded 07-06), so no profit-take involved; the PROFIT_ARMED flag is spurious (T12_Progress>90 on a below-price bearish 12wk target — KLAC/LMT bug). Valuation GROWTH_PRICED_IN (MoS_FV -1.97) amplifies only, does not originate. News neutral — general energy/staples pullback, no name catalyst. Reduce 89% = @ ~169.20 (Market, next open); residual held on stop 176.55. L8: pending_next_open, execute 2026-07-07, reprice if open gaps >1.5%. · news

decided @ 169.20 · stop 176.55 · t12w 123.67

Complete the KR residual de-risk: stable STRONG SELL - REDUCE (FailedBreakout_Down_20D) across 3 snapshots (conv ->65, weakening but still STRONG) with a deeply negative 12wk 41.95 (-29.3% vs 59.32 close) -- a signal-based trend exit (CLAUDE.md §9 / L2). News neutral (Giant Eagle acquisition, new CPO, retail-media/AI push vs margin pressure; analyst consensus above price but no override of an active trend-sell). Reduce 89% = @ ~59.32 (Market, next open); residual on raised chandelier stop 55.72. Valuation NA (no amplify). Trim-discipline G1-G3 N/A (bearish downtrend signal). L8 pending_next_open, execute 07-09, reprice if open gaps >1.5%. · news

decided @ 59.32 · stop 55.72 · t12w 41.95

Stable STRONG SELL - REDUCE (FailedBreakout_Down_20D) across both 07-06 close and 07-07 pre-open snapshots; negative 12-week projection (31.92 vs 48.91 close) confirms a rolling-over trend — a valid signal-based exit per CLAUDE.md §9 / learnings L2. Book is at cost (seeded 07-06), so no profit-take involved; the PROFIT_ARMED flag is spurious (T12_Progress>90 on a below-price bearish 12wk target — KLAC/LMT bug). Valuation GROWTH_PRICED_IN (MoS_FV -2.65) amplifies only, does not originate. News neutral — general energy/staples pullback, no name catalyst. Reduce 89% = @ ~48.91 (Market, next open); residual held on stop 51.47. L8: pending_next_open, execute 2026-07-07, reprice if open gaps >1.5%. · news

decided @ 48.91 · stop 51.47 · t12w 31.92