This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.

HK End-of-Day Verdict 3 Sept 2026, 16:00:00 GMT+8

HK Semiconductors — End-of-Day Verdict

EOD 2026-09-03 (hksemicon)

Recommendations

0 to acquire · 0 to dispose

TickerNameCloseActionAction 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
0992Lenovo Group Limited30.54HOLD30.5429.68n/a33.56 (+9.9%)43.52 (+42.5%)Lenovo is holding a constructive base and we stay long the full. The signal is a stable FailedBreakdown_Up_20D at conviction 63, the close of 30.54 sits well clear of our 29.68 stop, and the 12-week projection of 43.52 still implies roughly +42% upside, so there is no sell basis on any of the three origination classes. Valuation is neutral at FAIR_VALUE (MoS -0.62), which neither helps nor hurts, and no dated news landed in the window. Operational nuance: conviction slipped only, far short of the 20-point collapse that would be an ADR-0012 fire tick, and the position is not armed (Gain_ATR 0.10, T12_Progress 70.17). STOP HELD at 29.68: the model drifted Stop_Price UP to 30.09 but that is just 0.30 ATR below the close, inside one day's chop on a 4.9%-ATR name, and a 2.0x ATR chandelier from the 30.60 peak close computes 27.58 (lower), so the monotonic non-decreasing rule per HS-3a keeps 29.68. ADR-0020 not fired, Peak Stop 29.68 = live, 0% erosion. Scorecard cell FailedBreakdown_Up_20D|buy +4 is held report-only per learnings section C (L1/HS-5 chase discount) and was not applied.
2382Sunny Optical Technology70.60HOLD70.6066.19n/a76.86 (+8.9%)85.43 (+21.0%)We hold all of Sunny Optical and raise the trailing stop rather than trim. Conviction is recovering strongly on a FailedBreakdown_Up_20D, the 12-week projection of 85.43 is about +21% above the 70.60 close, and valuation reads UNDERVALUED with quality passing, so nothing originates a sell even though the position is 2.55% underwater on a one-day-old 72.45 cost. The Apple keynote on 09-10 is a direct optics read-through five sessions out, and the catalyst-aware hold says trail through the event rather than pre-trim. Operational nuance: STOP RAISED 64.73 -> 66.19, the same 2.0x ATR chandelier convention anchored on the 72.45 entry, recomputed with today's compressed ATR of 3.129; it is monotonic up per HS-3a and sits 1.41 ATR below the close, outside daily noise. The model's own Stop_Price of 69.66 is REJECTED at 0.30 ATR below the close, the same too-tight defect rejected on 09-02 (systemic finding 2). ADR-0012 NOT armed: Gain_ATR is -0.59 and T12_Progress reads 82.64 only because that field is Close/Target_12Week rather than progress from cost (systemic finding 1) - a trim here would be exactly the HS-2 target-exhaustion-dressed-as-profit-take error. ADR-0020 not fired, Peak Stop re-based up to 66.19 = live, 0% erosion.
AI - HK SemiconductorsNO ACTIONn/an/an/aNo trades today: the book stays fully as-is with HK (69.2%) in cash. Across all twelve names in the universe not one printed a buy trigger - six are outright bearish AVOID signals, four are stuck on consensus-conflict NO TRADE, and the two we own are simply holds - so there was nothing to buy at any price, and neither holding gave a stop break, sell label or profit-take trigger to sell into. Sitting on cash through the 09-04 index rebalance, 09-07 US payrolls and the 09-10 Apple keynote is a deliberate posture rather than an oversight. Operational nuance: buy pool and anticipate slot both empty. Loudest near-miss was 1347 Hua Hong joining the HSI on 09-04 - rejected on four grounds (AVOID cv58, 12wk -43% vs close, Quality_Pass False firing the hard gate on a no-DCF UNDERVALUED mark, and HS-5/L6 barring the inclusion-gap chase after this book watched a prior 1347 +8.4% gap round-trip 201.80 -> 185.50). 6869 YOFC touched WATCH cv55 on 09-02 then collapsed to NO TRADE cv0, and would have been blocked anyway by the quality gate at MoS -12.03. Counterfactual record supports the restraint: no_chase_queue n=8, median 4wk +23.19% saved.

Outcome & track record

Accuracy at the time of this record.

18 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 →
-0.138 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.

This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.