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 22 Jul 2026, 16:00:00 GMT+8

Hong Kong — End-of-Day Verdict

3 DISPOSE + 3 ACQUIRE (queued). Sells free HK (HSBC reduce 1.65M + 2513 full exit 1.18M + 7709 half-trim 50k); buys need HKif all three pullback limits fill - strongly. 0 stop breaches; monitor clean; 2513's 07-21 pending exit never filled (re-issued). HK (unchanged - all pending next open).

Recommendations

3 to acquire · 3 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
0005HSBC HDGS158.90SELL 85.46%158.90153.73157.27 (-1.0%)160.95 (+1.3%)164.28 (+3.4%)The settled close flipped HSBC to a Strong Sell - Reduce at Sell-Rank 1, conviction 70, on a reversal-from-overbought signal with RSI at 71 after a strong run. That is a signal-based reason to bank the bulk of a bank position that has stretched, so the model's 85% de-risk stands. Valuation is currency-blind here (NA_CCY) so it neither originates nor amplifies. Operational nuance: this is a REVERSAL sell, not a profit-take, so the trim-discipline gates (G1-G3) do not apply - they guard momentum breakouts, and ReversalDown_Overbought is the valid-exit contrast case. Honest caveat: the model's own forward view is only mildly positive (12wk +3.4%), so this is a risk-management reduce on the reversal + RSI extension, not a dead-view exit; the specific ReversalDown cell is report-only (n=1) but the broader HK SELL family is reliable (HK-3). L4: acted on the settled close, not an intraday whipsaw. L8: executes MARKET at the next open 07-23; reprice if the open gaps >~1.5%. Reduce (HSBC 400-lot), leaving 1,600 on a 153.73 stop. · news
2513Knowledge Atlas1175.00SELL ALL1175.001213.53 (+3.3%)1164.04 (-0.9%)1236.47 (+5.2%)Re-issue of the standing full exit: yesterday's pending exit at ~1219 never filled (no trade recorded, still held), and the name has faded to 1,175. It prints a bearish FailedBreakout_Down_20D for a third straight session on the book's actionable down-trend sell cell (bias +2), sits ~28% underwater against a 1,635 cost, and has zero fair-value coverage to lean on - there is no floor and no forward case (4wk -0.9%, and the +5.2% 12wk sits inside a -18%/+72% band, i.e. noise). Operational nuance: L3/HK-5 - this de-risk has been flagged for six sessions; execute it promptly at MARKET rather than re-recommend indefinitely. The label softened from Strong Sell-Reduce to AVOID but the exit thesis (persistent bearish, no valuation floor, deeply underwater) is unchanged. Extreme variance: >15% intraday moves on four of the last five sessions - stay MARKET and expect slippage. L7: check hktech before executing - execute once. Frees HKthe book's largest single risk line. · news
7709XL2CSOP HYNIX 2x55.76SELL 50%55.7644.0058.46 (+4.8%)55.22 (-1.0%)56.45 (+1.2%)De-risk a deeply underwater (-51% vs 114.56 cost) 2x leveraged ETF that has printed a bearish breakdown-family signal for a third straight session, conviction 61, with a dead forward view (4wk -1.0%, 12wk +1.2%) - the underlying's SK Hynix catalyst has now passed and the ETF fell into it. Operational nuance: the leveraged-ETF doctrine says trim 50% on a low-conviction (cv<70) reversal/breakdown rather than full-dump at an oversold low (RSI 38.7) - the 7747 full-exit that missed a +5.8% bounce is the lesson this respects. So bank half (, ETF 100-lot) and hold 900 on the manual 44.00 trail (unbreached; the scan's 84.35 stop is above spot and invalid for a long). Escalates to full exit only on conv>=70 + price below the manual trail. L8: MARKET at the 07-23 open. Frees HK. · news
3690Meituan83.65BUY — QUEUED @77.2277.2272.6183.76 (+8.5%)82.47 (+6.8%)82.66 (+7.0%)Top-ranked acquire: a fresh PullbackBuy_VWAP BUY at conviction 73 - the book's single most reliable buy cell (HK-2, bias +3) and the one setup the learnings say to prefer over any breakout/momentum chase. Adjusted conviction 76 clears the floor comfortably. Valuation is blind (NA) so it neither helps nor hurts and the quality gate is off. Operational nuance: the model entry 77.22 is ~8% below the 83.65 close, so this is a no-chase GTC buy-limit that fills only on a pullback to VWAP - queued, not bought at the close (HK-4: no-chase stays strict). Size (100-lot) ~HK at the book's 20% new-buy weight. Expires in 3 sessions (07-27) or on conviction falling below 65; stop 72.61. · news
