Recommendations
2 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 |
|---|---|---|---|---|---|---|---|---|---|
| O39 | OCBC Bank | 31.85 | ADD — QUEUED @30.79 | 30.79 | 29.74 | 31.76 (+3.2%) | 31.81 (+3.3%) | 33.18 (+7.8%) | OCBC is the book's strongest signal of the session - a pullback-to-VWAP buy at conviction 90 and the number-one buy rank, on a bank whose 1H2026 interim dividend rose 15% to Swith the lowest bad-loan ratio and highest capital ratio of the three Singapore banks. The shares closed two ATR above the model's trigger, so this is a limit order waiting for a pullback rather than a purchase at today's price, and the valuation is a mild negative (priced slightly beyond its residual-income anchor) which trims the size but does not block a top-up of an existing holding. Ops: L1-favoured family; no-chase queue (eD -2.02) per SG-6 (cv90 < 95 threshold, so no immediate tranche). scorecard_bias 0 - PullbackBuy_VWAP|buy is n=4 hit_1w.000, below the sample floor, stated not applied. Quality_Pass False does not bind - gate blocks new entries only, this is an ADD. Size cut from the model's to by (60% of Sallocated to the #1 name). EXECUTION CONFLICT: this 30.79 limit sits 1 cent below tonight's raised trail of 30.80 - fill only on an intraday touch that does NOT settle below 30.80; a settled close below 30.80 cancels this queue and arms the trail exit on the core 1,000 instead. 12wk 33.18 exceeds the DCF column (32.21); band is OVERVALUED_VS_RI and the RI model governs for a bank. NFP 09-04 is a HIGH-confidence two-sided event inside the 3-session queue life. · news |
| D05 | DBS Group Holdings | 77.60 | ADD — QUEUED @75.88 | 75.88 | 73.89 | 77.38 (+2.0%) | 77.18 (+1.7%) | 79.99 (+5.4%) | DBS flipped into the pullback-to-VWAP setup mid-session and closed the day at conviction 84, the book's second-best buy signal, backed by a 1H2026 payout of Sper share including a Scapital return and a 14.8% capital ratio. As with OCBC the shares finished well above the model's entry trigger, so this is a patient limit order into weakness, sized small because the book's available cash is only 1.3% of its value - the constraint here is cash, not conviction. Ops: L1-favoured family, three consecutive prints above the 65 floor. eD -1.73 - no-chase queue; cv84 < SG-6's cv>=95 immediate-tranche threshold. scorecard_bias 0 (cell below sample floor). Quality_Pass False does not bind on an ADD. PROVENANCE: this is NOT the 08-31 @77.20 buy-stop, which was withdrawn 09-01; price is re-derived from the live anchor row and the stale level must not be republished. Size cut from the model's to by (40% residual = S). BLOCKING: the 08-31 D05 card is still UNRECONCILED for a second session (trigger traded 09-01, high 77. unchanged at 6,900) - record it TAKEN or SKIPPED before acting on this card. 12wk 79.99 far exceeds the DCF column (57.01) but that column is unreliable for financials on this book (it prints 0.36 for U11); the RI/FV anchor governs. · news |
| O39 | OCBC Bank | 31.85 | HOLD | 30.79 | 30.80 | 31.76 (+3.2%) | 31.81 (+3.3%) | 33.18 (+7.8%) | The one position in the book sitting on a real, verified profit (+22.8% against a known cost of 25.93) has reached the level where the model watches for signs of the run ending. One mild warning did appear, but the signal itself strengthened to its highest reading of the month, so the right response is to raise the protective stop and keep holding rather than take money off the table. Ops: ADR-0012 armed (Gain_ATR 11.25) - tick audit: (a) FIRES on RSI_DIV; (b) no, cv (+37); (c) no, Est_12W 33.18 > 1.02x31.85=32.49; (d) no. Single Tier-1 tick on a fresh bullish cv90 print -> G1 substance applies (family is PullbackBuy_VWAP, not literally on G1's Breakout/MomentumContinuation list, but every condition G1 protects is present) -> TRAIL-RAISE not trim. SG-8 independently bars re-publishing the stalled de-risk. Trail 30.52 -> 30.80 (2.0x ATR chandelier 31.85 - 1.052); first actual move since it was set. NOTE: collides by 1 cent with tonight's 30.79 ADD limit - the ADD is subordinate to the trail. |
