US End-of-Day Verdict 3 Aug 2026, 16:00:00 GMT-4

Sector ETFs — End-of-Day Verdict

1 ACQUIRE (QUEUED), 0 DISPOSE -- the bar did not move, the board did. XLF: STRONG BUY at conviction 100, the engine's maximum, on a fifth consecutive buy print with the trail running -> -> 100 and momentum doubling from; queued as a GTC limit at for, 2.31% below the close and just above the anchored VWAP -- the deepest genuine discount on the board -- with a properly-sized 1.17 ATR stop at and 1.99:1 to the objective. It is the ONLY one of seven bullish candidates to pass the arithmetic gate, and it passes at two tenors rather than one. The other six fail in two clean ways: XLI (cv100), DIA (cv90), IWM (cv73) and SPY (cv70) all print 2.20 ATR stops 3.2-4.0% away against forward estimates of at most +1.6%, while XLP (adj 63) and SKYY (63) fall below the conviction floor and MAGS fails the anticipate-slot's 2-of-3 test. Yesterday's two queues both went unfilled and are now split rather than rolled: XLF is REPRICED to the level the live row prints, because everything that justified it got stronger; SPY is WITHDRAWN, because everything that justified it got weaker -- conviction, the family changed to BreakoutUp_52W, the order type changed to a buy-stop the price has already run past, and the stale limit is 3.29% out-of-the-money on a name at a record high. That halves the commitment to (25.0% of the) and removes the SPY/XLF correlation problem entirely. The uncomfortable fact is recorded rather than smoothed: XLF printed the maximum conviction on a session where it gained 0.87% against the S&P's 1.79%, its third straight session of lagging, and it is the worst S&P sector of 2026 -- the book resolves that by refusing to pay the market price, and by time-boxing the trade as a four-week swing since the 12-week estimate is negative for a sixth straight session. The tape was a record-high semiconductor day (PHLX +6%, Marvell +14%, Intel +10%, Palantir +30%) on which this book's own semis complex ended a five-session blackout by printing a BEARISH label at conviction 76 and 69 -- a finding, not a trade. Nothing is spent unless financials mean-revert 2.31% to the limit, and nothing is lost if it expires 08-07 unfilled. The book remains UNRECONCILED for a third session on the 07-30 IBB card, which must be recorded TAKEN or SKIPPED.

