SG Mid/Small Cap — End-of-Day Verdict
NO ACTION (both sides) - 25th written EOD verdict since 07-03; the book's best session of the standalone era (+1.76%, 10/11 names up) with an empty buy pool, because the reopen gap that caused the move also widened ranges and tripped the spread gate on 6/11 names
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 |
|---|---|---|---|---|---|---|---|---|---|
| BN2 | Valuetronics Holdings | 0.99 | WATCH | 0.99 | 1.09 | n/a | n/a | 1.21 (+22.2%) | We are still refusing to act on Valuetronics because its price history is corrupted, not because of anything the business did. Last week's dividend triggered a 28.6% back-adjustment of the entire price series against a dividend actually worth about 4.7%, so every indicator derived from it is wrong: the stock is shown as heavily overbought when it has simply traded flat at 0.99. Tonight the one accidental safeguard that had been keeping this row out of consideration stopped applying, and the bad data produced a live sell-side signal - it happened to point the safe way, but that was luck. We will not trade this name in either direction until the data reconciles. OPERATIONAL: MUTED ON SIGNALS. Liquidity gate passed tonight (2.53% range vs 3.00% cap) after failing by 0.03pp on 08-07, so the corrupted row reached the cascade and printed cv59 off a fabricated RSI 79.9. EMA200 0.6944 vs 0.99 close (+42.6%); ladder healing by absorption only (EMA20 0.7874 -> 0.8067). T12 escalated 1.10 -> 1.21 (+22.2%) - discard. Yahoo now republishes the broken 0.2857 factor as an 'SGD 0.30 dividend' (0.2857 x 1.05 = 0.30 exactly), so news cross-checking returns the corruption. Re-arm: EMA ladder must re-converge with the raw series FIRST, then a settled favoured-cell BUY >= 65 with Quality_Pass repaired. Systemic finding 2, severity-1, now unmitigated. · news |
| PCT | PC Partner Group | 2.97 | WATCH | — | — | n/a | n/a | 4.05 | PC Partner is the one name in this book where the business case and the chart are both improving, and it is the one we keep being unable to buy. Its momentum has strengthened three sessions running, it carries the only genuinely positive 12-week outlook in the mandate at about +36%, it trades around fair value rather than at a stretch, and the press reports first-half earnings likely to double on better pricing and margins. What stops us is a mechanical screen that rejects any stock whose daily trading range is wide - and this stock's range was wide precisely because the market reopened after a holiday. We are holding fire rather than overriding our own risk screen. OPERATIONAL: Momentum pillar 26 (08-06) -> 35 (08-07) -> 68 (08-11), Trend 77, RSI 60.1, +1.71% to 2.970, turnover SGD 4.35m. Gate-failed on a 5.05% range (2.85-3.00) vs the 3.00% cap, so conviction was never scored. Re-arm is a two-legged AND: settled PullbackBuy_VWAP/MomentumContinuation BUY that clears the gate AND prints raw cv >= 65; each leg has now been met on a different day, never together. Last printed shape: 2.74 limit / 2.56 stop / 3.17 target. DO NOT tune the 3% gate to bridge this - MSCAP-1 / SGM-3, forward-test rule. Primary exhibit for the champion/challenger shadow test (systemic finding 1). · news |
| AWX | AEM Holdings | 9.54 | WATCH | — | — | n/a | n/a | 9.40 | AEM is the strongest business in this book and we have not owned a share of it for thirteen sessions. It rose over 4% today, it is the most heavily traded stock in the mandate by a wide margin, and management has just raised full-year revenue guidance to S-600m, implying growth of 38-50%. Our screen still refuses it - not on valuation and not on the business, but because a wide daily trading range trips a filter intended to catch illiquid stocks. Rejecting the most-traded name in the universe on its most-traded day tells us the filter is mismeasuring, and that is now logged as an engineering fix rather than something to override by hand. OPERATIONAL: +4.26% to 9.540 on SGD 53.74m turnover (book record, 1.6x prior night); Liquidity gate failed on a 7.02% range (8.91-9.58). Trend 62, Momentum 52, Consolidation_Bullish, RSI 56.7, close +9.58% vs VWAP. T12 re-rated 8.48 -> 9.40, gap to close narrowed -7.3% -> -1.5%. News dates UNVERIFIED (one item quotes a garbled '512% 1H2026 return') - undated background, cannot promote a name with zero bullish prints (bounded-news doctrine). MSCAP-2 for a 13th session: FAIR_VALUE label on MoS -10.780. No re-arm; await a real bullish cascade output. · news |
