Singapore — End-of-Day Verdict
39th cycle: **the buy side evaporates in a single session and the de-risk escalates to a market order** — S68, which printed the model's ceiling `cv100` two sessions ago, collapses to `HOLD cv63` and its unfilled Slimit is **formally withdrawn**, leaving the buy pool **structurally empty with no Buy_Rank populated anywhere on the book**; **U11's L2 combo re-fires for a second cycle with all three horizons now *deteriorating***, and because the identical 08-26 card was never executed, SG-1 escalates it from an open-reference sell to an **unconditional market sell**; the sell pool produces its **two loudest prints of the month** (J69U `cv84` Sell_Rank #1, C07 `cv71` Sell_Rank #2) and SG-2 refuses **both**; and a news pass turns up a **prospective** ex-dividend trap on C07 — Sex 09-01 — letting the book pre-empt for once the exact arithmetic that has misread B61 and HMN for eighteen sessions
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 |
|---|---|---|---|---|---|---|---|---|---|
| U11 | United Overseas Bank | 40.78 | SELL 30% | 40.78 | 41.27⚠ | 40.65 (-0.3%) | 40.75 (-0.1%) | 40.42 (-0.9%) | Singapore's second-largest bank has now closed below its Sprotective trail for ten straight sessions, and the model's own 1-week, 4-week and 12-week price paths have all turned negative at the same time - the first triple-negative reading of the cycle. Critically, the trail was already rebased for the Sdividend that went ex on 17 August, so this is genuine price weakness rather than dividend arithmetic. The fundamentals remain sound (Q2 profit +10%, record wealth-management fees, a S asset-management disposal booked in August), which is why this is a partial trim that banks roughly S and keeps 70% of the position invested, not an exit. OPS: L2 origination (decisive stop-break + negative 12wk) - the engine still prints NO TRADE cv0, so the label alone would never have fired this. SG-1 escalation applies: the identical 30% card issued 2026-08-26 for execution 08-27 was NOT executed (monitor: static at 12,246), so this is carry cycle 2 and the order escalates from open-reference to an UNCONDITIONAL MARKET SELL at the 08-31 open. SG-8's stop-republishing threshold is >=3 cycles - not yet reached, so re-publication is still correct. Trim-discipline gates G1/G2/G3 are out of scope (they govern profit-trims on uptrending names; U11 is below EMA20 and its PROFIT_ARMED flag is spurious per SG-7, no Avg_Cost). Valuation is neutral: MoS_FV -0.51 is nowhere near the -3 amplify threshold, so no tier bump; DCF_per_share 0.36 is a broken bank-RI artefact and is ignored. Residual keeps stop 41.27 tonight; re-base the residual trail next EOD post-fill. · news |
| O39 | OCBC Bank | 31.07 | HOLD | 31.07 | 30.42 | 30.93 (-0.5%) | 30.94 (-0.4%) | 32.13 (+3.4%) | Sitting on a 19.8% unrealised gain and recovering - the signal improved from NO TRADE back to HOLD and the 12-week path (S) is still 3.4% above today's Sclose. Nothing here justifies banking the gain today; the protective stop stays at S, which locks in the bulk of the profit. OPS: ADR-0012 armed-but-NOT-fired. Genuinely armed per SG-7 (real Avg_Cost 25.93, Gain_ATR 10.56). All four fire ticks miss: (a) no second Exit_Warning code; (b) conviction ROSE; (c) dead-12wk needs T12 < 1.02x close = 31.69, actual 32.13 - misses by S, a wider miss than the two prior four-cent misses; (d) NO TRADE streak broken at 1. Mandatory trail-raise: 2.0x ATR chandelier = 31.07 - 0.973 = 30.10, BELOW the existing 30.42 watermark, so the trail is HELD at 30.42, never lowered. SEPARATE and unchanged: the 08-21 SELL card remains outstanding for a 31st static cycle and is suppressed per SG-8 - clear with 'trade sell sg O39 300' or 'trade skip sg O39'. L7: sgotr holds 1,100 O39 against the same pool - sg governs, execute once. |
