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 2 Sept 2026, 16:00:00 GMT-4

Sector ETFs — End-of-Day Verdict

EOD 2026-09-03 (sectoretf)

Recommendations

3 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 ETF70.43BUY — QUEUED @69.8269.8266.2170.78 (+1.4%)71.56 (+2.5%)78.44 (+12.3%)Mega-cap tech broke to a 52-week high on the day Fed Governor Waller signalled he would hold rates steady, collapsing September hike odds from 63% to 50% and pulling the 10-year Treasury yield down to 4.76% from its highest level since 2023. MAGS closed +2.43% with the highest conviction reading anywhere in the book (90) and the strongest forward view on the board -- a 12-week estimate 12.4% above the entry -- while still trading 4% below where it started the year, so this is a laggard breaking out rather than a leader extending. Queued below market at 69.82 rather than bought at the 70.43 close: no-chase (Entry_Distance_ATR -0.53, closed 0.5% off the session high). Stability is the weak point -- one STRONG BUY print after a NO TRADE two sessions back, and this name was withdrawn from the queue on 09-01 and declined on 09-02 -- so it leads on conviction, not on trail. BreakoutUp_52W cell is n=3, report-only, no bias applied. Widest stop of the three at 3.11 ATR; risk 646 USD = 1.29% of book. Executes into the 09-04 08:30 ET payrolls print -- reprice or stand down if the open gaps beyond ~1.5%. Expires after 3 sessions (09-07 is Labor Day). · news
SKYYFirst Trust Cloud Computing ETF163.92BUY — QUEUED @162.51162.51158.15164.91 (+1.5%)164.49 (+1.2%)169.51 (+4.3%)Cloud software was the single best performer in the book at +2.89%, and for two independent and nameable reasons at once. Falling interest rates lift long-duration software valuations more than anything else on the tape, and the 10-year fell to 4.76% after Fed Governor Waller signalled a rate hold; separately Snowflake surged 20.6% on 35% revenue growth, a large earnings beat and raised full-year guidance, describing enterprise-AI consumption as an accelerating flywheel -- which re-rated enterprise software broadly. This is the most stable signal in the book: four consecutive escalating bullish prints (62, 73, 73, 85) with no whipsaw, in a Trending regime, and it sits in the book's best-supported buy family (PullbackBuy_VWAP, n=11, actionable bias +1). Critically, the exact objection that caused this book to decline SKYY on both 09-01 and 09-02 -- rate-duration risk with the 10-year at a two-year high -- is the thing that reversed today. Queued 0.87% below the close (Entry_Distance_ATR -0.37) -- the shallowest and most fillable of the three limits. Note the 4-week estimate (164.49) sits below the 1-week (164.91) and barely above the close: per SE-2 run this as a tactical swing to the 171.24 target with a planned exit, not a position hold. Risk 331 USD = 0.66% of book. Executes into the 09-04 payrolls print; re-validate at the open. · news
IBBiShares Biotech ETF213.81BUY — QUEUED @208.85208.85204.23213.81 (+2.4%)216.50 (+3.7%)225.33 (+7.9%)Biotech holds a second consecutive STRONG BUY at identical conviction (79), the steadiest trail of today's three acquires, and carries the cleanest forward structure on the board -- 1-week, 4-week and 12-week estimates rising in order to 7.9% above the entry. The underlying case is real: a stronger drug-development cycle, a friendlier FDA, heavy M&A and AI-accelerated research pipelines have carried the group up more than 20% since semiconductors peaked in June. But the news no longer supports this the way it did yesterday, and that is said plainly: IBB closed exactly flat (+0.00%) and XBI fell 0.60% on a day the Dow rose 1.2%. Part of last week's healthcare bid was defensive, and defensives do not lead a relief rally -- fund flows into the group have been unimpressive all year despite the price. It is ranked third and sized accordingly rather than defended out of yesterday's conviction. REPRICE, not a rollover: the 09-02 queue at 207.06 never filled (session low 211.49) and is cancelled and replaced at the live 208.85 entry per the action-price provenance rule. The limit sits 2.38% below the close (Entry_Distance_ATR -1.24), a deeper pullback than any of the last three sessions produced -- treat a non-fill as the base case. Smallest risk of the three at 273 USD = 0.55% of book. Named risk: a hot 09-04 payrolls print sends the long end back up and long-duration biotech is sold first, filling this limit on the wrong kind of weakness -- re-validate at the open before letting the GTC stand through the print. · news
