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 | 32.26 | HOLD | — | 30.78 | n/a | n/a | 45.97 | Hold Lenovo. The position is up 10.9% and still sits above every moving average with a 12-week view of HK (+42.5%), so there is no reason to sell a winner that is still trending; fair value is neutral (FAIR_VALUE, MoS -0.65) and it is the highest-quality name in this universe (F-Score 7, quality rank). But the signal has softened: the engine flipped from HOLD to NO TRADE on 'consensus conflict + low conviction', momentum fell and the pillar read went Consolidation_Bullish -> Conflicted. That is one print, not a trend, so today is a hold on a protective stop rather than a trim. OPERATIONAL: no sell origination fires (CLAUDE.md S9) - label is NO TRADE not SELL-REDUCE, Sell_Rank blank, Exit_Warning blank. Conviction is the NoTrade artifact, not a real collapse tick, and it is moot because ADR-0012 is NOT ARMED (Gain_ATR 1.77 vs 3.00, T12_Progress 70.18 vs 90, no PROFIT_ARMED) - the 09-08 executor add lifted cost 27.4980 -> 29.0925 and structurally de-armed the position. L2 fails both legs (close 32.26 vs stop 30.78 = 4.6%/0.83 ATR cushion; T12 +42.5%). ADR-0020 n/a outside us. Valuation cannot originate and had nothing to amplify. 25.2% concentration is portfolio shape, never an origination class. STOP HELD AT 30.78 - the NO TRADE row printed no Stop_Price (Exit_Strategy 'none'), so the ratchet watermark carries; never lowered. Above the 29.0925 cost, so a stop-out banks +HK (+5.8%). RE-ARM TRIGGER for a 30% trim: a SECOND consecutive NO TRADE/Conflicted print at the 09-11 close (S9 'sustained NoTrade/Conflicted') OR a decisive close below 30.78. 12wk 45.97 marginally exceeds the 45.64 DCF anchor (+0.7%) - last ~1% of the path unfunded by valuation. · news |
| 0068 | MANYCORE TECH | 9.98 | WATCH | 9.98 | 9.56 | n/a | 12.72 (+27.5%) | 8.55 (-14.3%) | No buy in Manycore Tech. The label improved from AVOID to WATCH, but the model's own 12-week projection of HKsits 14.3% BELOW the HKclose - we do not buy a name whose forward view points down. The stock fell 10.9% today, trades below its 50-day and 100-day averages, earns no valuation support (no computable fair value, negative book earnings power, negative free cash flow) and faces a September Hong Kong IPO lock-up release wave. Watching, not buying. OPERATIONAL: HT-4 forward-view screen rejects it outright (Target_12Week 8.55 < Close 9.975) - the 12th such rejection in 14 sessions. Also sub-floor at cv 63 < 65, and WATCH is not an actionable bullish label so it never entered the buy pool. Signal is FailedBreakdown_Up_20D = the L1-DISCOUNTED chase cell (scorecard n=3, hit_1w 0.333). Valuation NA -> val_adj 0, quality gate off; but EPV -1.97 and FCF_Yield -1.15% are independently disqualifying, and Quality_Pass is False. Trend_Score decaying -> across four sessions. NOT QUEUED at any price - the block is the forward view, not the level. RE-ARM: requires Target_12Week > Close AND cv >= 65. · news |
| 0992 | Lenovo Group Limited | 32.26 | QUEUE CLOSED - filled @31.75 | 31.75 | 30.78 | n/a | n/a | n/a | Housekeeping, not a trade: closing out the 2026-09-04 queued add on Lenovo, which was already filled at HKand is now worth HK. No new money is committed. OPERATIONAL: Step 7 queue-expiry sweep. This row's expires_on is TODAY and it still read status 'queued' / execution_status 'pending_next_open' even though the Auto-Executor filled it on 2026-09-08 (txn hktech-20260908-0001, @ 31.75,. It is re-statused FILLED so it cannot be replayed a second time - exactly systemic finding #1(b) (a fill or withdrawal must re-status the originating ledger row, not merely be recorded in _state.json). The fill itself was UNAUTHORIZED: the 2026-09-07 verdict had withdrawn this GTC on the conviction leg (53 < 65) and the executor replayed it anyway. Marked to today's 32.26 the unauthorized is +HK; the of the error is positive, the governance defect is not remediated. Not reversed by trade - undoing it would be a portfolio-shape sell (CLAUDE.md S9). |
| — | AI - HK Tech | — | NO ACTION | — | — | n/a | n/a | n/a | No trades today. The Hong Kong tech book has no buyable setup - names print AVOID (the most this book has recorded), 6 print NO TRADE and 1 prints WATCH, so there was not a single bullish actionable signal to rank. The one holding, Lenovo, is up 10.9% and stays on a protective stop. Cash stays at HKor 74.8% of the book. OPERATIONAL: 0 ACQUIRE + 0 DISPOSE + 0 ANTICIPATE. Buy pool EMPTY for the first time in this window - not a floor rejection, an absence of candidates. Zero BUY - ANTICIPATE prints across 09-07..09-10, so the 4th slot is empty. One queue reconcile written (0992 09-04 GTC -> FILLED). : buys HKsells HKpowder unchanged HK; book HK = +2.53% since the 2026-07-06 HK seed. CARRIED BLOCKING: the Auto-Executor has now been silent TWO sessions (last_run_date still 2026-09-08) with its integrity defect un-remediated. Step 8a 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.