Latest & archive All books

Recommendations

1 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
XLF	Financial Select Sector SPDR	57.88	BUY — QUEUED @56.54	56.54	55.71	57.90 (+2.4%)	57.81 (+2.2%)	56.22 (-0.6%)	Financials have now strengthened for five straight sessions and the fund's rating has reached the maximum reading this system produces, with the healthiest trend in the basket, a momentum score that has doubled since yesterday, and the price sitting at a 52-week high. The wider market closed at a record on 4 August as the Treasury Secretary signalled a possible Strait of Hormuz deal within a day or two, which keeps oil soft and is a mild positive for lenders. Rather than pay the high, this is a GTC limit at -- 2.31% below the close and just above the anchored average price -- which is the deepest genuine discount available anywhere on this board; the stop sits 1.47% below it for roughly 2:1 against the objective. The honest caveat is participation and the source of the return: financials rose only 0.87% on a day the S&P added 1.79%, the third straight session of lagging the tape, and measured from today's close rather than from the limit the model forecasts essentially nothing over the next four weeks -- so every point of expected return here is the discount itself, and the position is deliberately time-boxed as a four-week swing rather than held. OPS: REPRICED from yesterday's unfilled limit (08-04 session low, never close) to the level the live row actually prints -- Step 7 provenance, no stale carry-forward. Conviction trail -> -> 100 (engine max), the multi-snapshot strengthening shape CLAUDE.md section 9 says to lean into; declined on four of those five sessions. PullbackBuy_VWAP = L1-favoured and this book's ONLY un-haircut actionable buy cell (bias +0, n=6). Trend 94, Momentum (named weak leg repaired), Regime Trending, Buy_Rank 1. Stop 1.17 ATR, correctly sized -- not the sub-ATR defect. Entry_Distance_ATR -1.88, the deepest discount on the board. ARITHMETIC GATE: the only one of seven bullish candidates to pass -- Est_1W +2.41% and Est_4W +2.25% from the limit both exceed the 1.47% stop distance (yesterday only the 1wk cleared); XLI/DIA/IWM/SPY all fail with max forwards of +0.8% to +1.6% against 2.20 ATR stops 3.2-4.0% away. SE-2 FIRES ON TWO CLAUSES: Est_12W is below both Est_4W and the close, and Est_4W is fractionally below the close -- sixth straight session the 12-week view contradicts the bullish label, so this is time-boxed with a planned exit at four weeks, NEVER a hold. SE-3/L5: calib_1w -0.08 degenerate, 1wk tenor low-reliability. Sized to the book convention of 25% of the declared ( -> =), not the raw 85.49%. Correlation risk from yesterday is GONE -- the SPY leg is withdrawn, so this is one order in one sector at one price, 25.0% of. Event risk in the queue window: ISM Services + ADP 08-05, NFP 08-07 (both inside), CPI 08-12 -- NFP/CPI are what actually decide bank NIM, and a below-market limit is the right exposure since a soft print pulls price toward the limit rather than gapping it away. Fill-risk precedent: yesterday's limit never came within 0.7%; the SPY analogue at 744.11 on 07-13 also expired unfilled -- expiring unfilled costs nothing. L8: place at the 08-05 open, reprice if it gaps >1.5%. · news
IBB	iShares Biotech ETF	190.15	WATCH	—	185.75	n/a	n/a	n/a	Carried open item, not a new decision: the biotech purchase proposed on 30 July was stopped out on its very first session and the book still has not recorded whether it was actually taken. Until that is written down the cash position is uncertain by roughly and the outcome cannot be graded honestly. The fund itself is recovering -- up 2.28% on 4 August and back to level with its anchored average price -- but it still produces no trade trigger, so there is no re-entry and no averaging down. OPS: ADR-0019 UNRECONCILED, THIRD consecutive session, managed_by human. Original card BUY 65 @189.96 market, execute_on 2026-07-31, stop 185.75; stopped out on the execution session (opened 189.14 = -0.43% gap inside the +/-1.5% band so the order stood, low 184.78 through the stop, close 186.41; the outcome scorer independently grades it filled/stop_hit). If TAKEN: -1.79%, -on, If SKIPPED: full Portfolio JSON still shows and the monitor reports trades_detected empty, so the carried default is NOT TAKEN -- an inference from an unchanged file, not a recorded decision; this run was unattended and could not ask. RESOLVE WITH: trade buy, or trade skip sectoretf IBB. Direction none so this line cannot contaminate the cells or calibration -- the 07-30 buy line remains the graded record. Today: WATCH 57 (trail B75 -> W49 -> W42 -> W57), +2.28% to 190.15, RSI 55.6, VWAP_Distance -0.09% (recovered from -2.49%), 0, FailedBreakdown_Up_20D so the -3 SE-1 haircut applies -> adj 54, below the floor regardless.

Outcome & track record

Accuracy at the time of this record.

28 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.