0857PetroChina H10.05BUY — QUEUED @9.179.178.7610.03 (+9.4%)10.10 (+10.1%)10.26 (+11.9%)Highest adjusted conviction in the book (78) and a re-arm of the queue withdrawn on 07-21: PetroChina prints PullbackBuy_VWAP BUY again at conviction 70, and it is the book's best-qualified name on valuation - UNDERVALUED with Quality_Pass True (MoS +0.236), which adds +5 and unlocks up to 1.25x sizing. Operational nuance: the model entry 9.17 is ~9% below the 10.05 close, a genuine no-chase pullback limit repriced to the live row (not the withdrawn 9.10). Sized up 1.25x on the valuation edge = (2,000-lot) ~HK. HK-2 anchor confirmed. Expires 07-27 or below cv 65; stop 8.76. · news
0992Lenovo24.10BUY — QUEUED @22.9822.9821.0824.03 (+4.6%)25.27 (+10.0%)26.40 (+14.9%)Third acquire: a PullbackBuy_VWAP BUY at conviction 66, adjusted 68 - above the floor and the third of three fresh prints on the book's reliable buy cell (HK-2, bias +3). FAIR_VALUE with Quality_Pass True, a small -1 valuation trim (MoS -0.25) but the DCF (43.5) sits well above spot so the quality gate passes and sizing stays 1.0x. Operational nuance: entry 22.98 is ~5% below the 24.10 close - the tightest of the three pullback limits, so the most likely to fill. (2,000-lot) ~HK. The 12wk estimate is the strongest of the three buys at +14.9%. Expires 07-27 or below cv 65; stop 21.08. · news
0939CCB H8.55HOLD8.13n/a8.97n/aHeld and strengthening - conviction climbed across the session on a FailedBreakdown_Up recovery, with the close (8.55) back above a rising 8.13 stop. No exit case and no fresh-add case (still below the add threshold), so hold and let the stop trail up. The book's largest share line by count; nothing to do today. · news
2888StanChart223.60HOLD218.00n/a234.76n/aHeld and firming - conviction into the close (223.60) on a FailedBreakdown_Up, just below the add threshold. No exit case; hold on the 218 trail into H1 results at month-end. Valuation is currency-blind (NA_CCY). Nothing to act on today. · news
0522ASMPT169.10HOLD156.50n/a178.15200.98Catalyst-aware hold into its own earnings, now estimated for the very next session. No-trade for a fourth straight snapshot, ~-4.7% vs cost, but the 12wk view (200.98, ~+19%) is the most positive in the book and valuation can never originate a sell. Operational nuance: no pre-trim into a binary one session away; the model's stance is add-after-print, not before. Hold on the 156.50 trail (just under recent session lows); only a confirmed lower high or a decisive stop break changes this before the numbers land. · news
1347Hua Hong161.50HOLD145.00n/an/a184.55Held around break-even (~+0.9% vs cost) on repeated no-trade prints - no entry case and no exit case. The spurious PROFIT_ARMED flag persists (systemic finding #2: T12_Progress is Close/Target_12Week, not progress from cost), so the profit-take path stays suppressed - the real gain is a fraction of an ATR, not a harvestable winner. Hold on the 145.00 trail; a fresh breakdown-family SELL print or the 12wk estimate turning negative means dispose immediately. · news
3750CATL609.50HOLD488.81n/an/an/aCatalyst-aware hold into H1 results in ~2 sessions. No-trade, ~-1.9% vs cost, no signal either way. FAIR_VALUE on valuation (cannot originate a sell). Hold on the 488.81 stop through the print; no pre-trim into the binary. · news

Outcome & track record

Accuracy at the time of this record.

112 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 →
-0.283 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.
21 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.