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 |
|---|---|---|---|---|---|---|---|---|---|
| E28 | Frencken Group | 2.30 | NO ACTION | — | — | n/a | n/a | 1.39 | Frencken printed the strongest conviction reading this book has produced all window - but it is a sell signal, not a buy one, and we own none of it. The shares have run 76.8% in 90 days to a 36.8x P/E against a 27.1x industry average, and today's failed 20-day breakout is a coherent topping print on a name that has already outrun its sector multiple; management guided flat revenue for H1-2026. We stay out. Operational: AVOID is not a short in a long-only book at (SGM-4b), so this generates no action either way. FAIR_VALUE band with a mild negative MoS takes a -5 haircut on the x5 curve - MSCAP-2 flag, lean on the sign not the magnitude. Row prints Entry=Stop=2.30 with a blank Target - degenerate short-side risk line, unusable. · news |
| 8AZ | Aztech Global | 0.56 | NO ACTION | — | — | n/a | n/a | 0.32 | Aztech screens cheap and passes our quality check, and it carries the largest valuation credit this book has ever issued - but it flipped from a bullish oversold-reversal read to an outright breakdown signal overnight, and our own twelve-week estimate still says the shares are worth 43% less than today's price on revenue shrinking 30% a year. A cheap valuation on a contracting business is not a buy case. Operational: adj 65 is numerically at the floor but the label is bearish and the book is long-only (SGM-4b); SGM-5 vetoes independently on the forward view. News search returned a 0.97 quote inconsistent with the 0.565 settled close and was discarded - treat 8AZ search quotes as unreliable here. Re-arm: Target_12Week must turn non-negative AND a settled bullish print at adj >=65. · news |
| PCT | PC Partner Group | 2.93 | NO ACTION | — | — | n/a | n/a | 3.41 | PC Partner is still the one genuinely constructive name we own nothing of - cheap, quality-passing, and the only positive twelve-week outlook on the book at +16.4% - and today it rose 1.0% on its heaviest trading of the week. It did not deteriorate; it simply dropped out of the model's investable set on a technicality, so we hold fire rather than reach for it. Operational: NO TRADE cv0 on 'Liquidity gate failed' - daily range 3.07% vs the SG 3.00% cap, on S turnover (47x the floor); range_to_atr was 0.86, a below-average bar by its own volatility, so it would clear the US/HK gates comfortably. Re-arm: settled BUY/STRONG BUY at adj >=65, preferably PullbackBuy_VWAP/MomentumContinuation (L1/MSCAP-1 bar a FailedBreakdown chase). CAUTION: SGD 0.10 ex-dividend 2026-09-16, 3.4% of close, ~8 sessions out - size and stop for the mechanical gap or wait until after. The 31.88 DCF print is a growth outlier; FV_per_share 3.006 EPV is the anchor. · news |
| S71 | Sunright | 0.41 | NO ACTION | — | — | n/a | n/a | 0.24 | Sunright is the deepest bargain on the book by our valuation work and remains completely untradeable - roughly Schanges hands in a day, and the shares swung 12% between high and low. There is no way to build or exit a position at a sensible price, and our twelve-week estimate points 42% lower regardless. Watchlist only. Operational: liquidity gate fails on both tests (spread 12.05% vs the 3.00% cap, turnover S vs the S floor); val_adj clamped at the +10 maximum, QPass True. Re-arm: turnover must clear S AND Target_12Week turn non-negative AND a settled bullish print at adj >=65. |
| — | AI - SG Mid/Small Cap | — | NO ACTION | — | — | n/a | n/a | n/a | No trades today. The whole SGX small-cap semiconductor universe closed higher, but not one name gave us a buy signal we would act on - the two strongest readings of the day were both sell signals on stocks we do not own, and the one genuinely attractive name (PC Partner) fell out of the model's tradeable set on a liquidity technicality despite its busiest trading day of the week. We keep the full Sin cash. Operational: 0 ACQUIRE + 0 DISPOSE + 0 ANTICIPATE. Acquire pool empty by construction (zero bullish actionable labels); dispose pool cannot exist at across all 11 names; no BUY - ANTICIPATE print so the 4th slot is empty. Honest note: with all 11 names up 0.0-3.8%, tonight's all-cash posture was a COST, not the SGM-2 risk-avoidance win the last eight entries leaned on. Scorecard 207 entries / 0 graded / calib_1w null -> f=0.30 per L5; all biases 0. Overlay stale 61d, down-weighted. Learnings ACTIVE block 48d old; six PROPOSED blocks inert. Step 8a NOT run (batch directive). |
Outcome & track record
Accuracy at the time of this record.
0 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.
0.000 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.