Singapore — End-of-Day Verdict
22nd cycle: a broad pre-FOMC melt-up flips FIVE names bullish at once; N2IU upgrades to STRONG BUY - ADD cv96 and becomes the book's FIRST acquire in 7 sessions (QUEUED @1.33); C38U prints STRONG BUY cv97 at BuyRank #1 and is blocked at the quality gate; both loud sells fail SG-2 for the fifth and sixth time
Recommendations
1 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 | 1.36 | ADD — QUEUED @1.33 | 1.33 | 1.28 | 1.36 (+2.3%) | 1.36 (+2.3%) | 1.35 (+1.5%) | Mapletree PanAsia has strengthened for three straight sessions on the pullback-buy setup we trust most, upgrading from a hold to the book's strongest buy signal while the units rose only 2.3% - so we are adding to a position we already own at the model's own entry price rather than chasing the close. The trust's flagship VivoCity mall is compounding well (net property income +7.6%, rents re-signing 14% higher, effectively full occupancy) and Hong Kong's Festival Walk is finally stabilising with tenant sales and footfall both up. The fair-value screen reads the units as expensive, but that reading rests on an earnings-power model that values them at 12 cents against its own discounted-cash-flow estimate of 1.13, so we halve the size rather than stand aside. Buying below the market with a defined 3.8% stop caps the risk on the new tranche at about SOPS: GTC limit 5,700u @1.33 (entry range 1.32-1.34), 114.38% x 10,000 x 0.5 size-mod, lot 100; qty_after 15,700. eD -1.14 extended -> queued not bought (L1 family favoured, SG-6 shallow -2.2% ~1 ATR limit, NOT a deep-pullback leak). MPACT 1Q 07-30 into the FOMC S-REIT reaction (+-3.5% tighten-trail) - likely fill scenario is an adverse print; the 1.28 stop is the condition of the trade. Withdraw on a DPU cut or raise; reprice if the 07-30 open gaps >1.5%. 12wk 1.35 exceeds DCF anchor 1.131 - growth beyond priced-in. Existing 10,000u trail unchanged 1.31. expires 2026-08-03. · news |
| C38U | CapitaLand Integrated Comm Tr | 2.51 | NO ACTION | 2.50 | 2.36 | 2.51 (+0.4%) | 2.51 (+0.4%) | 2.49 (-0.4%) | CICT printed the strongest buy signal on the book and we are still not buying it. The quality screen fails outright on a name we do not yet own, and the valuation work puts it further beyond fair value than anything else we track - its own twelve-week projection sits below today's price. Newly acquired Paragon mall is under enhancement review, which is constructive long-term but does nothing for the entry case now. Fourth straight rejection while the units ran 2..51. OPS: adj 87 = highest on book, blocked SOLELY by the hard quality gate (Quality_Pass F, F-Score 5, Q-Rank 17) on a new entry (, gate not waived). Trail: WATCH cv50 07-27 -> BUY cv61 07-28 -> STRONG BUY cv97 07-29. CICT 1H is Aug 12 per issuer (horizon file wrongly carries 08-05 - finding #8). · news |
| BUOU | FRASER L&C TRUST | 1.00 | NO ACTION | 1.00 | 0.97 | 1.00 (+0.0%) | 0.99 (-1.0%) | 1.00 (+0.0%) | Frasers Logistics flipped from a strong sell to a strong buy in two sessions, and that whipsaw is itself the reason to wait. The trust reports on 30 July - the very day we would buy - and it has a well-established habit of European acquisitions with discounted share placements, one of which could knock 3-5% off the units overnight. Its own four-week projection sits below today's price, contradicting the buy signal on the same row. We already own a large stake and our remaining cash would only fund a 2.6% top-up, so the patient course is to see the results first. OPS: adj 80 clears the 65 floor but