Recommendations
0 to acquire · 0 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 |
|---|---|---|---|---|---|---|---|---|---|
| 2888 | STANCHART | 232.20 | HOLD | — | 222.07 | 234.08 | 231.91 | 224.39 | Standard Chartered stays untouched. Yesterday's 30% de-risk executed cleanly at the open ( at HK, banking HK), and the remaining stake is now behaving better, not worse: the signal came back to life at conviction 51 after three straight sessions of zero, and the shares sit comfortably above our HKrisk line. There is no news and no fresh reason to cut further. OPERATIONAL: the ADR-0012 profit-take is armed on paper (PROFIT_ARMED, T12_Progress 103.48) and tick (c) reads as fired (Target_12Week 224.39 < 1.02 x close 232.20) -- both are SUPPRESSED as the HF-2 T12-inversion artifact, and the arm/tick are the same single fact (the 12wk target sitting below the close). Gain_ATR is 0.54, nowhere near the 3.0 arm threshold, so this is a +1.56% position, not a winner near target. HF-2 requires an independent bearish basis (decisive close < stop, or a fresh high-conviction formal SELL) before a profit-take on an inverted-12wk name; neither exists. Separately, the 08-06 state.json post_execution_trigger pre-committed that the CLAUDE.md section 9 N-consecutive-NoTrade count RESETS on the executed cut and that a further trim needs a NEW basis -- decisive close below 222.07, a non-PROFIT_ARMED Exit_Warning code, a fresh high-conviction formal SELL, or a fresh 2-consecutive-NoTrade run from 08-07. NONE is present: today prints HOLD cv51 (label change from NO TRADE, streak broken at the executed cut, exactly as pre-committed). Stop HELD at 222.07 -- the 2.0-ATR chandelier computes to 219.01, BELOW the ratchet floor, so no raise is available (never lower a ratcheted stop). ADR-0020 trail-erosion: stop 222.07 vs peak_stop 222.07 = 0% erosion, no fire, and the line was cut today regardless. Residual stop-risk on = (232.20-222.07) x 1100 = HK (0.56% of book, down from HKpre-trim). Note the stop still sits HKbelow the 228.63 cost. News: none found 08-06->08-07; the only item is the HKdividend that went ex on 08-06, already handled under HF-5 (no reprice). · news |
| 3968 | China Merchants Bank | 48.68 | HOLD | — | 48.10 | 48.54 | 48.92 | 49.26 | China Merchants Bank filled this morning exactly on our limit -- at HK -- and closed the day a whisker higher at HK. The signal has cooled from buy-grade to a watch reading on day one, which is common and is not a reason to undo a position we deliberately queued: no sell label, no warning flag, the shares are above our HKrisk line and the three-month view still points modestly higher. We hold and set the stop. OPERATIONAL: textbook HF-6 -- a freshly-filled entry that fades on day one is HELD, not cut, when no exit leg fires. Checked each leg: no SELL/REDUCE label (WATCH), Exit_Warning blank, NOT profit-armed (Gain_ATR ~0 on a same-day fill), close 48.68 > stop 48.10, and the 12wk is NON-inverted (Est_12W 49.26 = +1.19% vs close), so L2's stop-break-AND-negative-12wk combo fires on neither leg. Conviction round-trip 71 (08-06 buy print) -> 40 today is a fade, not a breakdown. FIRST STOP SET at 48.10 (the anchor Stop_Price) -- this closes the monitor's `3968 no_stop` warning, which was expected on a same-day fill. Stop-risk = (48.68-48.10) x 2000 = HK (0.06% of book); note 48.10 sits HKBELOW the 48.65 cost, so a stop-out realises roughly -HKValuation NA_CCY (CNY financials vs HKD price) so val_adj 0, but the underlying screen is the book's best: Quality_Pass TRUE on an F-Score of 5 and Quality_Rank 4.5. KNOWN ARTIFACT: the anchor scan wrote 8 seconds BEFORE the executor booked the fill (scan 17:00:19, txn 17:00:27), so the CSV still shows blank Avg_Cost/Held for 3968 and stale for 2888 -- this is learnings backlog item (b), the same executor/scan race flagged on 07-17. Position data here is taken from hkfin_transactions.jsonl and input/hkfinportfolio.json, which are authoritative. News: none found. · news |
| 2388 | BOC Hong Kong Holdings | 50.20 | NO ACTION | 49.51 | 48.29 | 49.80 (+0.6%) | 50.03 (+1.1%) | 50.02 (+1.0%) | BOC Hong Kong was the day's only buy-grade name and we are still passing on it. The signal is the right kind -- a pullback entry rather than a chase -- but it is simply not strong enough to clear our bar, the shares screen as expensive against what the bank's own book earns, they fail our balance-sheet quality test, and the model's own three-month projection sits BELOW today's close. Buying something the model expects to go nowhere is how this book has lost money before. We pass. OPERATIONAL: rejected on three independent grounds, any one sufficient. (1) FLOOR -- conviction 56 + val_adj -1 + bias 0 = adjusted 55 vs the 65 floor, short by 10. (2) HARD QUALITY GATE -- Quality_Pass FALSE (F-Score 4, Quality_Rank 9.) bars a NEW entry outright; 2388 is unheld so the gate applies at full force. (3) HF-3 INVERTED-12WK -- Target_12Week 50.02 and Est_12W 50.02 both sit BELOW close 50.20 (-0.36%); HF-3 is explicit that HK-financials buy-chases into inverted/negative 12wk views are near-uniformly stop-first (every filled hkfin buy graded so far -- 1299, 3988 x3, 2388 -- stopped first) and must be queued-at-pullback or rejected. HF-4 does NOT fire (Est_1W 49.80 > entry 49.51), so the 1-week view is the one healthy leg, which is why this is a rejection and not a red flag. Signal trail is choppy rather than building: BUY cv60 (08-04) -> WATCH 46 -> WATCH 50 -> BUY cv56, i.e. below the 08-04 print. Not queued: a GTC limit at 49.51 would be buying the same inverted-12wk exposure three days later, and no queue can cure a hard quality-gate block. RE-ARM TRIGGER: revisit only when BOTH Target_12Week and Est_12W clear the close AND raw conviction reaches 66+ (65 floor + the 1-point valuation haircut) -- the quality gate would still need addressing separately at the next fairvalue refresh. Valuation is one of only 3 non-NA reads in this book: OVERVALUED_VS_RI, P/FV 1.27, MoS -0.269 -> val_adj -1. News: none found. COUNTERFACTUAL CAVEAT: the scorecard's no_chase_queue bucket (n=6) shows this book's passes have COST a median 1.79% at 1wk and 11.32% at 4wk -- advisory only, never blended into conviction, but it is the systemic finding of this run. · news |
Outcome & track record
Accuracy at the time of this record.
4 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 →
0.184 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.
50 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.