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 19 Aug 2026, 16:00:00 GMT+8

Hong Kong — End-of-Day Verdict

0 ACQUIRE + 1 DISPOSE + 0 ANTICIPATE

Recommendations

0 to acquire · 1 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
3750CATL628.50SELL 85.46%628.50614.95625.19 (-0.5%)622.63 (-0.9%)610.15 (-2.9%)CATL has flipped to a sustained sell signal at the settled close and we are banking two thirds of a losing position. The stock printed STRONG SELL - REDUCE on four consecutive prints into the close on the failed-breakout-down pattern, which is this book's best-sampled and most reliable exit signal, and the model's own forward view is now below today's price at every horizon it publishes - one week, four weeks and twelve weeks. The shares closed at 628.50, which is the exact floor of the HK-651.80 range at which the company placed new stock in April, so the placement overhang caps the near-term recovery too. We keep for the battery cycle behind a tightened risk line. OPERATIONAL: This card is the pre-committed 08-11 WATCH line firing exactly as written - re-arm was 'close < 625.50 OR a fresh SELL-REDUCE', and the fresh SELL-REDUCE leg triggered; it is therefore a NEW origination, not a re-issue, so HK-11 does not bind. Adjusted 38 = 37 + 0 val (sell-side is amplify-only; MoS -1.16 does not reach the -3 tier-bump bar) + 1 (FailedBreakout_Down_20D|sell n=61 ACTIONABLE, live bias +1 per HK-7). The 65 floor governs buys, not a stable-label exit - book precedent (08-18 1347 @adj 58, 08-17 1347 @adj 1). PROFIT_ARMED on this row is SPURIOUS and NOT the origination: the position is -16.4% underwater vs 751.80 with Gain_ATR -5.26; the arm came from T12_Progress 103.01 alone, the same false-arm pattern caught on 1347 on 08-13. Sizing: 300 x 85.46% = 256.4 -> rounded DOWN to the 100-share board lot = 200, keeping this a trim rather than a full exit. Residual hold the 614.95 trail (0.58xATR below the close, tighter than the model's 581.62). HK-12 pass - not degenerate. Proceeds HK.
0700Tencent447.20HOLD447.20430.00443.27 (-0.9%)439.46 (-1.7%)451.26 (+0.9%)Tencent prints a sell signal today but stays untouched, because the exit was already retired to a single line in the sand at 430 and the shares closed at 447.20, comfortably above it. The trend is genuinely weak - the stock trades below its 200-day average and the model sees it going essentially nowhere over twelve weeks - but the holding is only, about 1% of the book, and we do not re-open a retired exit thread on a line that has not been touched. OPERATIONAL: STRONG SELL - REDUCE cv73 held across all five intraday prints, Sell_Rank 2, adjusted 74 = 73 + 1 (FailedBreakout_Down_20D|sell, live bias). Not acted on: exit RETIRED 08-17 to the 430.00 WATCH line, which is also the news overlay's own standing override for this ticker - close 447.20 is 3.9% above it, NOT re-armed. HK-11 binds. Re-arm: decisive settled close < 430.00 OR a fresh SELL-REDUCE after a re-arm. DATA ERROR carried forward - Avg_Cost 1.00 in input/hkportfolio.json produces Unreal_PnL_Pct +44,620% and Gain_ATR 37.13, which is what falsely sets PROFIT_ARMED on this row; the arm is meaningless until Cost is fixed. Systemic finding, unchanged from 08-18.
0522ASMPT158.50HOLD158.50155.02163.99 (+3.5%)178.81 (+12.8%)178.61 (+12.7%)ASMPT broke through its risk line today, closing at 158.50 against a 163.69 stop, but we are holding rather than selling. The rule this book uses for a stop break is that it only becomes an exit when the forward view has also turned negative - and here it has not: the model still sees 178.81 in four weeks and 178.61 in twelve, roughly 13% above today's price, the best forward view of any holding in the book. So we re-base the risk line lower and give the position room. OPERATIONAL: Stop breach -3.17% confirmed by the monitor. L2 requires stop-break AND negative 12wk; the 12wk leg does not fire (+12.7%), so no exit. Label HOLD cv42 across five prints, no Exit_Warning, NO ADR-0012 arm (T12_Progress 88.74 < 90, Gain_ATR -3.26 negative). Trail RE-BASED 163.69 -> 155.02 (the model's own Stop_Price; 0.30xATR below the close - deliberately tight, one decisive down session resolves it). Quality gate F (F_Score 6, FCF_Yield -0.34%) and MoS -4.71 GROWTH_PRICED_IN, but neither originates a sell (CLAUDE.md section 9) and the position is held, so the gate does not bind. Position -19.28% vs 196.35.
0939CCB - H9.05HOLD9.058.689.03 (-0.2%)9.14 (+1.0%)9.06 (+0.1%)China Construction Bank remains the steadiest holding in the book - a buy-side signal held at full conviction across every read today, and the largest position by share count. Nothing to do but nudge the protective stop up. OPERATIONAL: HOLD cv73 x4 stable. PROFIT_ARMED (T12_Progress 99.89) but NO fire tick: (a) no other Exit_Warning code, (b) conviction flat, (c) 12wk 9.06 > 1.02 x 9.05 is marginal but the row carries no second warning, (d) label HOLD not NO TRADE. Armed-but-not-fired -> mandatory trail-raise, not a trim. Trail RAISED 8.66 -> 8.68 (the model's own Stop_Price, 1.99xATR below the close). Re-arm from 08-11 retirement unchanged: fresh SELL-REDUCE or decisive close < 8.46.
0981SMIC73.00HOLD73.0069.6373.16 (+0.2%)75.92 (+4.0%)76.31 (+4.5%)SMIC closed up on the day and, notably, held firm while Hua Hong fell - the two Chinese foundries are not trading as one basket. The engine has no actionable view, and the model still sees mid-single-digit upside over four and twelve weeks, so we hold behind an unchanged stop. OPERATIONAL: NO TRADE cv0 on the last two prints, PROFIT_ARMED (T12_Progress 95.66). Tick (d) needs 2 consecutive NO TRADE and technically has them - but Avg_Cost is BLANK, so ADR-0012's no-guess rule blocks any profit-take, same as 0823. Thread also RETIRED 08-14. Trail HELD 69.63 (0.77xATR below the close; tighter than a 2.0xATR chandelier at 64.27). Est_4W +4.0%, Est_12W +4.5%. MoS -5.31 GROWTH_PRICED_IN does not originate a sell (CLAUDE.md section 9).

Outcome & track record

Accuracy at the time of this record.

176 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 →
-0.249 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.