Singapore — End-of-Day Verdict
21st cycle: the SG-2 discipline pays off four times in three sessions (BUOU/9CI/S63 sells all evaporate, 9CI while the price ran +4%); N2IU is the best buy setup in a week and dies on the same EPV anchor; O39 profit-take fires a third time and stays FROZEN
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 |
|---|---|---|---|---|---|---|---|---|---|
| N2IU | Mapletree PanAsia Com Tr | 1.33 | HOLD | 1.33 | 1.28 | 1.33 (+0.0%) | 1.33 (+0.0%) | 1.35 (+1.5%) | Mapletree PanAsia produced the book's best buy setup of the session -- a BUY - ADD at the top of the buy ranking on a pullback-to-VWAP entry, our most reliable family, priced right at the model level with no chase required and momentum still moderate. We are not adding, for two reasons. The intrinsic-value layer scores the units far above fair value and takes the maximum ten points off conviction, dropping it from and just under our 65 bar; and quarterly results land on 30 July, one session after a Federal Reserve decision that moves every Singapore REIT at once. Adding into that stack is not a risk worth taking when conviction is already short. We continue to hold the existing 10,000 units. Buy-limit 1.33 (entry range 1.32-1.34), stop 1.28, target 1.41, eD -0.0 (at the model entry, no extension), RSI 55.9, BuyRank #1. Quality gate is OFF here -- Q_Pass=F but this is an ADD to a held name, not a new entry. val_adj -10 from MoS -9.91, which is EPV-anchored (FV/sh 0.122 against the same row's DCF of 1.131 at a 1.33 price) -- systemic finding #1, now four candidates in two sessions. scorecard_bias +0 (PullbackBuy_VWAP|buy n=1, report-only). Had it cleared: 32.95% x 10,000 =, x0.5 size modulator (MoS <= -3) = (lot 100) SEst ~80% coverage: 1wk 1.33 [1.30-1.36] (0.0%) / 4wk 1.33 [1.26-1.42] (0.0%) / 12wk 1.35 [1.16-1.51] (+1.5%); Target_Price 1.41 is the trade's risk objective, distinct from the 4wk estimate. Horizon: MPACT 1QFY26/27 on 07-30 (+-3%, hold-through) stacked on the 07-29 FOMC S-REIT reaction (+-3.5%, tighten-trail). Trail held at 1.31 (the model stop has loosened to 1.28; keep the tighter line). RE-ARM: a BUY - ADD reprint holding cv >= 75 after MPACT prints, or a fairvalue run that moves N2IU off the EPV anchor. · news |
| C38U | CapitaLand Integrated Commercial Trust | 2.47 | WATCH | 2.45 | 2.39 | 2.47 (+0.8%) | 2.46 (+0.4%) | 2.46 (+0.4%) | CapitaLand Integrated printed a BUY on our preferred pullback entry, but it fails on every layer that matters for a brand-new position: it does not pass our quality screen, it trades far above any defensible estimate of intrinsic value, and its own 12-week projection sits marginally below today's close, so the model does not even forecast a gain. No entry -- the third straight rejection of this name, and the signal has decayed on its own after each decline. Quality gate BLOCKS the new entry (Q_Pass=F, F-Score 5, Quality_Rank 17). val_adj -10 (MoS -42.60, the deepest in the book; EPV 0.057 vs DCF 1.528). adj 61 - 10 = 51, well under the 65 floor. Buy-limit 2.45, stop 2.39, target 2.60, eD -0.42, RSI 61.7, BuyRank #2. Decay record: STRONG BUY cv97 on 07-23 -> WATCH cv50 07-24/07-27 -> BUY cv61 today. Est ~80%: 1wk 2.47 / 4wk 2.46 / 12wk 2.46 against a 2.47 close. CICT 1H results 08-05. |
| O39 | OCBC Bank | 28.93 | NO ACTION | 28.93 | 28.15 | 28.86 (-0.2%) | 28.43 (-1.7%) | 28.78 (-0.5%) | OCBC has run 38% year-to-date and roughly 18% in the past month, and today the case for banking half the gain hardened again: the signal collapsed outright, and the 12-week outlook now projects no further upside at all. Independent commentary agrees the shares are fully valued -- price/book of 1.9x against a 1.2x historical average and a forward P/E of 15.1x against 9.6x. This is the only holding where we know the real, so the profit-take is a genuine banked gain rather than a one. It is not being re-issued solely because the account has not been reconciled for twenty-one sessions. ARMED: real Avg_Cost 25.93, +11.56%, Gain_ATR 5.60 (>= 3.0), T12_Progress 100.5% (>= 90). TWO TICKS FIRED: (b) conviction collapse (-44, well past the 20-point threshold) and (c) dead forward view (28.78 < 1.02 x 28.93 = 29.51) -> profit_take_2 = 50% (500 of S). G1 N/A (NO TRADE cv0, not a cv>=80 breakout). MoS -0.63 > -3, no tier bump. FROZEN under SG-8 (21st cycle, zero change) -- carried behind the single sync flag, NOT re-published. Trail held at 28.15 (the armed 2.0xATR chandelier computes to 27.86 today; the ratchet does not step down). Q2 results 08-07. · news |
