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.

US End-of-Day Verdict 27 Aug 2026, 16:00:00 GMT-4

Sector ETFs — End-of-Day Verdict

EOD 2026-08-28 (sectoretf)

Recommendations

1 to acquire · 0 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
MAGSRoundhill Magnificent Seven ETF68.74BUY69.0768.6169.49 (+0.6%)69.89 (+1.2%)75.76 (+9.7%)Buy the Magnificent Seven basket. MAGS was one of only five names on a 22-ETF board to close green on a session where a hawkish Jackson Hole keynote from Chair Warsh pushed September rate-hike odds to 57% and knocked semiconductors down more than 3% — relative strength earned against a hostile tape, not with it. The signal has strengthened four sessions running (WATCH → BUY 75 → STRONG BUY 85), it is the board's only STRONG BUY and its top-ranked buy, the moving-average ladder is fully stacked and the forward view is intact rather than inverted — the model sees roughly +10% over twelve weeks. The honest caveat is that the wider 2026 story runs the other way: the Mag-7 basket lagged the S&P badly through the first half and Goldman reads institutional flow as rotating out of megacap tech into semis, so this is sized and timed as a swing, not a position. OPS: STRONG BUY 85, adj 83 after the -2 FailedBreakdown family haircut. At-market (range -), stop. = — model was 85.49% x x 20% = but max_position_pct 60% is applied against the DECLARED capital, not the inherited phantom equity (systemic finding 3, 2026-08-27), which caps it at Defined risk 434 x = = 0.40% of book. PRINCIPAL WEAKNESS, stated plainly: the stop is 0.50 ATR = 0.67% below entry — inside one normal session's noise, and this book has declined names on exactly that ground before (SKYY 07-30 at 0.58 ATR, DIA 07-31 at 0.74 ATR). It is taken anyway because the 0.50-ATR stop is a hardcoded family constant today, not a name-specific artefact (systemic finding 3 this run), and MAGS clears every other gate. SE-2 does NOT fire: Est_12W > Est_4W > close. Payoff 1.78x on Est_4W. 1wk estimate carries the SE-3 low-reliability flag (calib_1w -0.08); the f=0.30 fallback gives vs the model's — same neighbourhood. Executes at the 2026-08-31 open; reprice if it gaps >1.5%. VOID if the open gaps below. · news
Financial Select Sector SPDRNO ACTIONn/an/an/aFinancials print a buy and rank second on the board — banks were among the handful of green sectors as Warsh's hawkish keynote steepened rate expectations — but the trade is declined on its forward view. The model sees the fund essentially flat twelve weeks out (+0.4%) while projecting +1.8% at four weeks, an inverted horizon that on this book means a short tactical swing at best, and the apparent 3.8x reward-to-risk is manufactured entirely by a stop sitting less than half a percent below the entry. Not enough forward reward to commit capital that MAGS uses better. OPS: BUY 75, adj 73 (bias -2), Buy_Rank 2. Trail SBUY 88 → SBUY 78 → BUY 66 → BUY 75, re-strengthening. Entry = close, stop = 0.49 ATR = 0.48%. SE-2 FIRES: Est_12W < Est_4W. RE-ARM: take it if Est_12W lifts back above Est_4W with conviction ≥70, or on a genuine below-close pullback limit ≥1.0 ATR wide. NOTE: this is the name the 08-26 card queued at and the 08-27 card recorded as FILLED on the tape (08-27 low). That fill was never written to input/sectoretfportfolio.json, the monitor now reports XLF as an exit, and this run therefore reverts to the JSON-literal branch: the book is FLAT with See ADR-0019 below.
First Trust Cloud Computing ETFNO ACTIONn/an/an/aThe closest call on the board and the best risk/reward among the names that were declined. Cloud software is in a genuine uptrend — fully stacked moving averages, the only Trending regime among the buy candidates, and a model view roughly +8% over twelve weeks — but it gave back 0.9% into Friday's hawkish repricing and the trade would mean buying same-session weakness with a stop barely more than one percent away. Wait for the tape to settle. OPS: BUY 69, adj 67 (bias -2), Buy_Rank 4. Payoff 2.15x Est_4W, the best on the board; non-inverted forward; RSI 65.7, not extended. Stop = 0.50 ATR = 1.16%. Yesterday's re-arm trigger was 'AND Entry_Price within 1.0 ATR of close' — the second leg is now met (Entry_Price == close) but the first is not. RE-ARM: take it on a close near -162 that still prints conviction ≥69, or on any print where the stop widens past 1.0 ATR.
