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 |
|---|---|---|---|---|---|---|---|---|---|
| AWX | AEM Holdings | 9.73 | WATCH | 9.73 | 9.50 | 9.97 (+2.5%) | 10.53 (+8.2%) | 9.99 (+2.7%) | AEM is the strongest business on this book - roughly +90% year to date, first-quarter revenue up 36% with net margin tripling, a reinstated dividend and a new AI/HPC customer guided to become its largest - and tonight it produced its first buy-side signal in over a month. We are still not buying it. The signal only appeared because the shares' daily trading range happened to narrow to 2.88% from 7.04%, slipping under the model's 3.00% volatility cap; nothing about the company or its very heavy S daily turnover changed. On our own numbers the price already discounts more growth than we can justify, our quality screen fails the name outright, and the model's own twelve-week view has narrowed to just +2.7% above tonight's close - so the reward for chasing it here is thin. We keep it on the watchlist and want a genuine valuation or quality re-rating alongside a proper trend signal before committing capital. OPERATIONAL: WATCH is not an actionable buy Action (Step 5 pool = STRONG BUY/BUY/ADD only); val_adj -10 clamped from MoS -11.233 drags cv62 to adj 52, 13 below the 65 floor; Quality_Pass False independently hard-blocks the new entry; FailedBreakdown_Up_20D is the L1/MSCAP-1 weak cell (US hit ~0.25, stop-first ~60%). Gate rotated from PCT (09-10) to AWX (09-11) emitting the identical unbuyable object - systemic finding 1. Re-arm: quality/valuation re-rate PLUS a settled PullbackBuy_VWAP/MomentumContinuation BUY at adj >=65; the range condition is now satisfied so liquidity is no longer this name's binding constraint. · news |
| BN2 | Valuetronics Holdings | 0.95 | WATCH | 0.95 | 0.93 | 0.96 (+1.3%) | 0.99 (+4.2%) | 0.81 (-14.7%) | Valuetronics came the closest of any name to earning a place in the book tonight - its reversal signal strengthened materially and the shares are now the most oversold in the mandate - but it still falls short, and the reasons it falls short are the right ones. The setup is a bounce attempt in a downtrend: our twelve-week view on the shares is about 15% BELOW tonight's price, the company fails our balance-sheet and cash-flow quality screen, analysts have been cutting their sales forecasts, the share count has grown nearly 7% in a year diluting existing holders, and a published intrinsic-value estimate puts the stock around a quarter above what it is worth. Buying a rising signal into a falling forward view on a company failing its quality test is the trap, not the opportunity. We want to see the quality screen actually pass and a proper trend signal establish itself before we treat this as a buy. OPERATIONAL: cv41 -> 58 (+17) with RSI 31.7 -> 28.9, the closest approach to the 65 floor in weeks but still 7 short; Quality_Pass False hard-blocks the new entry irrespective of conviction; val_adj 0 (MoS -0.051 rounds the x5 to zero, so MSCAP-2's over-penalty does not bite here); Entry_Distance_ATR -0.31. Shape matches the inert MSCAP-11 proposal (strong print on a QPass-False name is not durable) but the hard gate settles it independently. Re-arm: Quality_Pass True AND a settled favored-cell BUY at adj >=65 persisting >=1 further session. · news |
| 8AZ | Aztech Global | 0.56 | WATCH | 0.56 | 0.56 | n/a | n/a | 0.32 (-42.9%) | Aztech printed a bearish failed-breakout signal for the sixth session running and the signal is weakening rather than turning. This is a long-only book that holds no Aztech shares, so a bearish reading generates no action in either direction - it simply confirms there is nothing to buy here. The shares do screen as cheap on our valuation work, but cheapness alone has never been a reason to open a position, and the model's twelve-week view sits roughly 43% below tonight's price. OPERATIONAL: cv56 -> 46 (decaying); val_adj +5 (UNDERVALUED MoS +0.272, QPass True) lifts adj to 51, still 14 below floor and pointing the wrong way. SGM-4b: a bearish print confirms an empty buy pool, it never populates one. |