| BUOU | Frasers Logistics & Commercial Trust | 1.00 | NO ACTION | 1.00 | 0.97 | 0.99 (-1.0%) | 0.98 (-2.0%) | 0.99 (-1.0%) | Yesterday's loud sell signal on our largest REIT holding has already reversed. Frasers Logistics printed a STRONG SELL at the top of the sell ranking on Monday and is back to a constructive hold today, on a bullish consolidation pattern, at an unchanged price. This is exactly why we require a broken stop before acting on a Singapore sell label rather than trading the first print -- and it is the fourth such fade on this book in three sessions. Holding all 170,198 units and nudging the protective stop up. Position SSTRONG SELL - REDUCE cv76 (07-27) -> HOLD cv68 (07-28) in one session. Close 0.995 is +2.6% above the 0.97 model stop; 12wk 0.99 (-0.5%, still marginally negative) -- the SG-2 stop-break leg fails. MoS deepened -6.03 -> -6.25 but valuation cannot originate a sell (CLAUDE.md 9). SG-7 blocks the profit-take path (PROFIT_ARMED spurious, no Avg_Cost). Trail raised 0..97 (model line). ARM STANDS: a settled close < 0.97 fires a 50% trim immediately, escalating to 80% if the 12wk stays negative. News: the EUR294.9m German/Dutch logistics acquisition announced in May is still guided to complete 'by August 2026' with no dated confirmation -- an undated in-window catalyst. Logged optional so the counterfactual grader prices the restraint. |
| 9CI | CapitaLand Investment | 2.61 | NO ACTION | 2.61 | 2.44 | 2.62 (+0.4%) | 2.66 (+1.9%) | 2.67 (+2.3%) | CapitaLand Investment's strong sell signal from Monday has vanished entirely -- and the price rose 4% while it did. Had we acted on that print we would have sold into a rally. Holding. STRONG SELL - REDUCE cv71 (07-27) -> NO TRADE cv0 (07-28); close 2..61 (+4.0%), now +7.0% above the 2.44 stop line; 12wk +2.3% (2.67) positive -- both SG-2 legs fail by a wider margin than yesterday. MoS -1.76 > -3, nothing amplifies. Position S -- immaterial either way. Frozen SG-8 100% suspension. CLI 1H results 08-06 (+-4%, hold-through). |
| A17U | CapitaLand Ascendas REIT | 2.55 | NO ACTION | 2.55 | 2.57⚠ | 2.56 (+0.4%) | 2.56 (+0.4%) | 2.53 (-0.8%) | Ascendas REIT remains the only holding below its stop, but the breach has narrowed sharply -- from 3.5% on Monday to 0.8% today as the price recovered from 2..55. A shrinking three-quarter-percent gap is not a decisive break, and the position is worth about a thousand dollars. Holding. Close 2.55 vs the 2.57 recorded stop (-0.78%, narrowed from -3.5%); 12wk 2.53 (-0.8%) has now turned marginally negative, so both SG-2 legs are NOMINALLY met for the first time -- but L2 requires a DECISIVE close below stop, and a breach that is closing, on a NO TRADE cv0 row (liquidity gate failed), is not it. MoS -8.08 would amplify a valid exit but cannot originate one. Frozen SG-8 card. Stop held at 2.57 -- a settled close below 2.50 (a clean ~2.7% break) makes this decisive and fires the 100%. CLAR 1H 08-05. |