Invesco QQQ (Nasdaq-100)NO ACTIONn/an/an/aThe Nasdaq-100 tracker prints a buy but it is a buy into the teeth of the selloff it just suffered. QQQ closed down 0.6% as rate-hike odds jumped to 57%, and the model's stop sits 0.6% below the entry — inside a single session's normal range on a wrong-side-of-the-tape entry. The twelve-week view is good (+8%); the next few days are not the moment. OPS: BUY 69, adj 67 (bias -2), Buy_Rank 3. Payoff 1.53x Est_4W, non-inverted forward (Est_12W > Est_4W > close). Stop = 0.50 ATR. RSI 51.9, Ranging regime. RE-ARM: a green close that holds above EMA20, or a pullback limit ≥1.0 ATR below the close.
Tech Select Sector SPDRNO ACTIONn/an/an/aDeclined on reward, not on trend. Tech closed down 1.6% and the model's own four-week estimate lands barely a quarter of a percent above the entry price — for the risk taken, there is almost nothing to win over the horizon that matters. OPS: BUY 69, adj 67 (bias -2), Buy_Rank 5. Payoff 0.25x on Est_4W vs entry — the worst payoff among the above-floor names and the purest instance of the flat-forward problem on this board. Forward is non-inverted at twelve weeks (Est_12W, +5.6%) but the four-week leg is dead. Stop = 0.50 ATR.
SPDR S&P 500 ETFNO ACTIONn/an/an/aTwo points short of the action threshold once the signal-family discount is applied. The broad index fell 0.2% on the Warsh reaction and the entry is simply the closing price, so there is nothing to wait for and nothing to queue below the market. OPS: BUY 66 → adj 64, under the 65 floor on the -2 FailedBreakdown haircut alone. Buy_Rank 6. Payoff 1.57x Est_4W — it clears the reward bar and fails only on conviction. Stop = 0.50 ATR. Non-inverted forward (Est_12W, +2.9%). RE-ARM: any print at conviction ≥67 makes it actionable.
Consumer Discretionary SPDRNO ACTIONn/an/an/aThe day's best performer on the board — consumer discretionary rose 1.2% while tech and semis were sold — but it lands two points under the action threshold after the signal-family discount. Worth watching if the rotation into cyclicals has legs. OPS: BUY 65 → adj 63, below floor. Buy_Rank 7. Trail WATCH → 52 → BUY 65 — first buy label in four sessions. Est_12W (+8.5%) is the strongest twelve-week read on the board. Price is pinned to EMA20/50/200 within a percent — no trend to lean on yet. RE-ARM: a second consecutive buy print with conviction ≥70.
Health Care Select Sector SPDRNO ACTIONn/an/an/aHealthcare is the only name on the board printing the preferred pullback-to-VWAP setup, and it is the only one with a properly sized stop — but conviction is three points short and the model's twelve-week view is actually BELOW today's price. A cheap entry into a fund the model expects to be lower next quarter is not a trade. OPS: BUY 62, adj 62 (PullbackBuy_VWAP bias +0), Buy_Rank 8, below the 65 floor. Entry = 0.94 ATR below the close — genuinely queueable, and the stop is 1.16 ATR, clearing the width test every FailedBreakdown row fails today. DEAD FORWARD: Est_12W < close (-2.7%). RE-ARM: take it the moment Est_12W turns positive vs the close AND conviction reaches 65 — this is the highest-quality setup geometry on the board and only the forward view is stopping it.
Global X AI & Tech ETFNO ACTIONn/an/an/aYesterday's near-miss on chase grounds has resolved itself — the AI/tech fund gave back 0.5% and the signal collapsed from a strong buy to a watch in a single session, vindicating the decision not to buy it one cent off the session high. OPS: STRONG BUY 85 → WATCH 62, a 23-point conviction collapse. Below floor. Est_12W (+7.6%) keeps it on the watchlist. RE-ARM on a fresh buy print with a queueable below-close entry.
SPDR Dow Jones Industrial AvgNO ACTIONn/an/an/aThe Dow tracker was the market's most resilient major index on Friday, essentially unchanged while the S&P and Nasdaq fell — but the signal cooled back to a watch and there is no actionable print. OPS: WATCH 61, below floor. Trail WATCH → BUY 70 → WATCH 61. Entry_Price == close, nothing queueable. Est_12W (+1.9%) is a thin twelve-week case.
