Recommendations
0 to acquire · 0 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 |
|---|---|---|---|---|---|---|---|---|---|
| 0992 | Lenovo Group Limited | 31.30 | HOLD | 31.42 | 30.78 | 31.94 (+1.7%) | 33.44 (+6.4%) | 44.60 (+41.9%) | Hold Lenovo. The one held position printed a constructive PullbackBuy_VWAP label at conviction 62 after a single-session abstention, and yesterday's pre-committed exit trigger - a second consecutive NO TRADE print - was tested at today's close and did not fire. Valuation supports the hold rather than opposing it: the stock is the highest-quality name in the book on the freshly regenerated file (quality rank, Bargain flagged) and sits at fair value, while the 12-week projection of 44.60 lands just below the 45.64 DCF anchor - so the forward view does not require growth beyond what is already priced in, and an independent 1-year analyst target of 44.72 corroborates it. The position is up 7.6% and the trailing stop already sits above cost, so the downside is a locked-in gain. OPERATIONAL: stop HELD at 30.78 - the anchor row printed a fresh 29.22 stop (entry 31.42 less 1.25xATR 1.762) and the ratchet is NEVER lowered; Peak Stop stays 30.78. Cushion is now thin at 1.66% / 0.30 ATR below the 31.30 close with ATR at 5.6% of price, so an ordinary session can reach it - accepted deliberately: a stop-out banks +HK (+5.8%) rather than round-tripping the gain through the 09-14 US CPI / 09-17 FOMC window. ADR-0012 NOT ARMED (Gain_ATR 1.25 vs 3.00, T12_Progress 70.18 vs 90, Exit_Warning blank) and Gain_ATR is falling 1.77 -> 1.25, moving away from arming; the 09-08 executor add lifting cost 27.4980 -> 29.0925 continues to de-arm it. No trail-raise (that applies only to armed-but-not-fired). L2 fails both legs. ADR-0020 n/a outside us and vacuous (0% erosion, never cut). The sustained-abstention trigger is RETIRED, not carried - a fresh case must start from a new first print. The 24.6% concentration is portfolio shape, never an origination class (CLAUDE.md S9). Estimates use f=0.30 (L5 mandatory; calib_1w -0.183, negative a 4th session). · news |
| 6869 | YOFC | 183.50 | WATCH | 164.89 | 145.63 | 170.19 (+3.2%) | 182.55 (+10.7%) | 216.13 (+31.1%) | Not buying YOFC - refused for a fifth straight session, and this time four separate screens reject it. The decisive one is quality: the company fails the profitability and cash-flow screen outright, which blocks any new position on its own. On top of that the shares trade at roughly fourteen times our estimate of intrinsic worth with no computable cash-flow valuation at all, the model's own four-week target of 182.55 sits BELOW today's 183.50 close (the objective is behind the price), and the entry the signal actually wants is 11.3% lower than where the stock trades - buying here is chasing. The stock has round-tripped since we first refused it at 183.60 on 09-09, so the withhold is now roughly flat on price rather than the 3.4% saver it looked like yesterday; that does not change the decision. A trade refused on a failed quality gate and a floored valuation penalty stays refused whether or not the price recovers. OPERATIONAL: Quality_Pass False + -> Step 6 hard gate BLOCKS the new entry (dispositive alone); F-Score 6, Quality_Rank 13.5/21, Bargain False. MoS_FV -13.01 -> val_adj = clamp(round(-65.05), -10, 0) = -10 (the floor) -> adj conviction 71 - 10 + 0 = 61 < 65 sub-floor; size modulator would have been 0.5x had it survived. FV_per_share 13.10 vs 183.50 = 14.0x the intrinsic anchor, DCF not computable. HT-4 forward-view fails the 4wk leg (Target_Price 182.55 < Close 183.50 = -0.52%) even though T12 216.13 = +17.8% - HT-4 requires BOTH. Same R:R failure as 2382: reward 17.66 = 1.146 ATR vs risk 19.26 = 1.25 ATR = 0.917. Entry_Distance_ATR -1.21 = the steepest chase on the board. Withheld 09-03, 09-04, 09-09, 09-10, 09-11. NOT QUEUED. Re-arm requires the quality gate to pass first - nothing downstream can unblock it. · news |