| ME8U | Mapletree Industrial Trust | 1.94 | NO ACTION | 1.94 | 1.90 | 1.94 (+0.0%) | 1.94 (+0.0%) | 1.92 (-1.0%) | Our largest REIT position drifted mildly negative on the 12-week view today, and the valuation gap widened to the deepest in the book. Neither is a sell case, and the forward-view wobble has a mundane explanation: Mapletree Industrial goes ex-dividend on 30 July for 3.1 cents, about 1.6% of the price, which mechanically accounts for the whole of the negative drift. The price is still comfortably above its stop. Holding all 118,279 units. Position SAVOID cv62 (firming from cv43 but still not a formal sell label); close 1.94 is +2.1% above the 1.90 trail -> the SG-2 stop-break leg fails. 12wk 1.92 (-1.0%). NEWS OVERRIDES the bearish read: ex-date 07-30 for SGD 0.031 (~1.6%) -- the estimator does not adjust for scheduled ex-dividend dates (systemic finding #4b), so this negative 12wk is an artifact rather than deterioration. MoS deepened -9.42 -> -17.72 but cannot originate a sell (CLAUDE.md 9); its EPV anchor is 0.104 against a DCF of 1.411 (finding #1). SG-7 blocks the profit-take path. Trail held 1.90. MIT's 1Q reported 23 Jul -- no in-window earnings event remains. · news |
| J69U | Frasers Centrepoint Trust | 2.28 | HOLD | 2.28 | 2.23 | 2.28 (+0.0%) | 2.30 (+0.9%) | 2.36 (+3.5%) | Frasers Centrepoint came alive today, jumping from no signal to the book's strongest hold on a bullish failed-breakdown reversal, with a 12-week view 3.5% above the current price -- the market's verdict on the third-quarter update released Monday. There is a momentum-divergence warning on the row, but we do not trim a name whose signal and forward view are both improving. Holding 28,000 units and raising the stop. Position SHOLD cv73 (NO TRADE -> cv73 in one session), the highest conviction on the held book. Exit_Warning carries RSI_DIV, but the 9c valuation-elevate path CANNOT trigger: MoS_FV is NA (no valuation coverage), so the required MoS < -3 pairing is unavailable -- and G3 blocks in any case (bullish signal, 12wk +3.5%). SG-7 blocks the profit-take path (PROFIT_ARMED spurious). Trail raised 2..23 (model line). FCT 3QFY26 first-session reaction was today. |
| C07 | Jardine C&C | 27.49 | HOLD | 27.49 | 25.99 | 27.44 (-0.2%) | 28.05 (+2.0%) | 28.45 (+3.5%) | Jardine Cycle & Carriage stays the only genuinely cheap name we own -- below our estimate of intrinsic value, with a 12-week view 3.5% higher -- and it reports first-half results on 30 July. Holding through the event. There is no case for pre-trimming into a catalyst on a name that is undervalued, above its stop, and carries no formal sell label. AVOID cv52 on RallySell_BearTrend, whose sell cell is actionably NEGATIVE (-3, n=9, hit 0.333) -- the label argues against itself here. Close 27.49 vs the 25.99 trail (+5.8% above); 12wk +3.5% (28.45). The book's only UNDERVALUED holding (MoS +0.64), so valuation argues a trim DOWN one tier, not up. Catalyst-aware hold-through: JC&C 1H 2026 on 07-30 (+-4%). Only S. |
| D05 | DBS Group Holdings | 73.91 | HOLD | 73.91 | 71.00 | 73.79 (-0.2%) | 74.04 (+0.2%) | 77.13 (+4.4%) | DBS pushed to another high and is deeply overbought going into Wednesday's Federal Reserve decision. We do not trim bank cores on an overbought reading alone, and the 12-week view still points 4% higher. Holding, not adding -- the entry is far too extended to chase -- and lifting the protective stop meaningfully to lock in the run. Position Sthe book's largest by value. Close 73.91, RSI 76.1, NO TRADE cv0 (consensus conflict + low conviction); 12wk +4.4% (77.13). SG-4/G3: no RSI-only trim, no chase. SG-7: PROFIT_ARMED is spurious (no Avg_Cost), so the profit-take layer cannot fire here. Trail raised 69..00 (2.5xATR chandelier). FOMC 07-29 (SG bank reaction 07-30, hold-through); Q2 results 08-06 pre-open (+-4%). |
| U11 | United Overseas Bank | 43.14 | HOLD | 43.06 | 41.93 | 43.02 (-0.1%) | 43.17 (+0.3%) | 44.51 (+3.4%) | UOB stays the best-behaved of our bank cores -- a constructive pullback setup, RSI a comfortable 60 while DBS prints 76, and a 12-week view 3.2% higher. It is a hold rather than an add: the label is HOLD, not BUY - ADD, so there is no add trigger to act on, and results land on 7 August. Position SHOLD cv60 (eased from 63) on PullbackBuy_VWAP; close 43.14 vs the 43.06 model entry (eD -0.10). 12wk +3.2% (44.51). MoS -0.60 (> -3, no amplify). Trail HELD at 41.93 -- the model line loosened to 41.53 and the ratchet does not step down. SG-4: add only on a non-extended BUY - ADD reprint. FOMC 07-29; Q2 results 08-07 (+-4%). |