Earlier record 3 August 2026 2 ACQUIRE (both QUEUED), 0 DISPOSE -- the first this book has authorised in five sessions. SPY: STRONG BUY at conviction 100, the engine's maximum and the highest print in this book's history, queued as a GTC limit at 746.78 (1.44% below the 757.67 close, exactly on the anchored VWAP) for with a properly-sized 1.17 ATR stop at 734.77 and roughly 2:1 to the 770.70 objective. XLF: first STRONG BUY at conviction 90 after four consecutive buy prints ->, queued at 56.47 (1.59% below the close, the deepest genuine discount on the board) for, stop 55.67, also ~2:1. What changed is not the bar but the board: this is the first session in ten where a candidate's own forward estimates sit above its own entry at every horizon, and the hand-applied arithmetic gate that rejected 4-of-4 on 07-31 today separates rather than rejects -- passing SPY and XLF at the 1-week tenor and failing DIA and XLI outright. DIA is declined a second session despite a record Dow close and conviction 93, because all three of its tenors sit BELOW the 533.68 buy-stop against a 2.85% stop, and it simultaneously prints a 12-week target 5.7% beneath its own 4-week target. XLI clears the floor at adj 72 and then places a 0.91 ATR stop exactly at the session low on the family that is 6-of-10 stop-first -- the identical defect that cost money on IBB. Both acquires are discount limits, not market orders, so the no-chase rule is fully satisfied and nothing is spent unless the market comes to the book's price; both are time-boxed as <=4-week tactical swings under SE-2, since SPY's 4-week estimate sits below its close and XLF's 12-week is outright negative for a fifth straight session. The tape was a clean mirror of Friday: Trump called off an Iran strike, Brent fell over 5%, the curve eased at every tenor, ISM Manufacturing printed 55.6, and the Dow closed at a record 53,178.41. The standing XLE regret closed in the book's favour -- the withdrawn 56.93 limit looked 4.6% away on Friday and XLE gave back 1.28% today on the de-escalation, precisely the adverse-selection risk the 07-29 card named in advance. committed if both fill (48.9% of the), preserved, nothing spent if both expire 08-06 unfilled. The book remains UNRECONCILED for a second session on the 07-30 IBB card, which must be recorded TAKEN or SKIPPED. Later record 5 August 2026 NO ACTION -- 0 ACQUIRE, 0 DISPOSE, 1 QUEUE WITHDRAWN. First session in twelve where every candidate fails for a reason the book can name in one line, and where the declines are not all one gate. XLF (adj 98, highest on the board) fails the ARITHMETIC GATE, not the floor: repriced to the buy-stop the live row prints, buys a new all-time high with a 3.56% / 3.00 ATR stop against forwards of -0.12% / +0.15% / -2.54%; fourth consecutive session of under-participation (+0.21% on a record Dow day) while the label printed 90, 100, 98. XLV (adj 73) clears the floor and fails on forecast + geometry: all three Est_* sit BELOW the entry (-0.29% / -0.29% / -3.85%) and the stop at is EXACTLY the session low, 0.53 ATR, in the family whose stop_first_rate is 0.70 -- LLY's +48% revenue quarter and XLV's record highs are real and do not repair that. IBB (adj 64) MISSES THE FLOOR BY ONE POINT and is the best-constructed row on the board -- the only clean ascending forward ladder (SE-2 does not fire), 4.00:1 vs its stop; the floor is a hard bar and is not waived for a good story, but it is armed at raw conviction >= 69. All four BUY - ANTICIPATE names (XLK 82/72, DIA 80/70, XLI 78/68, SPY 65/55) are SLOT-INELIGIBLE on the same rule -- each printed anticipation for the FIRST time today, where >=2 is required -- and all four independently fail the gate (XLK 9.59% stop vs +4.33%, DIA 4.11% vs +0.11%, XLI 5.08% vs +0.55%, SPY 4.10% vs -0.97%). DISPOSE: book is flat, nothing to trim or exit. Aug-5 tape was a clean defensive rotation -- Dow +0.49% to a record 54,349.12 while the Nasdaq snapped a 4-day rally (-0.83%), GLD +4.14% (the largest move on the board, printed NO TRADE cv0), XLV +1.27%, IBB +0.87%, XBI +0.74% against XLE -2.07%, SOXX -2.12% round-tripping yesterday's +6.80%, UFO -2.06%, QTUM -1.93%. Yesterday's own framing was confirmed rather than merely honest: 'if the limit does not fill there was nothing to make' -- it did not fill, and there was nothing to make. : committed, intact.