Singapore — End-of-Day Verdict
29th cycle: a broad −0.51% STI risk-off day guts the acquire side entirely — D05's queue is AUTO-WITHDRAWN as conviction collapses cv90 → cv60 one session before the 81c ex-dividend, leaving **NO ACQUIRE for the first time in seven cycles**; O39's profit-take is DOWNGRADED 50% → 30% as two of its three ticks clear; HMN hits **SG-8 cycle 4 and converts to a blocking holdings-sync flag**; U11 breaks its stop on the book's largest position; and the newly-actionable `NoTrade|sell` cell prints **−10 at 0/6**, a direct haircut on B61's carried card
Recommendations
0 to acquire · 2 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 |
|---|---|---|---|---|---|---|---|---|---|
| B61 | BT Sembawang Estates | 4.53 | SELL 30% | 4.53 | 4.62⚠ | 4.52 (-0.2%) | 4.52 (-0.2%) | 4.56 (+0.7%) | We are cutting a third of our Bukit Sembawang holding because the shares have now closed below our risk line for four straight sessions and the trend has gone flat: the stock has traded in a nine-cent band around Sall week, sits below its 20- and 50-day averages, and the model sees essentially no movement over the next month. The company itself is fine and the shares still screen as cheap on intrinsic value, which is exactly why we are trimming the minimum 30% rather than exiting - this is risk discipline on a broken stop, not a verdict on the business. OPERATIONAL: SG-1 cycle 3 / 3rd publication of an unexecuted card (08-07, 08-11, 08-12); SG-8 threshold REACHED - if unexecuted at the next EOD this converts to a blocking holdings-sync flag and is NOT re-published. Scorecard caveat stated openly: NoTrade|sell became ACTIONABLE tonight at -10 (n=6, hit_1w 0.000, hit_4w 0.000) - every NoTrade-basis sell this book has graded was premature; L2's negative-12wk limb also FAILS (Est_12W 4.56 = +0.7%). Card is therefore published at the 30% minimum with NO escalation and no tranche-2 re-arm. val_adj corrected +4 -> +2 (Quality_Pass False -> halve). Residual 18,400, stop held 4.62. Sdividend (ex 08-05, pays 08-15) = Son the pre-trim 26,300, unaffected by the sale. · news |
| O39 | OCBC Bank | 31.19 | SELL 30% | 31.19 | 29.73 | 31.07 (-0.4%) | 30.66 (-1.7%) | 31.33 (+0.4%) | We are banking a third of our OCBC gain after a 20% run. The bank's half-year results were genuinely excellent - record profit up 13% and a dividend raised 15% - but the share price has already paid for all of it: the stock is overbought, our own 12-week projection now sits below today's price, and one Yahoo Finance piece asks outright whether OCBC is fully valued after the surge. We keep and a raised trailing stop, so we stay long the franchise while taking real money off the table. OPERATIONAL: ADR-0012 profit-take, ARMED on a REAL Avg_Cost 25.93 (SG-7 sanctions the fire here and only here) - Gain_ATR 7.13, T12_Progress 99.55, +20.27%. ONE tick fires: (c) dead 12wk, Target_12Week 31.33 < 1.02 x 31.19. DOWNGRADED 50% -> 30% vs the carried 08-11 card because ticks (a) RSI_DIV and (b) conviction-collapse have BOTH CLEARED in one session (conviction rose, Exit_Warning is now bare PROFIT_ARMED) - profit_take_1 governs at 1 tick; L4 makes tonight's settle authoritative, and re-publishing 50% would be trading yesterday's tape. G1 not engaged (not a fresh cv>=80 breakout family); tick (c) fires regardless. MoS -0.608 is far from <= -3, so no tier bump. Banked S; residual 700; stop HELD 29.73 (live chandelier 29.71 - never lower). Forfeits Sof the 08-17 47c dividend, down from Sat the old size - disclosed, not decisive. · news |
| HMN | CapitaLand Ascott Trust | 0.86 | HOLD | 0.86 | 0.88⚠ | 0.86 (+0.0%) | 0.86 (+0.0%) | 0.87 (+1.2%) | No new decision on Ascott Trust tonight - the exit we called three times running has still not been executed, so re-deciding it a fourth time would be noise. The position remains one to sell in full; what is needed now is for the holdings file to be brought up to date, not another recommendation. OPERATIONAL: SG-8 FIRES at cycle 4 - card WITHDRAWN from the verdict and replaced by a single blocking 'holdings not synced - reconcile input/sgportfolio.json' flag. Decision NOT reversed, size NOT re-escalated (remains / 100% / Sa de-minimis stub). Technical case has STRENGTHENED: BreakdownDown_20D, STRONG SELL - REDUCE cv33, SellRank #2, 6th close below the 0.88 stop, below EMA20/50/200 (0.872/0.874/0.873), STRONG SELL at 10:00, 15:00 and the settle. Est_12W 0.87 = +1.2%, dead flat. Coliwoo Midtown S acquisition + STI Reserve List remains constructive and remains completely unpriced after six sessions - not an override. Re-arm on a reclaim of 0.88 with a bullish family print. Becomes actionable again the moment a fill is reflected. · news |
