This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.

SG End-of-Day Verdict 20 Aug 2026, 17:00:00 GMT+8

SG REITs — End-of-Day Verdict

EOD 2026-08-20 (sgreit)

Recommendations

0 to acquire · 1 to dispose

TickerNameCloseActionAction 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
N2IUMapletree PanAsia Com Tr1.28SELL 50%1.271.25n/an/a1.16 (-8.7%)MPACT bounced a cent to 1.27 today, but the trend case did not change: this was its TENTH consecutive close below the 200-day average (1.297), the model's twelve-week view still sits at 1.16 - roughly 8.7% below today's price - and, tellingly, the engine's confidence in the bearish FailedBreakout pattern ROSE from while the price went up. A green day on which the model turns more bearish is a bounce inside a downtrend, not a bottom. There is no bad news at the trust - the gateway portfolio is intact and the SAugust distribution is already banked as a receivable - and sector coverage is mildly constructive, with S-REIT borrowing costs easing and active capital recycling across the sleeve. That is precisely why this is a half-trim and not an exit: a distribution-payer de-rating on the rate path deserves a SMALLER position, not a closed one. Selling half returns ~Sof the book's only exposure to cash ahead of two more rate events - the PCE reaction on 27 Aug and Jackson Hole's reaction on 31 Aug - whose S-REIT impact bands are +/-2.5% and +/-3.0%. OPERATIONAL: pre-committed 08-06 ladder. BE PRECISE ABOUT WHICH CLAUSE IS LOAD-BEARING TONIGHT - clause (A) (decisive stop-break) technically breaks a fifth time at 1.27 vs the operative pre-trim stop 1.280, but at only -0.78% / -0.45 ATR it is the SHALLOWEST of the five and is NO LONGER DECISIVE in REIT-2's sense; on its own it would be a watch. The card fires on clause (B) alone: 10th consecutive close below EMA200 (-2.09% / -1.21 ATR) with Target_12Week 1.16 < close (-8.66%) - the L2 / REIT-2 basis. REIT-1 remains affirmative support for a second session (FailedBreakout_Down_20D is the family it trusts). The bull case WAS re-tested and did not clear: close +1c, RSI 38.57 -> 43.17, Momentum pillar off the floor, and Target_12Week UP for the first time in nine sessions (1.15 -> 1.16) - but volume expanded to 11.18M on the up-cent, the whole EMA stack is still above price, and clause (C') (settled close above EMA200 ~1.2972 with Target_12Week >= close) is the ONLY void condition and is not met. Residual trigger (A') (settled close < 1.25) moved FURTHER from firing: today's low was 1.26, so the 1.25 line was not tagged at all after two sessions of exact tags, and the close sits +0.89 ATR above it - portion stays 50%, not more. ACTION PRICE REPRICED UP 1.26 -> 1.27 off the LIVE anchor Entry_Price (= settled close); no carried-forward level. 72,550 rounded down to the 100-share SG board lot = 72,500; residual 72,600 keeps stop 1.25 (NOT re-based on today's bounce, NOT raised). Reprice if the open gaps outside 1.251-1.289. 1wk/4wk estimates UNAVAILABLE because the FailedBreakout_Down_20D row emits a blank Target_Price - systemic finding 7, tonight reproduced on FIVE of eight names. 12wk 1.16 is +13.7% above the 1.02017 DCF anchor - annotated, not applied. PROFIT_ARMED is a FALSE POSITIVE for the 13th session (Gain_ATR -2.67, T12_Progress 109.48% on a position down 4.51%) and is NOT the basis; note both readings moved the RIGHT way tonight as the price rose, confirming the metrics track price correctly but are never gated on profitability - systemic finding 1. Valuation freshly REGENERATED tonight (MoS -0.2449, val_adj -1) but barred from originating per CLAUDE.md section 9 / REIT-2 / SGR-1, and far too shallow to bump a tier. ADR-0020 moot (peak_stop 1.30, erosion 3.85%, class default-off). NINTH issue of the same card - the 08-07 through 08-19 cards all lapsed unfilled because the auto-executor has not run since 2026-08-03 (TWELVE trading days stale); today's miss cost Sand the cumulative gap NARROWED to ~Sas the price recovered, and last night's escalated event-timing risk did NOT materialise because the FOMC reaction session was benign - luck, not process. · news
J69UFRASERS CPT TR2.19WATCH2.192.13n/a2.22 (+1.4%)1.94 (-11.4%)No action - not held, and this is a bounce off a 52-week low rather than a turn. Frasers Centrepoint rebounded to 2.19 from yesterday's fresh low and the signal flipped to an oversold-reversal read at conviction 67, the highest in the book - but the label is WATCH, not a buy, and the model still projects 1.94 over twelve weeks, some 11.4% below today's price and the worst forward view in the sleeve. The setup also offers a 3-cent target against a 6-cent stop. Declining is not a close call. OPERATIONAL: never entered the buy pool (Action=WATCH is not a bullish actionable label). Three independent blocks even if it had: (i) SGR-3 - a +29-point single-session conviction jump into a bullish family with NO confirming same-family prior print; the 08-17 ReversalUp_Oversold 41 is separated by two BreakdownDown sessions, so it is the same whipsaw repeating, not a confirming second print, and that 08-17 instance faded within one session before the name made a fresh 52-week low - SGR-3 vindicated an 11th time; (ii) SGR-2 - Target_12Week 1.94 < close 2.19; (iii) reward:risk 0.5:1 on the printed entry/stop/target. Valuation NA (gate off, val_adj 0).