XLFFinancial Select Sector SPDR58.56WATCH58.5356.91n/an/an/aFinancials rose 1.56% as the rate-hold signal steepened the path for bank margins without a recession scare, and XLF still prints BUY at conviction 70 on a fresh 52-week breakout. But this book cannot tell whether it owns it. The 09-02 card was a MARKET order with execute_on 09-03; XLF opened at 58.00 and closed at 58.56, so it should have filled at roughly 58.00 for -- yet input/sectoretfportfolio.json still shows. If taken, the book is long up about 81 USD with a stop at 57.14; if skipped, it missed the move. The system must not guess. Per the provenance rule it is neither re-recommended at the stale 57.93 nor abandoned: it needs a human disposition -- trade buy sectoretf XLF 144 58.00, or trade skip sectoretf XLF. This is the 28th consecutive run flagging ADR-0019 UNRECONCILED on this managed_by:human book; outstanding names are XLF, XBI, IBB, XLY, MAGS, XLE and now XLV. Live entry if re-armed is 58.53 buy-stop, stop 56.91.
SPDR Dow Jones Industrial AvgNO ACTIONn/an/an/aBUY cv77, Buy_Rank 4, adj 76 -- the top near-miss. Closed +1.19% at 536.93, entry at market with Entry_Distance_ATR 0.0, and a respectable 2.34x reward-to-risk to the 546.24 target against a 532.94 stop. Ruled out on two counts: it is a one-print flip out of three consecutive WATCH readings (39, 35, 64, then BUY 77), and it sits in FailedBreakdown_Up_20D, the book's actionable-negative buy family (n=14, bias -1, hit_1w 0.429) that universal L1 and book lesson SE-1 both tell us to discount. Buying index beta at market 0.13% off the session high of a +1.2% day, 90 minutes before payrolls, is the chase this book is built to refuse.
Global X AI & Tech ETFNO ACTIONn/an/an/aBUY cv75, adj 74. Closed +1.93%, the third-best name on the board, riding the same falling-discount-rate lever that lifted SKYY. Clears the floor but loses the slot: whipsaw trail (BUY 67, NO TRADE 0, BUY 67, BUY 75), the discounted FailedBreakdown family, and an at-market entry 0.24% off the session high. SKYY expresses the same software-over-silicon theme with four stable prints and a holding-level catalyst instead.
Tech Select Sector SPDRNO ACTIONn/an/an/aBUY cv75, adj 74. Closed +1.29% after being one of only two red names on 09-02 -- the rate-duration veto lifting in real time. Declined on structure, not direction: a single BUY print off a decaying WATCH trail (62, 63, 57), the FailedBreakdown family haircut, and an at-market entry at the high. The theme is bought via MAGS and SKYY with better stability and better forward views.
SPDR S&P 500 ETFNO ACTIONn/an/an/aBUY cv75, adj 74. Closed +1.05% at 773.17, 0.11% off the session high. Broad index beta with no edge over the thematic acquires: 12-week estimate +2.2%, against MAGS at +12.4%. One-print flip from WATCH, FailedBreakdown family, at-market entry into a payrolls print. No action.
SPDR S&P Biotech ETFNO ACTIONn/an/an/aBUY cv66, adj 67 -- clears the floor by 2 points and is still declined. The trail is decaying (74, 69, 66) and it was one of only four red names today at -0.60% on a +1.2% Dow day, the clearest evidence that the defensive healthcare bid unwound with the rate scare. IBB carries the same theme with a stable double STRONG BUY and a materially better forward view; there is no case for owning both, and less for owning the weaker one.