FAILS the ACT-NOW stability leg on 4 counts - (i) STRONG SELL cv76 07-27 -> HOLD cv68 -> STRONG BUY cv90 (SG-2/SG-5 label instability), (ii) FLCT results 07-30 = execution date + placement risk (news override), (iii) Est_4W 0.99 < 1.00 action price vs Target_Price 1.11 (finding #5), (iv) L1 discounts BreakoutUp_52W (cell n=2, hit_1w 0.0). Model size 97,336u vs ~S effective cash -> ~4,500u ~S = 2.6% of the Sstake. RE-ARM: post-results BUY - ADD reprint holding cv>=75, settled close >=1.00, no placement. Trail held 0.97 (the 50% fire line). · news |
| S68 | Singapore Exchange | 24.03 | NO ACTION | 23.98 | 23.19 | 24.04 (+0.3%) | 24.10 (+0.5%) | 24.25 (+1.1%) | SGX is technically the cleanest setup on the book - the pullback entry we trust most, barely extended, momentum mid-range, and firming for three sessions - and it misses our buy bar by five points purely on valuation. The margin-of-safety shortfall is small in substance but the model applies its maximum penalty to it, so the add does not happen this session. Full-year results on 6 August are the next real test. OPS: adj 60 vs the 65 floor. val_adj -10 generated from a MoS of only -1.94 (FV/sh 8.16 vs same-row DCF 13.26) - this is finding #1b in its purest form: the x5 negative scaling saturates the -10 clamp at MoS <= -2, so S68 at -1.94 gets the same haircut as C38U at -42.59. eD -0.13, RSI 56.1, HOLD cv62 -> cv71 -> BUY - ADD cv70. SGX FY2026 08-06, horizon band +-5%, stance tighten-trail. Trail raised 23..19. |
| S58 | SATS | 4.73 | NO ACTION | 4.59 | 4.46 | 4.71 (+2.6%) | 4.70 (+2.4%) | 5.03 (+9.6%) | SATS is both too expensive on our valuation screen and already 1.3 ATR past the price we would be willing to pay, so the setup fails on entry discipline as well as on the numbers. It has faded once this week already. OPS: adj 50 (60 - 10 on MoS -2.40). eD -1.31 = extended, RSI 69.8. Quality_Pass True so the gate is not the issue. Second rejection in three sessions: cv73 07-27 -> cv0 07-28 -> cv60 today. |
| O39 | OCBC Bank | 29.78 | NO ACTION | — | 28.56 | 29.70 | 29.25 | 29.58 | OCBC is up 14.8% on our cost and the case for banking half of it has now fired on four separate occasions. The shares trade at 1.9x book against a 1.2x historical average and 15.1x forward earnings against 9.6x, momentum is diverging at a 78 relative-strength reading, and the model's own twelve-week projection no longer sits above today's price. The trim is not being re-issued only because the holdings file has not been reconciled for 22 sessions - this is a bookkeeping blockage, not a change of view. OPS: PROFIT-TAKE ARMED + FIRED, 4th distinct tick set - (a) RSI_DIV, (c) dead 12wk 29.58 < 1.02 x 29.78 = 30.38, (d) 2 consecutive NO TRADE prints. Real Avg_Cost 25.93, Gain_ATR 6.33, T12_Progress 100.68% - the ONLY genuinely armed row under SG-7. G1 n/a (cv0, not a breakout). MoS -0.63 > -3 so no tier bump: profit_take_2 = 50% = S @29.78. NOT RE-PUBLISHED - SG-8 freeze, 22nd cycle. Trail raised 28..56 (armed 2.0xATR). GE Q2 08-04 pre-read, OCBC Q2 08-07. · news |
| 9CI | CapitaLand Investment | 2.66 | NO ACTION | — | 2.52 | 2.66 | 2.70 | 2.78 | CapitaLand Investment has now flashed a sell signal three times in three sessions while the shares rose 6% through every one of them. The stock sits 4.3% above the level that would confirm a genuine breakdown and its twelve-week outlook is still 4.5% higher than today's price, so selling here would be selling strength. Singapore sell signals on this book reverse more often than not, and we require a confirmed break before acting. OPS: STRONG SELL - REDUCE cv79 SellRank #1, Portion 80.07% ( S), RSI 72.3. BOTH SG-2 legs fail wider than yesterday: close 2.66 = +4.3% above the 2.55 model stop; 12wk 2.78 = +4.5% POSITIVE. Cell ReversalDown_Overbought|sell n=1 -> report-only, bias stated not applied. MoS -1.76 > -3, nothing amplifies. Sequence cv71 (07-27) -> cv0 (07-28) -> cv79 (07-29), price 2...66. Frozen SG-8 100% suspension unchanged; trail raised 2..52. CLI 1H 08-06 (+-4%). |
