Recommendations
1 to acquire · 1 to dispose
| Ticker | Name | Close | Action | Action 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 |
|---|---|---|---|---|---|---|---|---|---|
| TSM | Taiwan Semiconductor Manufacturing | 399.09 | SELL ALL | — | 406.37⚠ | n/a | n/a | 363.69 | Exit the last 103 TSMC shares in full at the open. The stock has now closed decisively below our 406.37 risk level for a third straight session (399.09, -1.79% through the stop), the model's own 12-week view has fallen to 363.69 - roughly 9% BELOW today's price - and the signal engine has printed no tradeable setup three sessions running. TSMC's business is not the problem: Q2 profit hit a record (+77% y/y) and full-year growth guidance was raised above 40%. The market is repricing the cost of getting there - a -64bn capex guide, 14% above the prior range, and the margin compression that comes with it - and has taken the shares down about 15% in a month. That is a valuation debate we can re-enter later; it is not a swing-trade setup, and our stop has done its job. Booking the loss at roughly -8.9% (~-) frees of capital. OPERATIONAL: this is NOT a new decision - the 2026-07-25 verdict armed this same exit for the 07-27 open and it was never filled (auto-executor idle since 07-23), costing a further -1.1%. UT-11/UT-3/L3: an unfilled armed exit is an incomplete exit - re-fire at market-on-open, do not re-recommend a third time. PROFIT_ARMED on the row is the UT-9 arm-on-loss mislabel (Gain_ATR -2.46, position underwater) and is inert - this is loss-prevention, not a profit-take. Valuation amplify suppressed by G2 (DCF anchor vs price, ~9x broken). · news |
| AAPL | Apple | 336.91 | BUY — QUEUED @315.04 | 315.04 | 306.90 | 318.68 (+1.2%) | 327.16 (+3.8%) | 405.45 (+28.7%) | Place a good-til-cancelled buy limit at 315.04 for - we want Apple, but not at Monday's record high. The setup is the strongest this book has produced: a fifth consecutive strengthening session on the pullback-to-VWAP pattern that is our most reliable entry, with conviction at a maximum 100 and the #1 buy rank. The story behind it is the mirror image of what is hurting the rest of the book: Apple closed at an all-time high Monday, took the most-valuable-company crown back from Nvidia at a 94tn market cap, and is up 22% year-to-date precisely because it has NOT spent heavily on AI infrastructure - the same restraint the market is punishing TSMC and Dell for lacking. The catch is price: at 336.91 the stock sits about 6.5% above the model's entry, and on intrinsic value it is already expensive (margin of safety -3.16), so we halve the position size and refuse to chase - the order fills only on a genuine pullback to 315.04 or not at all. Risk is capped at the 306.90 stop, about one ATR below entry (on the full). OPERATIONAL: replaces (does not carry) the stale 07-25 GTC @307.05 - repriced to the live anchor Entry_Price 315.04 per the provenance rule, re-derived on current. expires_on deliberately set to 2026-07-29 rather than the mechanical anchor+3 (07-30): Apple reports FY26 Q3 on 07-30 AMC (Event Horizon band +/-5%, stance add-after-print) and FOMC lands 07-29 - leaving a live limit into an earnings print is the L6 gap-fill failure mode, so the order stands down before the event and the name is re-armed after the print. Size 0.5x on MoS <= -3; quality gate PASS; scorecard bias 0 (PullbackBuy_VWAP|buy cell n=2, report-only - L1 caution applies). 12wk 405.45 exceeds the DCF anchor - projection assumes growth beyond priced-in. · news |
| DELL | Dell Technologies Inc | 426.91 | HOLD | — | 423.45 | n/a | 488.13 | 608.35 | Keep all and keep the 423.45 stop. Dell fell 2.4% with the wider AI-hardware complex on Monday but nothing in the position's own case broke: the signal has held the same bullish reading at conviction 61 for three straight sessions, there is no exit warning on the row, and the model's 12-week view of 608.35 is still 42% above the current 426.91. The business news continues to support the thesis - 4bn of AI server orders in the last quarter, a 3bn backlog, FY27 revenue guided to -169bn (+47%) with ~$60bn of that from AI servers, and the stock up 213% year-to-date. The known bear case is unchanged and already in the price: AI server gross margins near 18% and the resulting UBS downgrade on 07-16. We are down 2.84% against our cost, which is noise on a 30-point ATR. OPERATIONAL: the model's own trailing stop has eased to 411.60 as the pullback ran; hold the higher recorded 423.45 (never ratchet down). Cushion is thin - the close sits only 0.81% above the stop into a two-sided FOMC on 07-29 (Event Horizon band +/-3%, stance tighten-trail). Read the asymmetry correctly: a close below 423.45 is a RE-EVALUATE/monitor trigger, NOT an automatic exit - L2 requires a stop-break AND a negative 12wk, and the 12wk is +42.5% positive, so G3 (do not cut a winner with a live forward view) governs. UT-10: no ADD - the two prior pullback ADDs (07-10 @411.73, 07-15 @419.03) both graded stop_hit; do not pyramid into a 30.6-ATR reset. 12wk 608.35 exceeds the DCF anchor. · news |
| SNOW | Snowflake Inc. | 272.92 | NO ACTION | 272.92 | 265.52 | 280.18 (+2.7%) | 297.12 (+8.9%) | 388.91 (+42.5%) | Not taken, and worth recording why - this was the cleanest entry available. Snowflake prints a buy at conviction 76 and the #2 buy rank with the close sitting exactly on the model entry of 272.92, so there is no chasing involved, and the 12-week view of 388.91 is 42% higher. It is blocked by our quality screen: the F-Score/free-cash-flow test fails, and the earnings-power valuation is actually negative at -a share, meaning the entire intrinsic case rests on a discounted-cash-flow projection of future growth rather than anything the business earns today. The news tape does not argue the screen is wrong - product revenue did reaccelerate to 34% growth and BofA named it a Q3 top idea, but the shares trade near 18x sales and insiders have been selling steadily (director Frank Slootman sold for on 07-20/21, an EVP sold on 07-14). We stand aside. OPERATIONAL: hard quality gate blocks NEW entries only (would not block an ADD to a held name); adjusted conviction 69 cleared the 65 floor before the gate. L1 also discounts the FailedBreakdown_Up_20D family it flipped into (own-book cell n=1, stop-first 1.0). Ledgered direction:none/optional:true so the counterfactual grader prices what the gate costs - fourth consecutive session this pattern recurs (see 07-11 Q4 measurement gap). · news |
Outcome & track record
Accuracy at the time of this record.
9 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 →
0.134 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.
24 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.