US Trades Decided by: Manual

US Main — Trades

Every virtual fill this book has made, with the model's reasoning — the fused signal, valuation and news case — behind each. Prices are model fills; quantities are never shown.

All books

Fill history

10 recorded fills.

Date	Ticker	Name	Side	Fill Px	Signal	Conv	Valuation	Why
2026-08-01	META	Meta Platforms	SELL	556.00	RallySell_BearTrend	63	FAIR	rationale Exit the whole Meta position. Q2 revenue grew 28% to 8bn but earnings missed, free cash flow collapsed ~91% to and 2026 capex was guided up to ~$135bn - the market cut the stock 8% on the print and it has not recovered. The shares sit 10% below our stop for a ninth session, 12% under our cost, and below every moving average we track, while the model projects no recovery at any horizon out to 12 weeks. Valuation is only fair, so there is nothing cheap to wait for. \| L2 fires both legs (decisive stop break -10.21% + dead 12wk: Est_12W 565.58 < 1.02x close 567.84). 2nd consecutive settled STRONG SELL - REDUCE (BreakdownDown_52W cv85 -> RallySell_BearTrend cv68, Sell_Rank 3). adj 63 sits below the 65 floor after the -5 RallySell_BearTrend\|sell haircut, but L2 is an independent exit basis. MoS -0.92 > -3 so no valuation tier bump. 8th re-fire, repriced 539..71. Stop held at the broken 620.00 for monitor continuity until the exit fills. Analyst median PT noted and rejected - trade the signal. · news decided @ 556.71 · stop 620.00 · t1w 557.12 · t4w 549.03 · t12w 565.58
2026-08-01	NET	—	BUY	258.00	—	—	—	—
2026-08-01	GOOGL	—	BUY	328.00	—	—	—	—
2026-07-22	GLW	Corning Incorporated	SELL	160.00	FailedBreakout_Down_20D	69	GROWTH PRICED IN	rationale Corning has now printed a bearish signal for six straight sessions and today it escalated to an outright STRONG SELL - REDUCE at the settled close, ranked the single worst name in the book to hold, with conviction rising through the session. The stock is trading below both its short- and medium-term trend lines, is our only large loser at roughly -19%, and rallied noticeably less than the rest of the book on a broadly explosive day - relative weakness that confirms the problem is company-specific rather than market-wide. Valuation offers no cushion either, so we take the full exit into today's bounce rather than hold a broken position for a sixth session. OPERATIONAL: Sell_Rank 1, 88.52 amplified to full by MoS_FV -11.34 (<= -3 tier bump). Stable family across J20C->J21C (cv 52/62/62/69), close 162.41 below EMA10 172.51 and EMA100 169.41. US-18 second-print bar met many times over; G-gates inapplicable (broken downtrend, CLAUDE.md 9 'valid exits still fire'); US-23 - the stale-bullish Est_12W +6.8% must not veto a signal/stop exit, and Est_4W is already negative. Card outstanding since 07-16, repriced +6.1% higher than the 07-21 ref of 153.10. Proceeds realized ~-Cross-book: executes ONCE here (L7). decided @ 162.41 · t1w 163.98 · t4w 160.33 · t12w 173.41
2026-07-22	OKTA	—	BUY	139.80	—	—	—	—
2026-07-21	SNDK	Sandisk Corporation	SELL	1388.06	NoTrade	—	GROWTH PRICED IN	rationale Sandisk has run 16% past its twelve-week target and the model's forward view has now gone negative (-13.7%), while the signal engine has fallen silent for four straight snapshots. That combination is the textbook case for banking half a winner rather than waiting for the reversal to announce itself. The NAND/HBM supercycle looks intact - the stock still sits above its 100- and 200-day trend lines and next week brings SK Hynix, Samsung and Lam Research read-throughs - so we keep the other half running with a raised trail. Operational: ADR-0012 profit_take_2, armed at T12_Progress 115.9% with ticks (c) dead 12wk forward [1200.54 < 1.02x1390.95 = 1418.77] and (d) 4x consecutive NO TRADE. Self-set 1300 trail is INTACT - pure armed + soft-2-tick fire, not a trail break. US-19: same 50% first-fire repriced +2.7% higher from 07-18, re-verified at the settled close, NOT an escalation. Valuation amplify (MoS -4.77 <= -3 would bump is G2-SUPPRESSED (DCF 1005.14 << Target_12Week, Rev_Growth 0.98 hyper-growth). Keep, trail 1300. Est_12W exceeds the DCF anchor 1005.14. Proceeds. decided @ 1390.95 · stop 1300.00 · t1w 1428.44 · t4w 1385.08 · t12w 1200.54
2026-07-21	MU	—	BUY	868.00	—	—	—	—
2026-07-18	META	—	BUY	635.00	—	—	—	—
2026-07-08	NVDA	NVIDIA Corporation	SELL	195.50	FailedBreakout_Down_20D	69	GROWTH PRICED IN	rationale Trim NVDA hard: the signal has printed STRONG SELL - REDUCE for 4 straight snapshots with a negative 12-week outlook (target 168 vs 195.55 close), and the news is structural not panic - a sector-wide chip selloff plus the Kyber NVL144 rack slipping to 2028 and hyperscalers building custom silicon. Stable strong-sell + neg-12wk is the valid trend exit (L2), unaffected by trim gates; the position is also PROFIT_ARMED. Keep 69, invalidate above 207.68. Closes the NVDA NO-TRADE bleed (US-1). · news decided @ 195.55 · stop 207.68 · t12w 168.00
2026-07-07	NVDA	NVIDIA Corporation	SELL	197.00	FailedBreakout_Down_20D	69	GROWTH PRICED IN	rationale Trim NVDA hard: the signal has printed STRONG SELL - REDUCE for 4 straight snapshots with a negative 12-week outlook (target 168 vs 195.55 close), and the news is structural not panic - a sector-wide chip selloff plus the Kyber NVL144 rack slipping to 2028 and hyperscalers building custom silicon. Stable strong-sell + neg-12wk is the valid trend exit (L2), unaffected by trim gates; the position is also PROFIT_ARMED. Keep 69, invalidate above 207.68. Closes the NVDA NO-TRADE bleed (US-1). · news decided @ 195.55 · stop 207.68 · t12w 168.00