Invesco QQQ (Nasdaq-100)NO ACTIONn/an/an/aBUY cv65, adj 64 -- below the 65 floor by a single point after the FailedBreakdown haircut. Closed +1.19% with a 12-week estimate of 788.97 (+9.9%), genuinely attractive, but the trail is a mess (BUY 69, WATCH 52, NO TRADE 0, BUY 65) and MAGS expresses concentrated mega-cap tech with conviction 90 instead of 65. No action; would re-arm on a second consecutive BUY print.
Energy Select Sector SPDRNO ACTIONn/an/an/aNO TRADE, conviction 0, NoTrade_Reason 'Consensus conflict + low conviction'. Closed -0.74%, one of four red names, completing a four-session collapse from STRONG BUY 85 to STRONG BUY 78 to BUY 63 to zero. This fully vindicates the 09-01 withdrawal of the 64.72 buy-stop. RSI still elevated at 67.6 with price 3.3% above EMA20 -- an extended name with no signal. No action.
Procure Space ETF (SpaceTech)NO ACTIONn/an/an/aAVOID cv68 on FailedBreakout_Down_20D, the strongest bearish read in the book for a second session (76 then 68). Closed +0.90% but RSI is 38.1 with price below EMA20, EMA50 and EMA200 and a Consolidation_Bearish consensus. Long-only book: no action, and firmly off the buy list.
Consumer Staples SPDRNO ACTIONn/an/an/aWATCH cv44, well below the floor, and the only defensive on the board to close red at -0.32%. Perfectly consistent with the day's rotation: staples are sold when the rate scare that bid them up disappears. Conviction decayed in one session. No action.
Consumer Discretionary SPDRNO ACTIONn/an/an/aNO TRADE, conviction 0, 'Consensus conflict + low conviction', after two sessions of AVOID. Closed +1.39%, a strong move, but price sits within 0.2% of EMA20, EMA50 and EMA200 simultaneously -- a genuinely undecided chart and an Indeterminate consensus. No action.
Industrial Select Sector SPDRNO ACTIONn/an/an/aNO TRADE for a fourth consecutive session. Closed +1.03% but RSI is 36.5 with price 2.4% below EMA20 and 2.8% below EMA50 -- a bounce inside a downtrend, not a turn. No action.
iShares Semiconductor ETFNO ACTIONn/an/an/aNO TRADE for a fourth consecutive session, and the more diagnostic of the two semiconductor reads. Closed just +0.15% on a +1.3% Nasdaq day, with Broadcom falling 4% despite an earnings beat on soft Q4 guidance. Price sits below both EMA20 and EMA50 with RSI 41.8. The AI trade has split into a bid software layer and an unbid silicon layer -- the most useful sector read this gauge produced today, and one that should propagate to ussemicon and ustech as context. No action.
VanEck Semiconductor ETFNO ACTIONn/an/an/aNO TRADE for a fourth consecutive session. Closed +0.39% against a +1.3% Nasdaq, below EMA20 and EMA50, RSI 45.7, Indeterminate consensus. Same silicon-layer non-participation as SOXX. No action.
Defiance Quantum ETFNO ACTIONn/an/an/aNO TRADE, conviction 0, back from an AVOID 69 print two sessions ago. Closed +0.97% but sits below EMA20 and EMA50 with RSI 44.8 and an Indeterminate consensus. No action.
iShares Russell 2000 ETFNO ACTIONn/an/an/aNO TRADE for a fourth consecutive session. Closed +0.40%, badly lagging the +1.2% Dow -- small caps did not participate in the rate-relief rally, which is itself a caution on its breadth. Below EMA20, RSI 46.7. No action.
SPDR Gold Shares ETFNO ACTIONn/an/an/aNO TRADE for a fourth consecutive session despite closing +1.85%, the fourth-best move on the board. Gold rallying alongside equities on a falling real yield is coherent, but consensus is Indeterminate and conviction is zero. Worth watching as a rate-cut proxy into the 09-15/16 FOMC. No action.
Destiny Tech100 IncNO ACTIONn/an/an/aBUY cv57, and the only non-ETF row in the book with an actual valuation read: OVERVALUED_VS_RI, MoS_FV -0.78, val_adj -4, Quality_Pass FALSE. Adjusted to 54, well below the 65 floor, and the quality gate would block a new entry regardless. Closed -1.38%, the worst name on the board, on a day everything rate-sensitive rallied -- its net-asset-value premium to private marks makes it a sentiment gauge rather than a tradeable. No action.

Outcome & track record

Accuracy at the time of this record.

39 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 →
-0.068 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.