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 |
|---|---|---|---|---|---|---|---|---|---|
| — | AI - SG Mid/Small Cap (all-cash) | — | NO ACTION | — | — | n/a | n/a | n/a | No action again, and for a second straight session the cash did real work. The semiconductor selloff extended: SK Hynix reported record quarterly revenue and a sixfold jump in operating profit that still fell short of expectations, and its shares fell 11-15%, dragging Kioxia, MediaTek, Micron, SanDisk and SoftBank down with them. The worry is structural rather than a one-day wobble - Chinese competition, a Chinese lithography advance, and how much of the AI buildout the chipmakers are financing themselves. This book is that supply chain, and today all eleven names fell and none rose, the basket down 4.34%, or 8.74% across the two sessions. Fully in cash it lost nothing, against roughly Stoday and Sover the two days had the seed been spread across the mandate. There is nothing to act on: ten of eleven names print no-trade with no entry price at all, the eleventh prints a bearish breakdown, and the twelve-week outlook is now negative on ten of eleven. The book is broadly oversold - median RSI around 35, Aztech at 26 - and that is explicitly not a buy trigger: oversold with no bullish print, a negative forward view and an unresolved structural catalyst is a falling knife (L6, SGM-5). Full SGD 2,000,000 retained; 18th written EOD verdict since 07-03. SGM-2 earned directly a second day; MSCAP-3 (all-cash book, trim/add deliverables N/A); MSCAP-1/MSCAP-4/SGM-3 (no without a settled strong-cell BUY >=65 clearing both gates - size is irrelevant); L8 (nothing pending at the Thu 07-30 open). |
| BN2 | Valuetronics Holdings | 1.05 | NO ACTION | — | — | n/a | n/a | 1.03 | The decision not to buy Valuetronics is now vindicated three times over. It was the sole name to clear our conviction bar on Monday, held back only by the quality screen - weak F-Score, shrinking revenue, priced above every intrinsic anchor. Conviction went to zero on Tuesday; today the signal has not merely decayed but inverted, printing a bearish 20-day breakdown at conviction 71 with a downside target of 0.81, on the heaviest volume of the whole sequence. The shares closed at 1.05, so an entry at 1.11 on Tuesday's open would be down 5.4% and would have been stopped through the 1.08 stop on day one. The twelve-week view has also flipped negative, from +1.9% to -1.9%. Long-only with no position, so a bearish print generates no action either way - a lone AVOID confirms an empty buy pool, it never populates one (SGM-4b). Re-arm unchanged and materially further off: a settled favoured-cell BUY at conviction >=65 WITH the quality screen repaired; conviction alone will not open this. The 2026-08-07 ex-dividend date, on a 132% payout ratio, remains the cleaner read on whether any trend exists beneath the dividend-capture bid. Recorded direction none / optional so the counterfactual grader prices the avoided entry. · news |
| PCT | PC Partner Group | 2.60 | NO ACTION | — | — | n/a | n/a | 3.25 | PC Partner is now the book's only name with a positive outlook at all - the model sees 3.25 twelve weeks out against today's 2.60, about +25% - on genuinely decent fundamentals: undervalued, passing the quality screen, F-Score 8. It also behaved best on the day, down 1.9% against a basket down 4.3%, and it is cheaper again, which widens the true margin of safety to roughly 10% once the two-day selloff is priced in (the valuation file still marks it at Monday's 2.82). But for a sixth consecutive session the model refuses to trade it on insufficient liquidity, and there is no buy signal to act on. Cheapness never opens a position here (SGM-5). Both sides stay on the record: the liquidity gate has priced this book out of its best idea for six sessions, and across those same six sessions it has spared it the drawdown. Top watch - re-arm on a settled favoured-cell BUY >=65 that actually clears the liquidity gate. Direction none / optional so the grader prices the non-entry. · news |
| 8AZ | Aztech Global | 0.71 | NO ACTION | — | — | n/a | n/a | 0.55 | Aztech was the sharpest fall in the book today, down 14.5% to 0.71 on more than three times its normal volume, leaving it deeply oversold with RSI at 26. We checked whether this was a dividend artefact and it is not - the declared one-cent dividend goes ex on 4 August, and a cent on 83 cents is 1.2%, nowhere near the move. This is a genuine mark-down, consistent with a company whose last full year saw revenue fall 30% and profit fall 43%. Priced against today's close it now screens undervalued by around 17%, and that is precisely the trap: it fails the quality screen, so the hard gate blocks a new entry no matter how cheap it gets, and the model's twelve-week view is 0.55 against 0.71, a further 22% lower. A textbook case of the gate doing its job on an oversold small-cap. No signal, no entry price, no action. Watch the 4 August ex-dividend and any sign of revenue stabilising that could repair the quality screen. Direction none / optional so the grader prices the non-entry. · news |
| S71 | Sunright | 0.47 | NO ACTION | — | — | n/a | n/a | 0.27 | Sunright holds the widest discount in the book - roughly 48% below intrinsic value on today's close, and it passes the quality screen - yet it is the clearest illustration of why cheapness alone never opens a position here. It fell 12.8% today to 0.475, and tellingly it did so on against 18,600 the day before, the first session in which the stock genuinely traded; the previous run of unchanged prints looks like stale marks rather than stability. Against that, the model's twelve-week view is 0.27, another 43% lower - the most negative forward view in the book - and it prints no trade on insufficient liquidity. Cheap and weak is not a signal (SGM-5). It needs both trend repair and liquidity before it is even a watch candidate, let alone a buy. Direction none / optional so the grader prices the non-entry. |
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.
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.