Learnings in force

What this book's own record has taught it.

Last reviewed 2026-07-18

Favor PullbackBuy_VWAP entries over momentum-chase families (FailedBreakdown_Up, BreakoutUp, MomentumContinuation).

PullbackBuy_VWAP|buy is the sole reliable buy cell (~0.77-0.86); the three momentum-chase families are negative traps (US-16). Treat a chase-family BUY as non-actionable unless raw conviction clears the US-16 floor; prefer the PullbackBuy.

Trim on the stop-break + negative-12wk combo; don't wait for the SELL-REDUCE label.

A decisive close below the recorded stop AND a negative 12wk is itself a signal-based exit (CLAUDE.md §9), even while the engine still prints NO TRADE.

Execute the de-risk promptly.

A trim recommended but not executed keeps bleeding; act when a held name closes below stop with a negative 12wk, don't re-recommend the same trim for days.

The settled close is authoritative; don't act on weak opens.

Intraday/BOD prints whipsaw. Flag intraday reads provisional; reserve firm acquire/dispose for the settled-close EOD.

Use f=0.30 as the STANDING default for the 1wk estimate (calib_1w is structurally unstable on this book).

calib_1w oscillates across zero week to week (-0.045 -> +0.113 -> -0.025); f=0.30 is the default, not just the degenerate-case fallback, until n grows or the scorer winsorizes calib (US-15).

Don't chase post-earnings gaps.

A blowout (e.g. MU +14% AH) often round-trips within days; wait for the pullback to the model entry.

Same-pool/cross-book shares execute ONCE.

The parent book's decision governs; sub-book runs flag conflicts and avoid double- the shared cash. Tuned sub-book configs can diverge on the same share — surface the conflict, don't silently double-act.

EOD recommendations execute at the NEXT market open, not the decision close.

Persist actions as `execution_status: pending_next_open` with an `execute_on` date; action prices are close-refs — reprice/re-validate if the open gaps >~1.5%; the 1wk/4wk/12wk horizons and the scorecard grade from the execution date/price; live holdings/monitor stay unchanged until the JSON is updated post-fill; the next BOD reconciles pending proposals against the actual open.