Recommendations
2 to acquire · 2 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 |
|---|---|---|---|---|---|---|---|---|---|
| 2382 | Sunny Optical Technology (Group) | 71.90 | BUY — QUEUED @72.21 | 72.21 | 64.49 | 72.73 (+0.7%) | 74.50 (+3.2%) | 75.11 (+4.0%) | Sunny Optical broke out of its 20-day range on the strongest conviction print in the book (cv85, Buy_Rank 1) after a genuine two-day build, and the intrinsic work agrees the price is not stretched - UNDERVALUED on a DCF of 83.30 against a 71.90 close, with the 12-week projection of 75.11 still sitting comfortably below that anchor. We buy the follow-through rather than the print: the order is a buy-stop at the model entry of 72.21, above today's close, so it only fills if the breakout actually extends. OPERATIONAL: RSI 74.3 and a Broadcom read-through event on 09-03 (two-sided +/-6%) plus the Apple keynote on 09-10 (+/-5%) are why this is a stop rather than a market buy, and why the size modulator is held at 1.0x instead of the 1.25x that UNDERVALUED + Quality_Pass would allow. HK-1 chase discipline satisfied by the second/third consecutive bullish print. Model Stop_Price 62.47 (-13.1%) REJECTED under HK-12 as too wide to bind; working stop is the 2.0x ATR chandelier at 64.49, risk HKNotional capped from the model's HKto HKto fit two acquires inside HKof. L7 CROSS-BOOK: hktech (4,200 @71.65) and hksemicon (3,800 @72.45) buy the same name today - / ~HKacross the family; hk is the governing parent, execute once. Target_Price 91.69 is the risk objective, NOT a forecast - Est_4W is 74.50. |
| 0992 | Lenovo Group Limited | 30.58 | BUY | 30.58 | 29.68 | 30.94 (+1.2%) | 32.03 (+4.7%) | 33.07 (+8.1%) | Lenovo is the most-confirmed setup on the book - five consecutive daily BUY prints with conviction strengthening, RSI a temperate 58.9, and the price sitting exactly on the model entry rather than extended above it. Fair value is neutral (FAIR_VALUE, DCF 45.62 vs a 30.58 close) so this is a technical entry with no valuation objection, and the model's own 12-week projection of 33.07 offers +8.1%, the best forward view of any acquire candidate today. OPERATIONAL: adjusted conviction lands at EXACTLY the 65 floor (70 raw, -3 valuation, -2 scorecard on FailedBreakdown_Up_20D|buy which grades hit_1w.417 over n=14) - taken because the five confirming prints and sub-70 RSI are precisely the confirmation HK-1 demands before acting on this chase-grade family, and sized smallest in the book accordingly. Stop 29.68 (-2.9%) accepted as printed - sane width, unlike the rest of the book. Risk HKL7 CROSS-BOOK: the family already holds (hktech 10,000 @27.50, hksemicon 14,000 @30.39); neither sub-book proposes 0992 today so there is no double-, but the family aggregate becomes. |
| 0939 | CCB - H SHS | 9.60 | SELL 30% | 9.60 | 9.16 | 9.64 (+0.4%) | 9.78 (+1.9%) | 9.75 (+1.6%) | CCB has run into a genuinely overbought condition - RSI 71.6 - and the signal engine flipped it to a Sell_Rank #1 REDUCE that has now held for three consecutive prints. More telling than the label, the model's own forward view has gone flat: it projects 9.75 twelve weeks out against a 9.60 close, so the remaining upside in the bank re-rating is roughly nil on this horizon. We bank a third of a HKposition into the strength and keep the rest for the re-rating. Valuation is silent here (NA_CCY, currency-mismatched statements), so this rests purely on the signal. OPERATIONAL: engine was 85.46%; deliberately cut TWO tiers to 30% because (a) the originating cell ReversalDown_Overbought|sell is report-only at n=2 with hit_1w.00 - no evidence base, (b) the label is one session old and flipped straight out of a STRONG BUY - ADD cv90 print on 08-31, and (c) all three model estimates are mildly POSITIVE (+0.4/+1.9/+1.6%). Retains. Avg_Cost is ABSENT for 0939 in hkportfolio.json so realised cannot be computed - systemic finding #2. L7 CROSS-BOOK: hkfin trims 0939 the same session - separate share ledgers, same name, execute both or neither at family level. L8 gap-guard is asymmetric on a de-risk: sell on a gap up (better fill) or a gap down (more urgent), reprice to the actual open. |
| 3750 | CATL | 592.00 | SELL 33.3% | 592.00 | 556.49 | 591.66 (-0.1%) | 594.55 (+0.4%) | 606.39 (+2.4%) | CATL has broken its stop and the news flow explains why: a Sinopec unit placed roughly US of stock and the third-largest shareholder is offloading, while US House Republicans press Commerce to tighten restrictions on Chinese grid components - with battery inverters named as a security concern for the first time. Supply overhang plus policy risk on a position already 21% underwater is a risk-control exit, not a view on lithium. We take one board lot off and keep two-thirds, because the offsetting case is real: the placements are off-market and Changjiang argues 30% lithium demand growth in 2026 makes this an accumulation window. OPERATIONAL: close 592.00 breaks the recorded 600.00 stop by -1.33%; signal is a stable STRONG SELL - REDUCE across four intraday and two daily prints on RallySell_BearTrend, the book's best actionable exit cell (bias +3, n=20, hit_1w.632). Adjusted conviction 58 is BELOW the 65 floor - carried on the CLAUDE.md section 9 stable-REDUCE origination plus the live stop break, with the sub-floor stated openly and size cut to the single minimum board lot (, board lot 100). Realised -HKagainst cost 751.80. Valuation MoS_FV -1.045 is not <= -3 so NO tier bump. Model estimates are flat-to-up (+0.4%/4wk, +2.4%/12wk) which is exactly why this is a minimum trim and not an exit. · news |
Outcome & track record
Accuracy at the time of this record.
188 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 →
-0.261 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.
49 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.