| J69U | FRASERS CPT TR | 2.16 | HOLD | 2.17 | 2.18⚠ | 2.16 (-0.5%) | 2.17 (+0.0%) | 2.20 (+1.4%) | The model is shouting its loudest sell of the month on this suburban-mall REIT - a top-ranked STRONG SELL at conviction 84 asking for an 80% cut of a Sholding - and we are declining it for the 15th time. The reason is simple: Singapore sell labels on this book reverse more than half the time, and every forward price path here is still pointing UP (12-week Svs today's S). Earnings are forecast to grow 8.4% a year on necessity-led suburban retail. The stock is a fraction below its Sstop, not decisively through it. OPS: SG-2 refusal, leg 2 fails - Est_1W 2.16 flat, Est_4W +0.46%, Est_12W +1.85%, none negative. Highest sell conviction the name has printed -> 84 whipsaw across three sessions, itself evidence of label instability). Breach is only -0.92%, not decisive. Re-arm trigger: a close below 2.10 (model live stop) WITH Est_12W turning negative flips this to a mandatory trim. Trail HELD 2.18 (model live stop 2.10 is lower). · news |
| C07 | Jardine C&C | 27.19 | HOLD | 27.19 | 26.56 | 27.27 (+0.3%) | 27.46 (+1.0%) | 27.26 (+0.3%) | A fresh second-ranked STRONG SELL is being refused because the news flow contradicts it outright: Chandra Asri agreed on 21 August to buy the Malaysia/Singapore car-dealership arm for S, with proceeds going to debt paydown, and a Sdividend goes ex on 1 September. This is the only name in the book that screens genuinely cheap on quality-adjusted intrinsic value. The stock is above its stop and every forward path is flat-to-positive - there is no exit case here. Hold through the dividend. OPS: SG-2 refusal on BOTH legs - no stop break (27.19 vs 26.56) and Est_12W +0.26%. Valuation argues the trim DOWN a tier anyway (UNDERVALUED, MoS +0.62, Quality_Pass True). CATALYST WARNING for the next EOD: Sex-div on 2026-09-01 is ~1.03% of price and will mechanically drop the close - do NOT read it as a stop break. Rebase the trail 26.56 -> 26.28 once ex. This is the B61/HMN ex-div arithmetic lesson applied prospectively rather than forensically. · news |
| B61 | BT SEMBAWANG EST | 4.33 | HOLD | 4.33 | 4.62⚠ | 4.33 (+0.0%) | 4.37 (+0.9%) | 4.42 (+2.1%) | Deeply oversold (RSI 26) and 6.3% below its recorded stop for a fifteenth session, yet still not a sell: the forward view is positive across all three horizons (12-week Svs Sclose), the label has whipsawed violently in four sessions, and the shares screen as undervalued. Roughly 40% of the headline stop breach is dividend arithmetic the August rebase under-corrected, not real damage. OPS: SG-2 leg 2 fails (+2.08% Est_12W). Label trail SELL70 -> HOLD57 -> STRONG SELL37 -> AVOID60 - textbook SG label instability, the exact behaviour SG-2 was written for. Trail HELD 4.62 (model live stop 4.44 is lower; never lowered) - but note this watermark is itself ex-div-inflated and remains the book's most overstated risk statistic. L7: sgotr also carries B61. |
| D05 | DBS Group Holdings | 76.15 | HOLD | 76.15 | 74.47 | 75.85 (-0.4%) | 76.31 (+0.2%) | 79.22 (+4.0%) | The book's largest single holding at S is behaving exactly as a rate-supported bank core should: steady HOLD prints, no stop break, and the strongest 12-week outlook in the book at S, some 4.0% above today's close. Nothing to do but hold and let the trail work. OPS: SG-4 bank-core discipline - hold through extension, add only on a non-extended BUY-ADD reprint. Conviction oscillating 53/63 is normal range-bound noise, not deterioration. Trail HELD 74.47 (model live stop 74.17 is lower). L7: sgotr holds 7,700 D05 against the same pool - sg governs. |
| ME8U | Mapletree Industrial Trust | 1.92 | HOLD | — | 1.89 | 1.92 | 1.91 | 1.89 | A S industrial-REIT position drifting sideways with no actionable signal and no stop breach. The forward view is flat at best - the 12-week estimate of Ssits 1.6% BELOW today's S - so this is the closest thing in the book to a slow-motion sell case, but it is not there yet. OPS: NO TRADE cv0, 'Consensus conflict + low conviction'. Stop 1.89 NOT breached (close 1.92). L2 needs BOTH legs - leg 2 (negative 12wk) is satisfied at -1.56%, leg 1 (decisive stop break) is not. Re-arm trigger: any close below 1.89 makes this a mandatory L2 trim next EOD - watch it. SG-3 applies to the valuation: ignore the GROWTH_PRICED_IN band, read the MoS magnitude, which at -17.5 would amplify a trim one tier IF a signal ever originates one. |
