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 |
|---|---|---|---|---|---|---|---|---|---|
| HMN | CapitaLand Ascott Trust | 0.83 | HOLD | 0.83 | 0.88⚠ | 0.83 (-0.6%) | 0.84 (+0.6%) | 0.86 (+3.0%) | Ascott broke to a fresh 52-week low on the strongest sell signal this book has produced in weeks, and we are still not selling it. Two independent reasons: the model's own 12-week view is +3.0% from here - this book's rule is never to act on a sell label alone without a negative forward view behind it - and the holding is Sso the trade would raise about Sbefore costs and change nothing about the portfolio's risk. Selling a hotel trust at an RSI of 27 into a projected bounce, for less than a rounding error of proceeds, is activity rather than discipline. Only floor-clearer tonight at adj 81 (bias 0, BreakdownDown_52W|sell n=1 report-only, printed not applied). Stop-breached -5.11% vs 0.88; below EMA20 0.8530 and EMA200 0.8690. SG-2 leg 2 fails. Stop HELD 0.88. Re-arm: a settled close with Est_12W turning negative makes this a live L2 exit. · news |
| U11 | United Overseas Bank | 41.26 | HOLD | 41.26 | 41.27⚠ | 41.21 (-0.1%) | 40.89 (-0.9%) | 40.44 (-2.0%) | UOB is the largest bank holding on the book and it stays untouched, because the line we published for cutting it was never reached. The shares closed a single cent under our stop - a rounding error, not a break - and then moved away from the trigger rather than toward it, while the signal itself recovered from a sell print during the day to a hold by the close. A 12-week view of -2.0% keeps us watchful, but cutting a half-million-dollar position on a line price did not reach would destroy the only discipline currently working here. Breach 0.02x ATR (vs 0.55x last night); close moved 40.99 -> 41.26 (+0.66%) while the label went STRONG SELL - REDUCE 37 intraday -> HOLD 44 at the close. L2 leg 2 met (Est_12W -1.99%), leg 1 effectively unmet. RE-ARM UNCHANGED: settled close <= 40.85 with Est_12W still negative arms the exit at the next open. Stop HELD 41.27. Far above EMA200 38.02, just under EMA20 41.34. US CPI bank leg released tonight 20:30 SGT; SGX first prices it Monday 14-Sep. · news |
| ME8U | Mapletree Industrial Trust | 1.91 | HOLD | 1.91 | 1.89 | 1.91 (+0.0%) | 1.90 (-0.5%) | 1.88 (-1.6%) | Mapletree Industrial is the one holding where the valuation case is genuinely extreme - priced further beyond its cash-flow anchor than anything else we own, by a factor of three - and where the model's 12-week view is finally negative. It still does not qualify for a trim, because the price has not broken its stop: 1.91 against a 1.89 line. Being expensive is not a reason to sell; being expensive while breaking down is. This is the name to watch. Deepest MoS on the book (-17.62) and the only held name pairing a negative Est_12W (-1.57%) with a deep negative MoS, but L2 leg 1 fails. Engine silent (NO TRADE cv0, 'Consensus conflict + low conviction'); PROFIT_ARMED is spurious per SG-7 (no Avg_Cost). Per CLAUDE.md SS9 valuation never originates a sell. ARM TRIGGER: one decisive settled close below 1.89 meets both L2 legs and makes a Sde-risk live. Stop HELD 1.89. · news |
| — | Keppel | — | NO ACTION | — | — | n/a | n/a | n/a | Keppel was the only buy signal on the board tonight and it is refused three times over. Its balance-sheet quality screen fails outright, which is a hard block on any new position; it is priced well beyond what its cash flows justify; and the model's own 12-week projection is slightly negative, so even on its own terms the trade has no destination. A first-ranked signal in a family this book has consistently lost money on is exactly the trade the scorecard exists to stop. Buy_Rank #1; adj 47 = raw 60 + val_adj -10 (MoS -3.0931 x 5, clamped) + bias -3 (PullbackBuy_VWAP|buy n=6 hit.333 ACTIONABLE) - 18 points below the 65 floor. Quality_Pass FALSE and BN4 is unheld, so the gate is hard. Entry 11.54 is +0.50 ATR ABOVE the 11.43 close - a break-up, not a pullback. Note the inversion vs 09-10: L4 now supports the BUY (WATCH 44 -> BUY 60 at the settled close) and it is refused on substance instead. RE-ARM: settled-close BUY reprint with Quality_Pass TRUE (not overridable) AND raw conviction >= 78 AND Est_12W >= +1.0% vs entry, at the then-live Entry_Price - no carried limit. |
| O39 | OCBC Bank | 31.60 | HOLD | 31.60 | 31.30 | 31.61 (+0.0%) | 31.79 (+0.6%) | 33.01 (+4.5%) | OCBC is our best-performing holding and we are leaving it alone. It has run far enough past cost to qualify for profit-taking, but nothing has actually deteriorated: the signal held, the 12-week view improved, and there is no warning on the tape. The rule here is that a big gain on its own is never a reason to sell - something has to go wrong first, and nothing has. ADR-0012 re-check on the only holding with a real Avg_Cost (25.93): ARMED (Gain_ATR 12.29, T12_Progress 95.73, +21.85%) with ZERO of four ticks - (a) PROFIT_ARMED alone, no second code; (b) conviction = -7, short of the 20-pt bar; (c) Est_12W 33.01 > 1.02 x 31.60 = 32.23, alive and improved from 32.81; (d) HOLD, not NO TRADE. Trail-raise assessed and DECLINED: chandelier 30.678 and model stop 30.68 both below the carried 31.30. Stop HELD 31.30 - tightest line on the book at 0.95% under the close. +4.29% since 10-Aug, the best held performer in the window. |
| C07 | Jardine Cycle & Carriage | 27.38 | HOLD | 27.38 | 26.44 | 27.39 (+0.0%) | 27.78 (+1.5%) | 27.60 (+0.8%) | Jardine C&C is the one cheap, quality-passing name we own, it is holding its ground, and the only change tonight is that we tighten the protective line underneath it. The stop moves up because the model's own level has risen above where ours sat - we are adopting its line, not inventing one - which locks in a little more of the position without touching the shares. Stop RAISED 26.33 -> 26.44: model Stop_Price prints 26.44 on a long-side FailedBreakdown_Up_20D / HOLD cv44, and the 2.0x ATR chandelier computes to 27.38 - 0.9402 = 26.4398 - both above the carried line. Close 27.38, stop now 3.43% below. UNDERVALUED, MoS +0.6201, Quality_Pass TRUE, Est_12W +0.80%. Position SNOT an ADR-0012 action: C07's PROFIT_ARMED is spurious under SG-7 (no Avg_Cost) and no profit-take was evaluated. Indonesia auto sales 09-15 and BI rate decision 09-17 are the read-throughs in the window. |
Outcome & track record
Accuracy at the time of this record.
131 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.