| U11 | United Overseas Bank | 41.95 | HOLD | 43.24 | 42.15⚠ | 41.91 (-3.1%) | 41.97 (-2.9%) | 42.09 (-2.7%) | UOB slipped below our risk line today, but we are holding. The move was part of a broad 0.5% fall across the whole Singapore market rather than anything specific to the bank, the shares are still above their 50- and 200-day averages, and our 12-week view remains positive. The break is also only half a percent deep - and next Monday's 88-cent dividend will mechanically reset that risk line 1.6% lower anyway. This is our largest holding, so it is the one we watch most closely. OPERATIONAL: stop BROKEN 41.95 vs 42.15 (-0.47%, 1st session) but SG-2 FAILS - Est_12W 42.09 is +0.3% POSITIVE, so L2's negative-12wk limb is unmet; break is not decisive; conviction RECOVERED; Entry_Distance_ATR +1.60 puts price well BELOW the model entry (pullback, not breakdown); holds above EMA50 41.54 and EMA200 38.03, only below EMA20 42.83; RSI 44.2 mid-range. MANDATORY EX-DATE REBASE: stop 42.15 -> 41.27 ON 08-17 (88c interim, -2.09%, confirmed by horizon + Yahoo) - left unrebased this prints a phantom breach on Sof stock, the identical B61 08-05 failure. WATCH TRIGGER: a close below 42.15 before 08-17 on a NEGATIVE Est_12W makes U11 the immediate lead dispose. · news |
| J69U | Frasers Centrepoint Trust | 2.21 | HOLD | 2.21 | 2.18 | 2.21 (+0.0%) | 2.21 (+0.0%) | 2.26 (+2.3%) | The model is shouting sell on Frasers Centrepoint for a ninth session and we are still not acting on it. The shares are holding above the level at which we would cut, and our own 12-week projection is more than 2% higher than today's price - so the label is not backed by either the price or the forecast. This book's track record on exactly this kind of sell signal is a coin flip at best, which is why we require confirmation before selling a Sposition. It is, however, the closest this has come. OPERATIONAL: STRONG SELL - REDUCE cv69, SellRank #1, portion 80.07% - SG-2 requires stop-break AND negative 12wk; close 2.21 is ABOVE the recorded 2.18 stop and Est_12W 2.26 is +2.3% positive, so BOTH limbs fail. RallySell_BearTrend|sell is ACTIONABLE at -3 (n=9, hit.333) -> adj 66; SG-5 rejects trusting an SG sell label alone. TWO GENUINE NEGATIVES: close is now below ALL of EMA20/50/200 (2.245/2.252/2.247) for the first time in the window, and the label FIRMED intraday (AVOID cv46 -> STRONG SELL cv62 from 15:00 -> cv69 at the settle). Stop held 2.18 (live model 2.15 - never lower). TRIGGER: a close below 2.18 with a negative Est_12W makes this an immediate dispose. |
| A17U | CapitaLand Ascendas REIT | 2.49 | HOLD | 2.49 | 2.49 | 2.49 (+0.0%) | 2.51 (+0.8%) | 2.51 (+0.8%) | Holding Ascendas REIT, with one piece of housekeeping due at tomorrow's open: the trust pays out 7.5 cents tomorrow, which mechanically knocks about 3% off the quoted price. We move our risk line down by the same amount so the payout is not mistaken for a fall. OPERATIONAL: MANDATORY EX-DATE REBASE - stop 2.49 -> 2.42 ON 08-13 (1H DPU 7.482c on a 2.49 close, -3.0%, quantum CONFIRMED in decisions/sg_horizon.json). Left unrebased it prints ~2.42 and trips a phantom breach - the identical B61 08-05 failure that cost this book a false alarm. NO TRADE cv0 ('Consensus conflict + low conviction') on all six of today's prints; Est_12W 2.51 = +0.8%. |
| C38U | CapitaLand Integrated Commercial Trust | 2.49 | WATCH | 2.48 | 2.43 | 2.49 (+0.4%) | 2.50 (+0.8%) | 2.50 (+0.8%) | CICT was the only stock in our whole Singapore universe still flashing a buy today, and we are not taking it. It fails our financial-health screen, it is the most expensive name on our board relative to what we think it is worth, and its half-year results landed today with the shares drifting lower rather than rallying. Any one of those would give us pause; together they are a clear no. OPERATIONAL: BUY cv60, BuyRank #1, 34.0m shares - HARD BLOCK, Quality_Pass=False on an UNHELD name, so the gate bites. Would fail independently: MoS_FV -43.12, the most extreme in the book -> val_adj -10 -> adj 50, FIFTEEN under the 65 floor. 1H 2026 results landed 08-12 per the horizon file and the tape faded 2.50 -> 2.49 - the print produced no bid. Three independent reasons, one verdict. · news |