| S63 | ST Engineering | 10.72 | NO ACTION | — | 10.33 | 10.75 | 10.88 | 10.75 | ST Engineering flashed a sell while the shares were rising, and it sits nearly 4% above the level that would confirm a real breakdown with a twelve-week outlook still marginally positive. The signal family behind it is a coin flip over 72 graded trades, so on its own it is not a reason to sell a Sposition. The stock does look richly valued, which would deepen a trim once a genuine sell case exists - it cannot create one. OPS: STRONG SELL - REDUCE cv67 SellRank #2, Portion 80.07% ( S). Both SG-2 legs fail: close 10.72 = +3.9% above the 10.32 stop; 12wk 10.75 = +0.3% positive. Cell +0, n=72, hit 0.50 (most-sampled on the book). MoS -9.23 <= -3 would bump a valid trim one tier but cannot originate (section 9). Re-fired from cv0 while price rose 10..72. LARGEST declined trim by dollar value - WATCH: a settled close below 10.32 with a negative 12wk meets SG-2 on both legs and fires 80%. Trail held 10.33. |
| ME8U | Mapletree Industrial Trust | 1.96 | HOLD | — | 1.90 | 1.96 | 1.95 | 1.95 | Mapletree Industrial is our largest property holding and it is holding above its trailing stop with no sell signal. Its slightly negative twelve-week reading is an artefact: the trust goes ex-dividend on 30 July for 3.1 cents, about 1.6% of the price, which our projection does not adjust for. The valuation screen reads it as the most expensive name we own, but expensive alone has never been a reason to sell a name that is trending. OPS: AVOID cv0 (decayed from cv62), no formal sell label. Close 1.96 = +3.2% above the 1.90 trail -> SG-2 stop-break leg fails. MoS -17.72 deepest on book but cannot originate (section 9). 12wk 1.95 (-0.5%) explained by the 07-30 ex-div S - finding #6. Trail held 1.90. S. |
| A17U | CapitaLand Ascendas REIT | 2.58 | HOLD | — | 2.57 | 2.59 | 2.59 | 2.58 | Ascendas REIT has climbed back above its stop and the breach that ran for two sessions is now fully closed - the units recovered from 2..58 in three days. The exit we declined twice is vindicated again. OPS: monitor reports ZERO breaches (close 2.58 vs the 2.57 stop, recovered 2.48 07-.55 07-.58). 12wk flat 2.58 (0.0%). Both SG-2 legs now fail where yesterday both were nominally met. MoS -8.08 cannot originate. Frozen SG-8 card, trail held 2.57. The <2.50 decisive-break trigger STANDS. CLAR 1H 08-05. |
| J69U | FRASERS CPT TR | 2.28 | HOLD | — | 2.23 | 2.28 | 2.29 | 2.35 | Frasers Centrepoint lost its signal without losing any price - the units closed unchanged at 2.28, comfortably above our trailing stop, with a twelve-week outlook 3.1% higher. A momentum divergence flag is present but the trust has no valuation coverage, so there is nothing to pair it with and no sell case. OPS: HOLD cv73 -> AVOID cv0 on NoSignal at an unchanged 2.28 close - conviction collapse without a price move or a sell label. RSI_DIV present but the section 9c elevate CANNOT trigger (MoS_FV is NA, so the MoS < -3 pairing is unavailable); 12wk +3.1%. Close +2.2% above the 2.23 trail. Trail held 2.23. S. |