| BUOU | FRASER L&C TRUST | 0.93 | HOLD | 0.93 | 0.96⚠ | 0.93 (+0.0%) | 0.93 (+0.0%) | 0.96 (+3.2%) | A S logistics-REIT stake sitting 3.1% under its stop for a ninth session, but the model still projects the price back up to Sover twelve weeks - a 3.2% recovery - and the signal is only a mild AVOID, not a sell. Oversold at RSI 30. Holding. OPS: SG-2/L2 leg 2 fails (+3.23% Est_12W, the largest positive 12wk limb among the breached names). First non-NO-TRADE print in four sessions. SG-3: MoS -5.89 is past the -3 amplify threshold, so IF a signal originates a trim the portion bumps one tier - but valuation cannot originate it. Trail HELD 0.96. |
| N2IU | Mapletree PanAsia Com Tr | 1.27 | HOLD | 1.27 | 1.31⚠ | 1.27 (+0.0%) | 1.28 (+0.8%) | 1.28 (+0.8%) | Closed exactly on its Sstop - a technical breach of 0.00% - which is not a decisive break by any reading. Forward paths are flat-to-slightly-positive and the position is small at SHold, watch the next close. OPS: second consecutive at-stop print. L2 leg 1 (decisive break) not met at 0.00%; leg 2 fails too (+0.79% Est_12W). Trail HELD 1.31. L7: sgreit carries 145,100 N2IU against the same pool - a far larger stake than sg's own; sg governs execution but any sgreit verdict on this name must not double-act. |
| S63 | ST Engineering | 10.64 | HOLD | — | 10.51 | 10.66 | 10.59 | 10.56 | No actionable signal, no stop breach, forward view essentially flat. A S position marking time. OPS: NO TRADE cv0, consensus conflict. Close 10.64 above the 10.51 trail. SG-3: MoS -10.38 would amplify a trim one tier if one ever originates. Trail HELD 10.51. |
| U13 | UOI LTD | 8.70 | HOLD | — | 8.03 | 8.69 | 8.71 | 8.68 | Structurally untradeable by the model - the liquidity screen rejects it every session - so this S insurance holding is held by default rather than by decision. No stop breach; forward view flat. OPS: NoTrade_Reason 'Liquidity gate failed', unchanged for weeks. Any exit here must be worked manually over multiple days, not sent as a single market order. Trail HELD 8.03. |
| HMN | CAPITALAND and ASCOTT STAPLE U | 0.86 | HOLD | 0.86 | 0.88⚠ | 0.86 (+0.0%) | 0.86 (+0.0%) | 0.88 (+2.3%) | An Srump position showing an 18-session 'stop breach' that is almost certainly an unrebased distribution rather than price damage, with a positive 12-week path. Too small to matter and there is no genuine exit case. OPS: SG-2 leg 2 fails (+2.92% Est_12W). Breach -2.84%. The 0.88 watermark has never been rebased for the distribution - operational follow-up carried from 08-24. Trail HELD 0.88. |
| 9CI | CapitaLand Investment | 2.66 | HOLD | — | 2.62 | 2.66 | 2.68 | 2.75 | Carries the strongest forward view in the book on a percentage basis - the 12-week estimate of Sis 3.4% above today's Sclose - with no stop breach. Small position, clear hold. OPS: NO TRADE cv0 (HOLD 54 on 08-26 did not persist). Close 2.66 above the 2.62 trail. Best Est_12W upside among held names; if the buy pool ever reopens this is the natural add candidate. Trail HELD 2.62. |
| A17U | CapitaLand Ascendas REIT | 2.43 | HOLD | 2.43 | 2.42 | 2.43 (+0.0%) | 2.44 (+0.4%) | 2.44 (+0.4%) | A token Sposition with a mildly negative label, no stop breach and a flat-to-positive forward view. Below any sensible trading size. Hold. OPS: AVOID 50, fifth consecutive session out of the sell pool. RallySell_BearTrend|sell carries an actionable -3 scorecard bias (n=9, hit.333) - the book's worst sell family - so even a label upgrade here would be haircut. SG-3: MoS -7.76 would amplify one tier. Trail HELD 2.42. |
Outcome & track record
Accuracy at the time of this record.
120 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.