iShares Russell 2000 ETFNO ACTIONn/an/an/aSmall caps fell 1.4%, the second-worst of the broad indices, and remain a watch. Higher-for-longer rate expectations are a direct headwind to the most leveraged part of the market. OPS: WATCH 44, well below floor. Trail →. RSI 45.3, price below EMA20 and EMA50. Est_4W is BELOW the close.
Defiance Quantum ETFNO ACTIONn/an/an/aThe quantum-computing basket fell 1.8% and lost its signal entirely after four sessions of watch prints. OPS: NO TRADE cv0, down from WATCH 64. Close broke below EMA20 and EMA50. Est_12W (+8.7%) is attractive but there is no entry to act on.
Consumer Staples SPDRNO ACTIONn/an/an/aConsumer staples rose 0.4% — the classic defensive bid on a hawkish day — but conviction is far below the threshold and the model's four-week estimate sits below today's price. OPS: WATCH 51, below floor. Est_4W < close. Flat-to-nothing forward; the defensive bid is a market signal, not a trade here.
SPDR Gold Shares ETFNO ACTIONn/an/an/aGold fell 3.2% as Warsh's hawkish message lifted real-rate expectations, and the signal engine has no read on it. Notably, the long-running cost of having declined this name has now collapsed: the streak opened 2026-08-08 at and peaked at +6.1% forgone on 08-27; after today's drop to it is +2.6%. OPS: NO TRADE cv0, 'Consensus conflict + low conviction'. Est_4W is 2.6% BELOW the close — the forward leg fails outright. Still declined; the decline is now cheap.
VanEck Semiconductor ETFNO ACTIONn/an/an/aSemiconductors fell 3.5% — the worst name on the board — and the engine prints nothing at all, a day after printing an AVOID label on a +3.1% session. This book's semiconductor read has been uninformative across the entire NVDA reaction and its unwind. OPS: NO TRADE cv0. Close broke below both EMA20 and EMA50. Est_12W < Est_4W — inverted. See systemic finding 4.
iShares Semiconductor ETFNO ACTIONn/an/an/aThe other semiconductor proxy, down 3.2%, and silent for a fifth consecutive session — cv0 through the entire NVDA run-up, the reaction and now the unwind. OPS: NO TRADE cv0. Est_12W is BELOW the close. RSI 43.1 and price under EMA20/EMA50.
SPDR S&P Biotech ETFNO ACTIONn/an/an/aBiotech fell 3.5% on the rate repricing — long-duration equity is the most rate-sensitive corner of the market and it traded like it. No signal. OPS: NO TRADE cv0 for a third straight session. Close held just above EMA20. Est_12W (+5.2%) keeps it on the watchlist.
iShares Biotech ETFNO ACTIONn/an/an/aLarge-cap biotech fell 2.7%, no signal. The BUY - ANTICIPATE print from four sessions ago has fully lapsed — three consecutive NO TRADE prints since — so the anticipate slot has nothing eligible today. OPS: NO TRADE cv0. Est_12W (+4.4%). The 08-06 gap-guard card remains open under ADR-0019.
Destiny Tech100 IncNO ACTIONn/an/an/aThe book's only priced name on fair value, and the only one carrying a valuation penalty. Fell 1.9% and reverted to no signal after a single buy print. OPS: NO TRADE cv0. OVERVALUED_VS_RI, MoS -0.78 → val_adj -4 (residual-income basis; the fair-value file is 79 days old). Close against a anchored VWAP — still stretched. High-volatility, wide-band name; the model's 4wk band spans -.
Industrial Select Sector SPDRNO ACTIONn/an/an/aIndustrials fell 0.9% and remain the board's weakest momentum reading. No signal. OPS: NO TRADE cv0 for a fourth straight session. RSI 37.8 — the lowest on the board — with price below EMA20 and EMA50. Est_4W < close.
Procure Space ETF (SpaceTech)NO ACTIONn/an/an/aSpace fell 1.9% and continues to print an avoid label. Long-only book, so this costs nothing directly, but the read is negative. OPS: AVOID 48 on FailedBreakout_Down_20D, fourth consecutive avoid print. RSI 37.0, price below EMA20/EMA50/EMA200 — the only name on the board under all three. No target price printed.

Outcome & track record

Accuracy at the time of this record.

35 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 →
-0.080 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.

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.