A17UCapitaLand Ascendas REIT2.46WATCH2.432.43n/an/a2.22 (-8.6%)No action - unheld, and the model's strongest read in the whole book points DOWN. Ascendas REIT edged up to 2.43 but conviction rebounded to a book-high 71 and the signal flipped into the bearish FailedBreakout family - the pattern this REIT book actually trusts. The twelve-week view is 2.22, roughly 8.6% below today's price, and the stop level sits right at the close. Nothing to buy, and nothing held to sell. OPERATIONAL: second of two names tonight where the engine HARDENED its downside view into an up-day (with M44U) - the night's key tell. Row lost its Target_Price on the family flip (it printed 2.43 last night under RallySell_BearTrend) - fourth instance of systemic finding 7 tonight. SGR-2 blocks on the negative 12wk; SGR-1 nulls the -7.65 margin-of-safety read for buy purposes. val_adj -10 applied strictly per formula clamp(round(MoS_FV x 5), -10, 0); the 08-19 entry recorded -8 using x1 - see systemic finding 11, impact nil.
M44UMapletree Logistics Trust1.17WATCH1.171.17n/an/a1.06 (-9.4%)No action - unheld, and a 38-point conviction leap in one session that leaps the WRONG way. Mapletree Logistics rose to 1.17 on heavy volume, yet the engine's confidence in its bearish FailedBreakout pattern jumped from - joint-highest in the book, pointing down, on an up-day. The twelve-week view of 1.06 is 9.4% below the price. OPERATIONAL: first of two names tonight where the model hardened its downside view into a bounce (with A17U). 26.30M units traded. Blank Target_Price - fifth instance of systemic finding 7 tonight. The 08-04 BUY 71 at 1.23 this book declined now trades -4.88%; SGR-3 graded correct an 11th time. SGR-2 blocks on the negative 12wk. Valuation NA.
HMNCAPITALAND and ASCOTT STAPLE U0.86WATCH0.85n/an/a0.77 (-9.9%)No action - unheld and bearish. CapitaLand Ascott closed 0.855 completely unchanged for a FOURTH session with conviction pinned at 60 on the bearish FailedBreakout pattern - the one name in the sleeve that did not participate in today's bounce at all. The twelve-week view is 0.77, nearly 10% below the actionable price, and the exit strategy on the row reads immediate. A flatlining name with a negative forward view is not an entry. OPERATIONAL: RSI frozen at 37.4655882135 to ten decimal places for a fourth session on bars that genuinely traded - contributes to systemic finding 8 (indicator window not advancing on unchanged-close rows). Blank Target_Price. SGR-2 blocks independently.
ME8UMapletree Industrial Trust1.92WATCH1.911.91n/an/a1.81 (-5.2%)No action - unheld and bearish. Mapletree Industrial closed 1.91 for a fourth straight session with conviction pinned at 52. The twelve-week view of 1.81 is the least-bad in the book and is still 5.2% below the price. Nothing actionable. OPERATIONAL: last night's finding that a conviction-52 bearish label materialised from a zero-information bar has now PERSISTED for a second session on a second such bar - RSI 47.1558402849 identical to ten decimals while the row traded 6.08M units on a 1.89-1.92 range. Reclassified under systemic finding 8: the indicator window is not advancing, so the label persists from a frozen state rather than being freshly computed. Declined on SGR-2's negative-12wk rule, which holds regardless. val_adj nulled by SGR-1 (EPV screen non-informative on distribution-paying trusts; raw formula value would be -10).
C38UCapitaLand Integrated Comm Tr2.40WATCHn/an/a2.30No action - the engine gives no trade at all (consensus conflict, zero conviction) for a third session. CapitaLand Integrated recovered a cent to 2.38, clawing back part of the 08-19 ex-dividend mark-down, and its twelve-week view of 2.30 is the least-negative in the book at -3.4%. The name now trades 4.0% below the 2.48 level it was chasing three sessions ago - declining to queue there was right, and the re-arm has never triggered. OPERATIONAL: the freshly regenerated valuation DEEPENED its margin of safety from -42.95 to -46.62 on a HIGHER price, which is the EPV screen behaving exactly as SGR-1 describes for distribution-paying trusts. Heaviest volume in the book again at 27.26M. RE-ARM UNCHANGED: a fresh PullbackBuy_VWAP at conviction >= 65 with Target_12Week >= close, queued at the THEN-LIVE Entry_Price - never at 2.48 or any carried-forward level.
BUOUFRASER L&C TRUST0.94WATCHn/an/a0.85No action - the engine gives no trade at all for a SEVENTH straight session. Frasers Logistics closed unchanged at 0.94 with its RSI at 26.9, the only deeply oversold reading in the book and the obvious temptation tonight. Declined: there is no entry, no stop and no target on the row to size or risk-define against, and the twelve-week view of 0.85 slipped further to 9.6% below the price - one of only two names whose forward view DETERIORATED on an up-day for the sleeve. Oversold is not a signal without a trigger. OPERATIONAL: this row still carries the book's only Bargain=True flag alongside a -5.85 margin of safety, and tonight's FRESHLY REGENERATED file reproduces the contradiction exactly (Val_Signal FAIR_VALUE + Bargain True from MoS_DCF +0.034, against MoS_FV -5.85 from the EPV anchor) - proving systemic finding 4 is a live logic defect, not a stale-file artifact. The flag has been true for seven sessions and the name has not risen once. RSI also frozen to ten decimals for a second session - systemic finding 8. SGR-1 nulls the valuation read; SGR-2 blocks on the negative 12wk.

Outcome & track record

Accuracy at the time of this record.

10 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 →
-0.043 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.
48 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.

This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.