| PCT | PC Partner Group | 3.10 | WATCH | — | — | n/a | n/a | 3.87 | PC Partner's brief buy signal from last night has already vanished - the model's trend and momentum readings fell into conflict and it is back to no-trade, with the price unchanged on the day. The name still carries much the most attractive forward view on this book, roughly 25% above the current price over twelve weeks, so it stays top of the watchlist. Tonight's refreshed valuation work also reset it from cheap to fairly valued, exactly as we flagged yesterday would happen once the stale numbers caught up with the recent share-price run - so it is now neither a bargain nor expensive, and any case for owning it has to come from the chart. Note the shares go ex-dividend 10 Singapore cents on 16 September, about 3.2% of the price: that drop is mechanical and should not be mistaken for a breakdown. OPERATIONAL: NoTrade_Reason verbatim 'Consensus conflict + low conviction'; Pillars Conflicted, Momentum 16, Regime Transitional; range 1.94% so liquidity was NOT the blocker tonight. Valuation label moved UNDERVALUED/+0.035 -> FAIR_VALUE/-0.031 on the 09-11 regen, validating the 09-10 prediction. Re-arm: settled PullbackBuy_VWAP/MomentumContinuation BUY at adj >=65; discount EMA-relative readings across the 09-16 ex-div gap. · news |
| S71 | Sunright | 0.41 | WATCH | — | — | n/a | n/a | 0.23 | Sunright carries the deepest valuation discount on the book and remains untradeable. Only about nine thousand Singapore dollars of stock changed hands today - roughly a tenth of the minimum liquidity we require - while the price swung nearly 10% between its high and low. At that size we could not build or exit a position without moving the price against ourselves, and the model's twelve-week view is around 44% below the current price anyway. Cheap and unsellable is not an investment. OPERATIONAL: NoTrade cv0 on liquidity gate failure, turnover S = 0.09x the S floor, range 9.76% - double-fails by an order of magnitude. val_adj +10 clamped (MoS +0.548, QPass True). SGM-5: cheapness never originates a buy absent a bullish chart plus liquidity clearance. |
| — | AI - SG Mid/Small Cap | — | NO ACTION | — | — | n/a | n/a | n/a | No action tonight - the book stays fully in cash at SThree of the eleven names in the mandate produced a signal and none of them is something we are willing to own. AEM, much the strongest business here, only generated a signal because its daily trading range happened to narrow below a volatility threshold in the model - nothing about the company changed - and it fails our quality screen with almost no upside left in the twelve-week view. Valuetronics came closest to qualifying and is the most oversold name on the book, but it is a bounce attempt in a downtrend with a forward view 15% below today's price, a failing quality screen, analysts cutting forecasts and shareholders being diluted. Aztech remains outright bearish. The wider group fell about 0.7% today, so sitting in cash was worth roughly Son the session. We would rather hold the cash than force a position that does not meet our standard. OPERATIONAL: 47th consecutive all-cash session since standalone inception 2026-07-06. Adjusted convictions AWX 52 / BN2 58 / 8AZ 51, all below the 65 floor; AWX and BN2 additionally hard-blocked by Quality_Pass False. No BUY - ANTICIPATE print so the 4th slot is empty; dispose pool structurally empty at (MSCAP-3). Fairvalue REGENERATED tonight (was 8.0 days stale), FetchedAt 2026-09-11 18:21:45, NA; PCT inverted UNDERVALUED -> FAIR_VALUE exactly as predicted 09-10 and 575 lost Quality_Pass, taking QPass-False to. Scorecard 237 entries, 0 graded, calib_1w None, all biases 0, f=0.30 per L5. Overlay stale 68 days; Event Horizon parent fallback with zero of 14 events matching any book ticker (verified programmatically). Monitor clean: no positions, no breaches, no drift, no open queues. 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.