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 | Standard Chartered | 223.00 | HOLD | — | 214.98 | 224.26 | 227.31 | 224.65 | We are holding Standard Chartered rather than trimming it, and today that required overruling the model's own profit-take trigger. The position is 2.5% under water (223.00 against a 228.63 cost), yet the scan flagged it PROFIT_ARMED — a false alarm caused by the twelve-week target sitting BELOW our purchase price, so the position reads as '99% of the way to target' while actually losing money. Taking a 'profit-take' on a loss is not a trade, it is a disguised capitulation, so the trigger is suppressed. Every genuine exit test still fails: no sell signal, conviction steady, the stop at 214.98 untouched with a comfortable 1.4-ATR cushion, and the twelve-week view still slightly above today's close. The bank's story is unchanged (the wealth-management pivot targeting US$200bn of new client assets) and H1 results land on 29 July — an event to trail into, not to pre-empt. The stop, not a discretionary exit, remains the plan. OPERATIONAL: HOLD FailedBreakdown_Up_20D cv50 (flat vs 07-17, no >=20pt collapse -> no tick b); Exit_Warning=PROFIT_ARMED SUPPRESSED per HF-2 (widened to the cost-inverted case): T12_Progress 99.27 arms only because Target_12Week 224.65 < cost 228.63 while Gain_ATR = -1.0 (a loss); fire-tick (c) (Target_12Week 224.65 < 1.02 x Close 227.46) is the SAME artifact counted twice -> also suppressed. No other Exit_Warning code (no tick a), not 2 consecutive NO TRADE (no tick d). No stop break (223.00 > 214.98 = 1.43 ATR cushion); 12wk non-inverted vs close (+0.74%); L2 combo fires on neither leg; overlay override (0005/2888 hold-above-stop) endorses. Stop held at 214.98 (tighter than the scan's widened 211.74 chandelier); NO trail-raise — the armed-but-not-fired trail rule presupposes a profit to protect, and tightening into a -2.46% drawdown manufactures the whipsaw the optional_trim counterfactual (n=20, +2.24% median saved) warns against. Remaining risk from cost = (228.63-214.98) x 1,550 = HK (~1.06% of book). HF-6 second session. Exit basis if it matures: decisive close below 214.98, or a 12wk turning negative alongside it. · news |
| 0005 | HSBC Holdings | 157.80 | NO ACTION | 158.67 | 150.53 | 155.30 (-2.1%) | 158.43 (-0.2%) | 161.11 (+1.5%) | HSBC is setting up for another run at its 52-week high and, for the first time, qualifies structurally for our anticipation slot — but we are standing down on one stubborn detail: the model's own one-week estimate (155.30) sits 2.1% BELOW the 158.67 price we would have to pay. That is the identical signature that preceded our Standard Chartered entry, which is now 2.5% under water. The shares are also overbought (RSI 72) heading into H1 results in roughly nine days. We would rather buy HSBC on a pullback the model believes in than pay up for a breakout it does not. The twelve-week view (+2.1%) is genuinely constructive, so this stays firmly on the watchlist. OPERATIONAL: BUY - ANTICIPATE AnticipationUp_VCP cv79, adj 79 (val_adj 0 NA_CCY, bias 0) -> clears the 65 floor. Anticipate-slot structural gates PASS for the first time: 2/3 ANTICIPATE prints in the last 3 snapshots (07-16, 07-20) at cv>=55, and HF-3's non-inverted-12wk leg passes (Est_12W 161.11 = +2.10% vs close 157.80). WITHHELD on HF-4 LOW_QUALITY_ENTRY: Est_1W 155.30 < entry 158.67 (-2.12%); the carried 07-17 re-arm requires Est_1W >= entry on the anticipate slot as well as the acquire slot. Entry non-extended (+0.32 ATR) but RSI 71.95 rising into a ~07-30 H1 print. RE-ARM: >=2/3 ANTICIPATE prints (cv>=55) WITH Est_1W >= entry AND non-inverted 12wk -> anticipate slot at 0.5x size; OR non-extended actionable BUY cv>=65 meeting the same two estimation legs -> acquire slot. |
| 2388 | BOC Hong Kong Holdings | 47.10 | NO ACTION | 47.10 | 46.24 | 46.89 (-0.4%) | 46.79 (-0.7%) | 46.85 (-0.5%) | BOC Hong Kong turned into the book's top-ranked buy candidate today, and we passed on it for four independent reasons. It fails our fundamental quality screen, so it cannot be opened as a new position at all. Its own twelve-week estimate (46.85) is BELOW today's close (47.10) — the model expects it lower in three months — and its one-week estimate sits below the entry price too. Finally, its signal family is the weakest buy pattern in our record (roughly one in four works, and six in ten hit the stop first) and is the very same pattern our losing Standard Chartered position now prints. A top rank on a poor setup is still a poor setup. OPERATIONAL: BUY FailedBreakdown_Up_20D cv68, Buy_Rank 1, at-entry 47.10 (EDist 0.0), RSI 58.4. adj_conv 67 = 68 + val_adj -1 (OVERVALUED_VS_RI MoS -0.174 -> round(-0.87)) + bias 0 -> clears the 65 floor but blocked by: (1) HARD quality gate Quality_Pass=False on a NEW entry (F-Score/FCF screen); (2) HF-3 inverted 12wk (Est_12W 46.85 < Close 47.10, -0.53%); (3) HF-4 Est_1W 46.89 < entry 47.10; (4) L1 family discount (FailedBreakdown_Up_20D ~0.25 hit, ~60% stop-first). RE-ARM: quality-pass turn (Q=True) AND non-inverted 12wk AND Est_1W >= entry, at cv>=65. |
| 0388 | HKEX | 397.40 | NO ACTION | — | — | 397.56 | 397.84 | 396.79 | HKEX remains off the buy list for a second session — the indicators are in conflict and the engine offers no trade, even as the shares rallied 6..40. The pullback entry that tempted us last week never arrived, and the exchange still screens as expensive against its residual-income value while failing our quality screen. Our patience has now cost nothing across four sessions. Nothing to do. OPERATIONAL: NO TRADE cv0 ('Consensus conflict + low conviction'), second consecutive session. OVERVALUED_VS_RI MoS -1.063 -> val_adj -5; Quality_Pass=False. RE-ARM: quality-pass turn (Q=True) with cv>=65 AND a non-extended pullback fill near the printed entry. |
Outcome & track record
Accuracy at the time of this record.
3 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.
22 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.