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 | 234.60 | HOLD | — | 223.61 | 236.08 | 232.44 | 226.48 | Standard Chartered slipped 0.6% to HKon a day when every name in the book fell, and we are staying with it: the position is still up 2.6% (about HK) and trades a comfortable HKabove our HKprotective stop. The one thing that did change is the strength of the signal, which weakened materially -- but not enough, and not for the right reason, to justify selling into a broad down day with no company news behind it. The bank screens on a foreign-currency basis so our valuation model has no usable read here; the decision is a technical one. Operational nuance: the scan again flags PROFIT_ARMED, again suppressed under ACTIVE HF-2 (12th consecutive session) -- the arm comes purely from an inverted 12-week target (226.48 < close 234.60, -3.46%) driving T12_Progress to 103.59, while Gain_ATR is 1.07 against a 3.0 threshold. ADR-0012 fire ticks: (a) no other Exit_Warning; (b) conviction = -17, three points short of the >=20 threshold -- the closest this tick has come; (c) reads fired but is mathematically implied by the same inversion that created the arm (systemic finding #2); (d) no NO TRADE print. L2 fails its stop-break leg (234.60 vs 223.61, 1.98 ATR cushion). TRAIL: no raise -- the 2.0-ATR chandelier computes to 223.49, HKBELOW the 223.61 ratchet floor, and the ratchet is never lowered; next raise when close > ~234.72. ADR-0020: stop 223.61 vs peak_stop 223.61 = 0% erosion, no fire. Stop-risk HK (0.60% of book). PRE-COMMITMENT next session: trim 30% if conviction falls >=20 from 40 (i.e. <=20), OR any non-PROFIT_ARMED Exit_Warning prints, OR two consecutive NO TRADE from 08-12, OR a fresh formal SELL/REDUCE at cv>=65; full exit on a decisive close below 223.61. · news |
| 3968 | China Merchants Bank | 48.34 | HOLD | — | 48.10 | 48.26 | 48.42 | 48.93 | China Merchants Bank fell 1.4% to HKand has slipped just back below our HKcost, with the trading signal going flat -- the model now sees conflicting evidence rather than a bullish case. We are holding rather than cutting, for one concrete reason: the loss is already fenced. Our stop sits at HK, just 24 Hong Kong cents below the close, so the most this position can cost from here is about HKbefore the stop takes us out -- and selling a four-day-old position into a market-wide down day has repeatedly cost this book more than it saved. The bank remains the highest-quality name we own on the fundamental screen, and its three-month view is still modestly positive. Operational nuance: PROFIT_ARMED printed again and is SUPPRESSED on the ADR-0012 arm criterion itself -- Gain_ATR is -0.30 (the position is at a LOSS) and T12_Progress 98.79 is merely Close/Target_12Week (48.34/48.93); true progress-from-cost is (48.34-48.65)/(48.93-48.65) = -111%. Neither ADR-0012 arm limb is met, so no profit-take is reachable. This is the FLAT/non-inverted 12wk variant (48.93 > 48.34, +1.22%), which ACTIVE HF-2 does not textually reach and HF-P15/HF-P1 remain inert. Fire tick (b) DID fire (conviction = -47) and is deliberately not honoured -- arm and tick are a conjunction; no arm, no profit-take. Not a sell on three tests: §9 'sustained' NoTrade = ADR-0012 tick (d) 2 consecutive, this is the 1st; L2 fails both legs (close above stop, 12wk positive); HF-6 is directly on point for a freshly-filled day-4 fade with no exit leg firing, backed by the optional_trim counterfactual (n=43, +2.36% median saved). Stop UNCHANGED at 48.10 -- the chandelier from close is 46.26, far below, and the ratchet is never lowered; next raise when close > ~50.18. ADR-0020: 0% erosion. PRE-COMMITMENT next session: trim 30% on a SECOND consecutive NO TRADE, booked as a signal-degradation trim NOT a profit-take (the position is not armed in substance); full exit on a decisive close below 48.10; re-arm to hold if conviction recovers to >=45 with a non-NO-TRADE label. · news |
| 2388 | BOC Hong Kong Holdings | 50.35 | NO ACTION | 50.02 | 48.96 | 50.22 (+0.4%) | 50.05 (+0.1%) | 50.39 (+0.7%) | BOC Hong Kong is the strongest buy signal in the book today and we are still passing on it. The problem is what we would be paying for: the model's own three-month target of HKsits just 8 basis points above today's HKclose, while the protective stop would sit 2.8% below -- risking roughly HKto make 4 cents. The bank also fails our fundamental quality screen (the weakest F-Score band in the book), and the stock has already run past the model's buy price, so we would be chasing it. We would rather own it cheaper, with a real forward view. Operational nuance: adjusted conviction 61 + val_adj -1 + bias 0 = 60 vs the 65 floor, short by 5; Quality_Pass FALSE is a HARD gate on a NEW entry (F-Score 4, Quality_Rank 9.5/11); flat 12wk (+0.08%) is HF-3's spirit if not its inverted letter; Entry_Distance_ATR -0.32 trips the no-chase bar (<= -0.25). HF-4 does NOT fire (Est_1W 50.22 > entry 50.02). Not queued -- a GTC limit cannot cure a hard quality-gate block. Fourth consecutive session withheld. RE-ARM TRIGGER (all four, on a settled close): Target_12Week > Close by >= 2% AND raw conviction >= 66 AND Quality_Pass TRUE at the next fairvalue refresh AND Entry_Distance_ATR > -0.25. · news |
| 0388 | HKEX | 405.40 | NO ACTION | 403.47 | 395.68 | 405.17 (+0.4%) | 408.72 (+1.3%) | 422.81 (+4.8%) | HKEX is the most improved name in the book and still a pass. Two things genuinely got better after its 2.7% fall: the entry pattern rotated to the pullback style we trust most, and the three-month view flipped from negative to +4.3% because the price dropped while the target held. What did not improve is the case for owning it: the exchange still screens as expensively priced against what its own business earns, it still fails our fundamental quality screen, and it reports interim results in six trading sessions -- buying a quality-gate failure into an earnings print is not a risk-defined entry. We are tracking it, not buying it. Operational nuance: conviction fell; 60 + val_adj -6 + bias 0 = adjusted 54 vs the 65 floor, short by 11 (a wider miss than yesterday's 7). Quality_Pass FALSE (F-Score 4) is a HARD gate on a NEW entry. HF-3 no longer fires -- Target_12Week 422.81 > close 405.40 (+4.29%), reversed from -0.30% inverted. HF-4 does not fire (Est_1W 405.17 > entry 403.47). Signal rotated BreakoutUp_20D -> PullbackBuy_VWAP, the L1-favoured family. Earnings + interim dividend 2026-08-19, 6 sessions out. RE-ARM TRIGGER (all four, on a settled close): raw conviction >= 71 AND Quality_Pass TRUE at the next fairvalue refresh AND Target_12Week > Close AND the 08-19 results are PAST, not pending. Two of the four legs moved the right way today -- track it. · news |
Outcome & track record
Accuracy at the time of this record.
6 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.