| 0068 | MANYCORE TECH | 9.10 | WATCH | 9.10 | 8.70 | 9.90 (+8.8%) | 11.76 (+29.2%) | 6.92 (-24.0%) | Not buying Manycore. The label improved to WATCH but the model's own twelve-week projection points DOWN 23.9% from here - worse than yesterday's -14.3% - so the forward view contradicts the bounce signal. The fundamentals give no cover either: earnings power is negative, which is precisely why no intrinsic value can be computed for it, and the company fails the quality screen. The stock also sits inside September's record Hong Kong IPO lock-up release wave, an overhang of supply the technical signal does not price. This is the classic falling-knife shape - an oversold-bounce label on a name in a confirmed downtrend. OPERATIONAL: WATCH is not a bullish actionable label so it is not in the buy pool; sub-floor at 63 < 65 regardless. HT-4 forward-view screen rejects it - this is the 13th HT-4 rejection in 15 sessions, the screen continuing to do the bulk of this book's protective work. Val_Signal NA -> val_adj 0 and the quality gate is OFF (NA rows pass Step 6), but that is a SILENCE, not an endorsement: Quality_Pass False, EPV_per_share -1.97 (negative earnings power = why the row is NA), Quality_Rank 9/21, Bargain False. Signal family FailedBreakdown_Up_20D is the chase family L1 explicitly discounts; this book's cell is n=3 hit_1w 0.333, report-only, no bias applied. NOT QUEUED - the block is the forward view, not the price. RE-ARM requires Target_12Week > Close AND cv >= 65, NOT a cheaper price. · news |
| — | AI - HK Tech | — | NO ACTION | — | — | n/a | n/a | n/a | NO ACTION: 0 ACQUIRE + 0 DISPOSE + 0 ANTICIPATE, plus one queue reconcile (2382 01-Sep GTC expired unfilled). The buy pool had two members for the first time in two weeks - Sunny Optical and YOFC, both on the pullback-buy signal this book rates most highly - and both were refused: YOFC on a failed quality screen and a valuation fourteen times intrinsic worth, Sunny Optical purely because the trade's stop is wider than its target, making the reward-to-risk 0.92. Lenovo is held unchanged at; yesterday's pre-committed exit trigger was tested at the close and did not fire. The book stays 75.4% in cash. OPERATIONAL: buys need HKsells free HKdry unchanged at HKDistribution 11 AVOID / 4 NO TRADE / 3 WATCH / 2 BUY / 1 HOLD (AVOID count fell. Valuation layer REGENERATED today (was 8.0 days stale, FetchedAt now 2026-09-11 17:17:36). Zero BUY - ANTICIPATE prints across six snapshots, 4th slot empty an 8th session. THREE ESCALATIONS: (1) the Auto-Executor has been silent three consecutive sessions (last_run_date 2026-09-08) and today produced its first priced consequence; (2) 20 ledger rows still carry open/queued status because the ADR-0012 sweep appends rather than re-statuses - a resuming executor would read 20 live instructions; (3) STRUCTURAL - every PullbackBuy_VWAP print in hktech arrives at R:R 0.917 by config geometry (1.25 ATR stop vs 1.145 ATR target), so this book is permanently unable to act on its own most reliable signal family while the no_chase_queue counterfactual sits at n=7 / +5.07% 1wk / +15.04% 4wk cost and the book has been ~75% cash for twelve sessions; HT-3's champion/challenger shadow test appears not to be running two months on. Step 8a (learnings review) NOT run - skipped by run directive. |
Outcome & track record
Accuracy at the time of this record.
13 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 →
-0.183 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.
50 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.