| 558 | UMS Integration | 2.52 | WATCH | — | — | n/a | n/a | 2.34 | UMS was the best performer in the book today, up over 4.5%, and it remains one we will not buy. The stock is priced well beyond what its earnings currently support, and it fails our quality screen on profitability and cash-flow consistency - a combination that would sink even a clean technical signal. Tonight it also became too volatile for our entry screen. Nothing here has changed our view; a strong day in a richly priced name is not an entry. OPERATIONAL: +4.56% to 2.520 on SGD 25.84m turnover (2nd largest); flipped Consensus conflict -> Liquidity gate failed on a 5.16% range. Double block intact and deepened: GROWTH_PRICED_IN MoS -4.016 (from -3.777) => val_adj -10 clamped, AND Quality_Pass False (SGM-4a). T12 2.34 = -7.1% vs close. Cannot clear absent a valuation AND quality re-rate. |
| 8AZ | Aztech Global | 0.60 | WATCH | — | — | n/a | n/a | 0.34 | The rebound signal we declined to buy on 6 August has kept falling - the stock is now about 3% below the entry we passed on, at a fresh momentum low, with revenue still contracting. Its valuation screen has actually improved since we declined it, but only because the price dropped; cheapness that comes from a falling share price is not a reason to buy. We stay out until the business shows revenue stabilising. OPERATIONAL: -0.83% to 0.600 (declined 0.62 entry => -3.2% cumulative), RSI 17.9 fresh low. Double-blocked: gate-failed on a 4.17% range AND Quality_Pass False. T12 0.34 = -43.3%, tied worst in book. MoS improved +0.168 -> +0.220 purely on the price fall - textbook SGM-5, cheapness never originates a position. Re-arm: FY2026 revenue stabilisation repairing Quality_Pass FIRST, then a settled favoured-cell BUY >= 65. Would grade -3.2% as a hypothetical refusal if the scorer could record it (systemic finding 6c). |
| M14 | InnoTek | 0.50 | WATCH | 0.50 | 0.50 | n/a | n/a | 0.28 (-44.0%) | InnoTek is again the only conventional signal our engine produced, and it is a bearish one - a failed attempt to break higher. We hold no shares and we do not sell short, so there is nothing to do. It is worth stating plainly why this matters: when the only clear reading in the whole book is a negative one, that confirms there is nothing to buy - it never creates something to buy. OPERATIONAL: cv56 (from 55), entry/stop 0.50, no Target_Price, T12 0.28 = -44.0% (worst in book). val_adj -7, Quality_Pass False => adjusted 49, below the 65 floor regardless of direction. Cleared the gate on a 3.00% range, exactly at the cap. SGM-4b, 9th instance. |
| S71 | Sunright | 0.49 | WATCH | — | — | n/a | n/a | 0.28 | Sunright remains the cheapest business on our list relative to its earnings power, and it is still the one we are least able to own. Only about Sof stock changed hands today - less than half the minimum daily turnover we require - so a position of any meaningful size could not be entered or exited without moving the price against us. The chart is also weak and the 12-week outlook deeply negative. A stock that cannot trade cannot be a holding at any price. OPERATIONAL: +4.26% to 0.490 on a 6.12% range and SGD 40,376 turnover - DOWN from 51,277 and 60% below the SGD 100k floor; the only Fail_Both in the book, 2nd session running. val_adj +9 (UNDERVALUED MoS +0.459, QPass True) would only lower the required raw conviction to 56, far above the chart (Trend 51, T12 -42.9%). SGM-5 stands. Needs genuine turnover AND trend repair, in that order. |
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.