| C07 | Jardine C&C | 28.10 | HOLD | — | 26.63 | 28.02 | 28.61 | 28.93 | Jardine Cycle & Carriage rose 2.2% and its sell signal disappeared entirely. It is the only holding our valuation work calls outright cheap, with a twelve-week outlook 3% above today's price, and it reports half-year results tomorrow - we hold through the event rather than pre-trimming into it. OPS: AVOID cv52 -> NO TRADE cv0 (Consensus conflict + low conviction) while price ran 27..10. 12wk +3.0% (28.93). Only UNDERVALUED holding (+0.64) so valuation argues any trim DOWN one tier. JC&C 1H 07-30 (+-4%) -> catalyst-aware hold-through. Trail raised 25..63. S. |
| D05 | DBS Group Holdings | 75.00 | HOLD | — | 72.07 | 74.61 | 74.79 | 77.17 | DBS closed at a new high of Swith a twelve-week outlook still 2.9% higher. Momentum is stretched at a 79 reading, but an overbought gauge on its own has never been a reason to sell a bank that is trending - we raise the trailing stop instead and hold into 6 August results. OPS: RSI 79.2 at a new high, 12wk 77.17 (+2.9%). SG-4/G3 - hold through the extension, no RSI-only trim, no chase. Trail raised 71..07 (2.5xATR chandelier; model stop prints 72.21). DBS Q2 08-06 pre-open, +-4%. S. |
| U11 | United Overseas Bank | 43.89 | HOLD | — | 42.33 | 43.91 | 43.99 | 45.67 | UOB firmed to a 71-conviction hold with a twelve-week outlook 4.1% above today's price and momentum at a comfortable 65 - far less stretched than DBS. We raise the stop and wait for a non-extended add signal rather than buying into the strength. OPS: HOLD cv60 -> cv71 on FailedBreakdown_Up_20D, close 43.89, RSI 65.1 vs DBS's 79.2. 12wk 45.67 (+4.1%). SG-4: add only on a non-extended BUY - ADD reprint. Trail raised 41..33 (model stop; chandelier computes 41.94). UOB Q2 08-07 (+-4%). S. |
| B61 | BT SEMBAWANG EST | 4.89 | HOLD | — | 4.78 | 4.88 | 4.89 | 4.91 | Bukit Sembawang firmed to a 63-conviction hold at 4.89 and remains one of only two holdings our valuation work calls cheap. No action beyond ratcheting the stop up. OPS: HOLD cv52 -> cv63. Trail raised 4..78 (2.5xATR chandelier). S. |
| HMN | CAPITALAND and ASCOTT STAPLE U | 0.90 | HOLD | — | 0.89 | 0.90 | 0.90 | 0.91 | Ascott Trust recovered to a modest hold after its half-year results landed benignly on 28 July. A small position; no action. OPS: NO TRADE cv0 -> HOLD cv43, close 0.905. Trail held 0.89 (chandelier computes 0.883, model stop 0.88). S. |
| U13 | UOI LTD | 8.60 | HOLD | — | 8.43 | 8.61 | 8.60 | 8.65 | UOI is permanently gated out of the signal engine by thin trading volume, so it is managed on the stop alone. Unchanged at 8.60. OPS: NO TRADE cv0, NoTrade_Reason 'Consensus conflict + low conviction'; liquidity-gated. Trail held 8.43 (chandelier 8.427). S. |
| VC2 | Olam Group Ltd. | 1.40 | HOLD | — | 1.28 | 1.39 | 1.37 | 1.33 | Olam is up 7.7% since we bought it and our valuation work still calls it cheap, but its twelve-week outlook has now been negative for two straight sessions - the first sustained warning since the fill. Not a sell yet; it would take a close below 1.28 to make the case. The position is still missing from the holdings file. OPS: NO TRADE cv0, NoTrade_Reason 'Liquidity gate failed', RSI 70.9, close 1.40 vs the 1.30 fill (+7.7%). 12wk 1.33 = -5.0% below the close, 2nd consecutive negative session. If it closes below 1.28 with the 12wk still negative, SG-2 both-legs is met and a 50% trim fires. Trail held 1.28. UNRECORDED: @1.30 filled 2026-07- file still shows - phantom 'exit' in the monitor for a 3rd day. |
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.