| S68 | Singapore Exchange | 24.03 | HOLD | 24.03 | 23.18 | 24.06 (+0.1%) | 24.14 (+0.5%) | 24.18 (+0.6%) | SGX keeps strengthening -- conviction up again to 71, a new high for the move, with full-year results on 6 August. Holding and raising the stop underneath it, which is the right posture into a two-sided binary: the swing sits on the final dividend and cost guidance against an already-high bar. Position SHOLD cv71, firming for a second session); close 24.03, stop raised 23..18 (model line); 12wk +0.6% (24.18). MoS -1.94 (> -3, no amplify). Horizon stance for SGX FY2026 on 08-06 is tighten-trail, not trim (+-5% / +-3.06 ATR, the widest band in the window). |
| S63 | ST Engineering | 10.53 | HOLD | 10.53 | 10.33 | 10.59 (+0.6%) | 10.70 (+1.6%) | 10.56 (+0.3%) | ST Engineering's strong sell signal from Friday has now fully decayed to nothing across two sessions -- the cleanest illustration this week of why we require confirmation before trading a Singapore sell label. Holding above the stop. STRONG SELL - REDUCE cv71 (07-24) -> AVOID cv53 (07-27) -> NO TRADE cv0 (07-28). Close 10.53 is +1.9% above the 10.33 trail; 12wk +0.3% (10.56). MoS -9.23 is the SG EPV band artifact (SG-3), not a sell basis. No in-window event -- ST Engineering reports 19 Aug. |
| B61 | BT Sembawang Estate | 4.86 | HOLD | 4.86 | 4.77 | 4.85 (-0.2%) | 4.85 (-0.2%) | 4.89 (+0.6%) | Bukit Sembawang grinds higher on a steady constructive hold, still screening below our fair-value anchor, with no scheduled event in the next fortnight. Holding and inching the stop up. Position SHOLD cv52 (steady); close 4.86, stop raised 4..77 (model line); 12wk +0.6% (4.89). UNDERVALUED (MoS +0.18) though the quality screen fails (F-Score 3) -- irrelevant to a held name. March year-end; 1H due ~Nov, genuinely uncovered by the horizon. |
| VC2 | Olam Group | 1.39 | HOLD | 1.39 | 1.28 | 1.38 (-0.7%) | 1.36 (-2.2%) | 1.32 (-5.0%) | The Olam buy-stop that filled at Sfive sessions ago is up 7% and the valuation case is intact -- our discounted-cash-flow estimate is around Sagainst a Sprice. Holding, no add at RSI 70. Two things to watch: the 12-week estimate has turned negative for the first time since the fill, and the trade still is not recorded in the holdings file, which is why the monitor reports a phantom exit for a second straight day. WATCH cv60 on FailedBreakdown_Up_20D (up from NO TRADE); close 1.39, RSI 70.0. Trail held 1.28 (the 2.5xATR chandelier computes to 1.265 -- the ratchet does not step down). 12wk 1.32 is -5.0% vs the close, the first negative forward view since entry: if it persists AND the price closes below 1.28, SG-2 both-legs is met and a 50% trim fires. Est ~80%: 1wk 1.38 / 4wk 1.36 / 12wk 1.32. RECONCILE (SG-8 item 1): record @ 1.30 dated 2026-07-23 into input/sgportfolio.json -- the file still shows. |
| HMN | CapitaLand Ascott Trust | 0.91 | HOLD | 0.91 | 0.89 | 0.91 (+0.0%) | 0.90 (-1.1%) | 0.91 (+0.0%) | CapitaLand Ascott Trust reported first-half results today and the units firmed slightly. No signal either way and the price sits comfortably above the stop -- the hold-through worked. Holding. Close 0.91 (from 0.915) vs the 0.89 trail; NO TRADE cv0; 12wk flat (0.91). CLAS 1H 2026 released 07-28 -- the event is now behind us with no adverse reaction. Only S. |
| U13 | UOI Ltd | 8.60 | HOLD | 8.60 | 8.43 | 8.61 (+0.1%) | 8.60 (+0.0%) | 8.65 (+0.6%) | United Overseas Insurance has drifted up to 8.60 and remains permanently signal-muted by the liquidity gate -- it simply does not trade enough volume for the engine to form a view. Managing it on the trailing stop alone, which we raise today. Position SNO TRADE cv0 every session (liquidity gate / consensus conflict); close 8.60, trail raised 8..43 (2.5xATR chandelier); 12wk +0.6% (8.65). H1 print typically late-Jul/early-Aug but no date confirmable -- a standing horizon discovery gap. |
Outcome & track record
Accuracy at the time of this record.
97 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.
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.