Hong Kong — End-of-Day Verdict
3 DISPOSE (CCB -50%->25,700; 7709 -72%->500; Link -36%->1,800) free ~HK; 0 ACQUIRE (buy pool = 1 falling knife, rejected). HSBC trail-raise 148.3 + Tencent both fired-but-slot-capped near-misses. Post-cash ~HK.
Recommendations
0 to acquire · 3 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 |
|---|---|---|---|---|---|---|---|---|---|
| 0939 | CCB (H) | 8.28 | SELL 50% | 8.28 | 7.40 | n/a | n/a | 6.71 (-19.0%) | Re-issue the biggest-position de-risk: CCB is the dominant HK line with a persistently negative model 12wk (6.71, -19%). The settled close bounced off the RSI-27 washout to NoTrade cv0 (RSI, but it is PROFIT_ARMED and fired a profit-take (tick c dead/negative 12wk 6.71<1.02x8.28, tick d 3x consecutive NoTrade) -> 2+ ticks -> 50% trim. Standing de-risk on this name has lapsed 6 sessions; the book exited Lenovo+YOFC instead. Bank half into the bounce at a better 8.28 (vs 7.83 yesterday); keep 25,700 for the oversold recovery. NA_CCY valuation-blind (val_adj 0). L2/L3 de-risk basis; L8 pending next open. · news |
| 7709 | XL2CSOPHYNIX (2x SK Hynix) | 82.74 | SELL 72% | 82.74 | 72.00 | n/a | n/a | 107.08 (+29.4%) | The 2x leveraged SK-Hynix ETF flipped from RallySell (reversal) to BreakdownDown_20D (breakdown-family, stable 2 prints) at the close and both overlay invalidators are now TRUE: close 82.74 << 101 stop AND signal turned breakdown-family -> the overlay override itself calls for full exit, and the leveraged-ETF rule permits a full/large exit on a breakdown-family print. Bank ~72% (sell) of the -28%-underwater leveraged bleed; keep only a 500-share rider through the 07-10 ADR debut (the leveraged-ETF learning: don't clean-exit into a two-sided +/-16% catalyst - 7747 05-13 full-exit missed +5.8%). scorecard BreakdownDown_20D|sell +4. Leveraged 12wk 107.08 and 4wk -50.15 are noise, not objectives. L7 (hk parent governs the shared 7709 share). L8 pending next open. · news |
| 0823 | Link REIT | 37.48 | SELL 36% | 37.48 | 36.50 | n/a | n/a | 32.91 (-12.2%) | Continue the twice-lapsed REIT de-risk. The settled close softened to AVOID/PP0 (model backed off the 85% trim), but it prints the book's reliable FailedBreakout_Down_20D sell cell (cv63, +4 bias) with a negative 12wk (32.91, -12.2%) and is PROFIT_ARMED + fired (dead-12wk tick c). Take the model's softened size -> 30% (rounds to lot 500 = sell 1,000, ~36%), keep 1,800. NA_CCY valuation-blind. L2/L4 (settled close = softened size); L8 pending next open. · news |
| 0005 | HSBC | 151.90 | NO ACTION | 151.90 | 148.30 | n/a | n/a | 168.73 (+11.1%) | PROFIT_ARMED and marginally fired (tick d only: 3x consecutive NoTrade consensus-conflict at close), BUT the 12wk is +11% POSITIVE (168.73 vs 151.9, tick c does NOT fire), the overlay carries an explicit HOLD-override (invalidated only on close below stop; close 151.9 > 148.3), and G3 forward-view gate fails on the positive 12wk. Slot-capped behind the 3 negative-12wk de-risks -> trail-raise not trim: hold the 12,000 stop at 148.3 (2.0x ATR chandelier). Execute a 30% profit-trim (~3,600) only if a DISPOSE slot frees OR the 12wk turns negative. Ledgered conditional so the counterfactual grader prices not-trimming. L8 pending next open. |
| 0700 | Tencent | 478.80 | NO ACTION | 478.80 | 430.00 | n/a | n/a | 417.59 (-12.8%) | Ran up to 478.8 (from 461) and the close is NoTrade cv0, but PROFIT_ARMED + fired a profit-take (tick c dead 12wk 417.59, -12.8%). A 30% trim is warranted into the strength, but the overlay override caps any trim at 50% + hold-core ~430 (news confirms the WeChat-AI/CXMT thesis) and it is only (~HK) -> slot-capped behind the 3 larger de-risks. Execute 30% (sell 60, keep 140) if a slot frees; core stop 430. Ledgered conditional. L8 pending next open. · news |
Outcome & track record
Accuracy at the time of this record.
90 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.