AI · US Semiconductors — 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.
Fill history
3 recorded fills.
| Date | Ticker | Name | Side | Fill Px | Signal | Conv | Valuation | Why |
|---|---|---|---|---|---|---|---|---|
| 2026-07-18 | TSM | Taiwan Semiconductor (ADR) | SELL | 392.12 | RallySell_BearTrend | 37 | GROWTH PRICED IN | rationaleTrim TSM by half: a record Q2 beat (rev +36% YoY, profit +77%, FY26 guide raised >40%) still sent the stock down ~3-4% on a raised -64B capex + margin worry, and the settled 07-16 close 409.74 is the post-print decider firing bearish. Signal escalated NO TRADE -> STRONG SELL - REDUCE (RallySell_BearTrend, Sell_Rank 1), the 12wk is dead (411.26, +0.4%), the name is -6.46% underwater, and the de-risk has gone unexecuted 4 straight sessions. Half, not full: still above EMA200 and the 375.56 stop, FY guide strong, 12wk not yet negative (no L2 full-exit basis). Escalated from the 30% catalyst-cap now that earnings passed. val_adj 0 (G2 DCF blind); PROFIT_ARMED spurious. Retains 211 on the 375.56 stop; further exit on a close < EMA200 358.93 or < 375.56. Frees. · news decided @ 409.74 · stop 375.56 · t12w 411.26 |
| 2026-07-08 | AMD | Advanced Micro Devices | BUY | 516.08 | FailedBreakdown_Up_20D | 66 | GROWTH PRICED IN | rationaleStable STRONG BUY x4 (conviction on a FailedBreakdown_Up_20D reversal at entry (RSI 57, EntDist 0). News confirms and matches the signal: AMD +8.3% Mon 07-06 recovering an ~11% late-week drop (the failed breakdown the signal caught); Goldman PT Wells Fargo; Advancing AI event 07-22/23. Valuation GROWTH_PRICED_IN, MoS_FV -17.1 -> bounded val_adj -10 lands adj at 66 (clears floor by +1). The FV anchor is a broken hyper-growth artifact (FV/sh vs; DCF for a 45% grower) -- the systemic note stands; expressed conservatively via size_mod 0.5. 12wk 786.67 EXCEEDS the DCF anchor -- projection assumes growth beyond priced-in; 4wk 637.29 Goldman Sized 95.34% x x 20% x 0.5 =. Execute at 07-08 open (L8), price is 07-06 close-ref, reprice on >1.5% gap. · news decided @ 552.05 · stop 527.04 · t1w 577.62 · t4w 637.29 · t12w 786.67 |
| 2026-07-08 | TSM | Taiwan Semiconductor (ADR) | BUY | 438.02 | FailedBreakdown_Up_20D | 76 | — | rationaleFirst of the seed. TSM is a stable-and-strengthening STRONG BUY (BUY x3 -> STRONG BUY, conviction on a FailedBreakdown_Up_20D reversal caught AT ENTRY (RSI 54.6, EntDist ~0 -- not a chase), F-Score 9 (best quality in book). News confirms: Citi raised PT to NT (Buy) on sustained AI demand and TSMC hiking leading-edge (7nm & below) prices 5-10%; earnings 07-16. NA_CCY valuation (ADR -- val blind, val_adj 0, scorecard 0), adj 76 clears the 65 floor comfortably. Sized 95.34% x x 20% buy-pct =, size_mod 1.0. Execute at the 07-08 open (L8); price is the 07-06 close-ref -- reprice if the open gaps >1.5%. Measured first tranche into the FOMC-minutes/CPI window (overlay: stage adds between rate events); ~71% kept dry. · news decided @ 451.79 · stop 438.91 · t1w 467.25 · t4w 503.31 · t12w 538.11 |
Learnings in force
What this book's own record has taught it.
Last reviewed 2026-07-18
Favor PullbackBuy_VWAP entries over breakout/FailedBreakdown chases.
`PullbackBuy_VWAP|buy` is the most reliable buy cell (US hit ~0.86); `FailedBreakdown_Up_20D|buy` chases are weak (~0.25, stop-first ~60%) and `BreakoutUp_52W` 4wk targets overshoot +55-75% and mean-revert. On ties prefer the PullbackBuy; discount breakout/FailedBreakdown chases. Hardened for ussemicon by USC-3 (this book's realized FailedBreakdown record is 3/3 stop-first).
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.
Distrust the model 1-week estimate when `calib_1w` is degenerate.
If the scorecard `calib_1w_fraction` is <=0 or wildly unstable, fall back to f=0.30 for the 1wk interpolation and flag it.
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.
(ARCHIVED for ussemicon — ADR-0014 standalone) Same-pool/cross-book shares execute ONCE.
Applied when ussemicon mirrored us (and when a sibling like ustech governed a shared QCOM share). Post-2026-07-06 ussemicon runs on its own US seed with no shared shares and no governing sibling — nothing to reconcile; own cash independently (USC-1). Retained only as the generic rule for a book that still shares a pool with us.
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.