Recommendations
0 to acquire · 1 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 | 234.00 | SELL 30% | — | 222.07 | 235.64 | 234.09 | 232.22 | We are selling 450 of our 1,550 Standard Chartered shares, about 30% of the holding, at roughly HKand keeping the rest. The shares have risen 23% above their 200-day average and remain 2.4% above what we paid, but our signal engine has now refused to give any opinion on the stock for two consecutive closes - its trend reading is still strong while its momentum reading has turned outright negative, and the two cancel out. Four straight lower closes and a three-month projection that sits slightly below today's price say the same thing: the advance has stalled. Crucially, our protective stop at HKsits below our purchase price, so the profit we are showing is not actually defended by it; selling a third banks HKof that gain for certain and cuts what is at risk if the stock reverses, while leaving 70% of the position to participate if the uptrend resumes. This is a stall being respected, not a view that the bank is in trouble - its July half-year results beat and it raised guidance. Operational: exit reason is CLAUDE.md section 9 'N consecutive NoTrade/Conflicted on a held position' (N=2: 08-04 and 08-05 both NoTrade cv0, 'Consensus conflict + low conviction'), firing the HARD escalation trigger pre-committed in the 08-04 state file. This is NOT an ADR-0012 profit-take: PROFIT_ARMED is suppressed for a 9th session under HF-2 (Gain_ATR 0.80 vs 3.0 threshold; T12_Progress 100.77 is merely 234.00/232.22 on an inverted 12wk), so the arm is void and ticks (c)/(d) are moot - this originates from a different, arming-independent door. Valuation NA_CCY blind (the -7.556 MoS is a USD-financials-vs-HKD-price artifact), val_adj 0, NO tier bump applied. Scorecard bias 0 (no actionable cells); the optional_trim counterfactual (n=36, +2.44% median saved @1wk) argues AGAINST and is consciously overridden - it is advisory only, its 36 samples are trims against intact signals, and L3 forbids re-deferring a written pre-commitment. Pillars Trend 70 / Momentum 15 Bearish (2nd session) / Vol 45, Conflicted, regime Transitional, RSI 63.34, conviction trail ->. 30% tier chosen deliberately over 50%/100%: the trend is intact (close 234.00 vs EMA200 189.40) and no stop break has occurred. 465 raw -> 450 on the 50-share board lot. Stop stays 222.07 on the remaining (2.0-ATR chandelier prints 220.52, below the ratchet floor); residual stop-risk HKHF-5: 2888 goes ex-dividend HKon 2026-08-06 - the mechanical ~0.09% mark-down at the execution open is NOT deterioration and must not trigger an L8 reprice (threshold 1.5%); HKforgone dividend is immaterial. Est_* per ADR-0013: 1wk +0.70%, 4wk +0.04%, 12wk -0.76% vs the action price. · news |
| 0388 | HKEX | 414.60 | NO ACTION | 412.42 | 396.75 | 414.90 (+0.6%) | 414.11 (+0.4%) | 413.39 (+0.2%) | HKEX broke higher today and was the strongest buy signal this book has produced in a week, but we did not buy it. Our valuation work says the exchange is already priced well above what its returns and book value justify, and its underlying financial-quality screen fails outright - a combination that bars a new position regardless of how good the chart looks. Even setting quality aside, the price penalty pulls our conviction score to 58 against a 65 minimum, and our own three-month projection sits fractionally below today's close, so the model is not forecasting the breakout to carry. We pass and keep the cash. Operational: BreakoutUp_20D BUY cv64, Buy_Rank 1, close 414.60 (+1.62%), Buy Stop @412.42, eD -0.36, RSI 69.46, Trend 75 / Momentum 41 / Vol 40. Blocked three ways: (i) adjusted 58 = 64 + val_adj -6 (MoS -1.118 degree-scaled) + bias 0, fails floor by 7; (ii) HARD quality gate, Quality_Pass FALSE, F-Score 4; (iii) HF-3, Est_12W 413.39 inverted -0.29% vs close. L1-discounted breakout family; BreakoutUp_20D|buy is 1/2 at n=2 (report-only). Re-arm: Quality_Pass TRUE plus actionable BUY at cv >= 71 raw with a non-inverted 12wk. · news |
| 3968 | China Merchants Bank | 48.76 | NO ACTION | 48.50 | 47.27 | 48.46 (-0.1%) | 48.26 (-0.5%) | 48.87 (+0.8%) | China Merchants Bank pulled back to the level we would like to own it and its signal improved from watch to buy, but two things keep us out. Our conviction score of 60 is short of the 65 we require before committing capital, and - more tellingly - our own one-week price projection sits fractionally below the price we would pay, meaning the forecasting model is contradicting the signal model. When our two engines disagree at the moment of entry, the historical record in this book says the entry is low quality. It stays on the watch list as the best-quality name we track. Operational: PullbackBuy_VWAP BUY cv60, Buy_Rank 3, close 48.76 (-1.42%), Buy Limit @48.50, eD -0.22, RSI 53.04, promoted from WATCH cv41. Fails floor by 5; HF-4 fires (Est_1W 48.46 < entry 48.50, the LOW-QUALITY-ENTRY signature that preceded the 07-17 fill). HF-3 passes marginally (Est_12W 48.87, +0.23% vs close). Still the only Quality_Pass TRUE name with a positive MoS (+0.223, NA_CCY). Re-arm: actionable BUY at cv >= 65 with Est_1W >= entry and a non-inverted 12wk. · news |
| 3328 | Bank of Communications | 7.34 | NO ACTION | 7.25 | 7.07 | 7.26 (+0.1%) | 7.27 (+0.3%) | 7.15 (-1.4%) | Sixth session running that Bank of Communications has come close and been passed over. Our conviction score of 60 is again below the 65 bar, and the three-month outlook has deteriorated further: the model now projects the shares 2.5% BELOW today's price in twelve weeks, worse than yesterday. Buying a bank our own forecast expects to be lower in three months is not a swing trade, and we are not lowering the standard to force a position. The limit order we cancelled yesterday is not being revived at its old price. Operational: PullbackBuy_VWAP BUY cv60, Buy_Rank 2, close 7.335 (-0.68%), Buy Limit @7.25 (live entry; the withdrawn 7.19 is NOT rolled and NOT repriced, provenance rule), eD -0.47, RSI 53.23. Fails floor by 5; HF-3 fails harder than 08-04 (Est_12W 7.15 = -2.5% inverted vs close, was -2.0%). Re-arm: actionable BUY at cv >= 65 with Est_12W non-inverted vs the CLOSE, not merely vs entry. · news |
| 3988 | Bank of China | 5.27 | NO ACTION | 5.27 | 5.13 | 5.24 (-0.6%) | 5.31 (+0.8%) | 5.25 (-0.4%) | Bank of China is now going the wrong way on the one measure that matters for entry: our conviction score has slipped to 56 from 60, further below the 65 minimum, and the stock has fallen through its 20-day average. The three-month projection is also fractionally below today's price, and the underlying financial-quality screen fails, which would block a new position on its own. Sixth consecutive session of waiting, and the case is weaker than when we started waiting. Operational: PullbackBuy_VWAP BUY cv56, Buy_Rank 4, close 5.27 (-1.22%), Buy Limit @5.27, eD 0.00, RSI 49.88. Conviction trail -> 56 (now falling). Fails floor by 9; Est_12W 5.25 inverted -0.4%; Quality_Pass FALSE (hard gate would block regardless); close below EMA20 5.30. Re-arm: actionable BUY at cv >= 65 with Quality_Pass TRUE and a non-inverted 12wk. · news |
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.
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.