| A17U | CapitaLand Ascendas REIT | 2.38 | HOLD | 2.41 | 2.42⚠ | 2.38 (-1.2%) | 2.40 (-0.4%) | 2.43 (+0.8%) | The model's loudest sell warning of the day is declined. The signal only appeared on the final bar of the session, the stock is already deeply oversold, and the twelve-week outlook still points higher - and the holding is worth under Stoo small to trade sensibly either way. Ops: SG-2 leg 2 FAILS - Est_12W 2.43 vs close 2.38 = +2.1% positive. Stop break present (-1.65%) but that is leg 1 only. One-bar family flip RallySell_BearTrend -> BreakdownDown_52W with cv58->85 on the 18:00 print; BreakdownDown_52W|sell is n=1 on this book - no evidence. RSI 34.0. MoS -7.72 clears SG-3's <=-3 amplify trigger but amplify-only means there must be a qualifying signal-based sell to amplify - there is none. Position = Ssub-threshold. Stop held 2.42. 3rd consecutive refusal. |
| J69U | FRASERS CPT TR | 2.15 | HOLD | 2.15 | 2.18⚠ | 2.15 (+0.0%) | 2.15 (+0.0%) | 2.19 (+1.9%) | Frasers Centrepoint Trust prints a strong sell label for the sixteenth time and is declined again for the same reason: the trust is oversold, its twelve-week outlook is still positive, and this family of sell signals has been wrong more often than right on this book. The wider Singapore REIT weakness is a rate repricing the model wants handled by tightening stops, not by selling into the low. Ops: SG-2 leg 2 FAILS - Est_12W 2.19 vs close 2.15 = +1.9%. Stop break -1.38% is leg 1 only. FailedBreakout_Down_20D|sell bias 0, n=73, hit_1w.507 / hit_4w.375 - worse than a coin flip at 4wk. RSI 34.3. Valuation NA - no amplify available. Event Horizon prescribes tighten-trail for the S-REIT leg. Position = SStop held 2.18. 16th refusal. |
| HMN | CAPITALAND and ASCOTT STAPLE U | 0.85 | HOLD | 0.85 | 0.88⚠ | 0.85 (+0.0%) | 0.85 (+0.0%) | 0.88 (+3.5%) | A sell label with almost no conviction behind it on a holding worth about Swhere the twelve-week view still points modestly higher. Nothing here justifies a trade. Ops: adj conviction 31 is 34 points under the 65 floor - fails outright. SG-2 leg 2 also FAILS: Est_12W 0.88 vs close 0.85 = +3.5%. Stop break -3.41% is leg 1 only. Position = Ssub-threshold. Long-standing note: HMN's multi-session breach is probably largely an unrebased distribution. Stop held 0.88. |
| S63 | ST Engineering | 10.49 | HOLD | 10.49 | 10.51⚠ | 10.50 (+0.1%) | 10.45 (-0.4%) | 10.43 (-0.6%) | ST Engineering slipped two cents below its stop with a marginally negative twelve-week view - technically the book's exit combination, but the breach is trivial and the company just posted first-half revenue up 11%, profit up 27%, a record S7bn order book and a raised dividend. This is noise around a strong business, so the position is kept and the level watched. Ops: L2 both legs met on the letter - close 10.49 vs stop 10.51 (-0.19%) AND Est_12W 10.43 vs 10.49 (-0.57%). REFUSED on three grounds: breach is 0.09 ATR, not the 'decisive close' L2 requires; signal is a bullish-family FailedBreakdown_Up_20D at cv34 with no Sell_Rank and Portion 0 - the engine asks for nothing; forward view is flat not negative (Est_1W +0.1%, Est_4W -0.4%). MoS -10.50 would amplify a sell if one existed; none does. RE-ARM: a settled close below ~10.40 with Est_12W still negative arms the L2 exit on a Sposition. Stop held 10.51. · news |
| PORTFOLIO | Singapore Main | — | WATCH | — | — | n/a | n/a | n/a | The Singapore bank leg reopened the buy side after a week of nothing, and both DBS and OCBC are queued as pullback limit orders rather than bought at today's prices. Nothing is sold: every sell signal on the book sits on a REIT that is oversold with a still-positive twelve-week outlook, which is the book's standing reason not to sell into this kind of rate-driven weakness. Ops: 42nd cycle. 2 ACQUIRE queued (O39 400 @30.79, D05 100 @75.88; combined Svs Spowder, Sresidual), 0 DISPOSE, 0 ANTICIPATE (no BUY - ANTICIPATE prints). Sell side refused on SG-2 for the 6th consecutive cycle - all three Est_12W positive. F34 near-miss adj 64, one point under floor. 7 stop breaches, no stop lowered anywhere. O39 08-21 trim withdrawn; O39 trail raised 30.52->30.80. BLOCKING: 08-31 D05 card UNRECONCILED for a 2nd session (ADR-0019). Step 8a NOT run. |
Outcome & track record
Accuracy at the time of this record.
122 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 →
0.229 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.