| C07 | Jardine Cycle & Carriage | 27.88 | HOLD | 27.88 | 26.38 | 28.05 (+0.6%) | 28.59 (+2.5%) | 28.51 (+2.3%) | Jardine Cycle & Carriage is the best value on our Singapore board - the only holding our intrinsic-value work flags as an outright bargain - and it came within four points of qualifying as a buy tonight. It did not quite get there, so we hold our existing and wait for a better-shaped entry rather than reaching. OPERATIONAL: adj 61 vs the 65 floor - the closest miss of the night. Only Bargain=True name in the book; UNDERVALUED, MoS_FV +0.616, Quality_Pass True -> val_adj +10, the book's largest positive valuation adjustment. L1 warns FailedBreakdown_Up_20D is a weak chase family (~0.25 hit, ~60% stop-first), so the 4-point shortfall is not one to reach across. RE-ARM: a PullbackBuy_VWAP reprint at cv>=56 clears 65 on the +10 alone. Stop 26.38. NOTE J36 goes ex-dividend 08-20 - correlated drag on JC&C. |
| S68 | Singapore Exchange | 24.96 | HOLD | 24.96 | 24.01 | 24.96 (+0.0%) | 24.84 (-0.5%) | 24.69 (-1.1%) | SGX opened strong and gave it all back, finishing lower on the day. We are holding, but the forward view has turned: our 12-week projection now sits just below the current price, the only bank or exchange holding where that is true. No action while the risk line is intact, but this one is no longer working for us. OPERATIONAL: printed BUY - ADD cv77 at the 10:00 open then decayed all session ->, close 25.22 -> 24.96 - a one-print open spike that faded is exactly what L4 says to discount. Est_12W 24.69 = -1.1% vs close. Stop HELD 24.01 (live chandelier 23.83 - never lower). Singtel Q1 update 08-18 reads across via the index. |
| 9CI | CapitaLand Investment | 2.73 | HOLD | 2.73 | 2.62 | 2.72 (-0.4%) | 2.73 (+0.0%) | 2.81 (+2.9%) | CapitaLand Investment reports half-year results tomorrow and has the healthiest forward outlook of our property holdings, so we hold through the announcement rather than pre-emptively trimming into it. OPERATIONAL: catalyst-aware hold - 1H 2026 results 08-13 CONFIRMED in the horizon file; no pre-trim, trail through the print, trim only on a confirmed lower high or a decisive stop break. HOLD cv53; Est_12W 2.81 = +2.9%, the best forward view in the held REIT/property complex. Stop 2.62. |
| S63 | ST Engineering | 10.28 | HOLD | 10.28 | 9.58 | 10.31 (+0.3%) | 10.42 (+1.4%) | 10.24 (-0.4%) | ST Engineering reports half-year results tomorrow. The shares have been drifting and our 12-week view is now fractionally below today's price, but we do not sell into a scheduled announcement - we hold through it and let the numbers decide. OPERATIONAL: catalyst-aware hold - 1H 2026 results 08-13 CONFIRMED. NO TRADE cv0 on all six prints, RSI 42.98, below EMA20 10.42 and EMA50 10.57. Est_12W 10.24 vs a 10.28 close = -0.4%, the weakest forward view in the book after S68. If the results land badly this becomes an L2 candidate at the 9.58 stop. |
| F34 | Wilmar International | 3.94 | WATCH | 3.93 | 3.83 | 3.95 (+0.5%) | 3.95 (+0.5%) | 3.92 (-0.3%) | Wilmar reported half-year results today and our signal on it weakened rather than strengthened, dropping it out of contention entirely. We were already close to passing on it yesterday; today's reaction settles it. OPERATIONAL: BUY cv70 (adj 60, near-miss) on 08-11 -> WATCH cv53 (adj 43) tonight after 1H 2026 results landed 08-12. No longer in the buy pool. Est_12W 3.92 is still -0.5% BELOW the 3.94 close - a negative forward view on a prospective new entry. RE-ARM on a cv>=80 reprint post-results with a non-negative 12wk. |
| S58 | SATS | 4.71 | WATCH | 4.73 | 4.59 | 4.70 (-0.6%) | 4.70 (-0.6%) | 5.09 (+7.6%) | SATS is not a buy today, but it is the best-shaped candidate we do not own: it passes our quality screen, screens as a bargain, is not chasing a rally, and carries the strongest 12-week upside of anything on our unheld list. Worth watching for an entry signal. OPERATIONAL: WATCH cv53 - not in the buy pool, no action. Entry_Distance_ATR +0.18 (below the model entry, NOT extended); Quality_Pass True; Bargain True; Est_12W 5.09 = +8.1%, the healthiest forward view of any unheld name. The 08-05 buy-stop @4.85 was withdrawn 08-06 and remains unfilled. Watch for a BUY reprint. |
Outcome & track record
Accuracy at the time of this record.
112 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 →
0.162 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.