This September 6 note uses September 4 settled prices. U.S. markets are scheduled to remain closed September 7 for Labor Day.

The climate is consolidation within a longer advance, with calm index volatility and uneven participation. SPY and QQQ remain above their 200-day averages, but neither has cleared its defined pivot. Memory leads the rebound; electricity and physical-AI supply chains are recovering from weak months. This supports selective setup review without assuming a broad advance has resumed.

Regime and what changes it

The recorded classifications are basing for SPY and QQQ, with compressed volatility and confirmed bases. SPY's classification cites a 6.35% range over 60 bars, below its stated 20% ceiling, with no fresh lows. The VIX check reads CALM at 14.53, using its published below-16 criterion. Its separate persistent-complacency condition is inactive.

Equal-weight and small-cap weakness over the month qualifies index resilience. The recorded breadth gauge is mixed. Dispersion is elevated: the monitored sample of 14 names moves 2.58 times as much as its benchmark on the report's 13-week measure. That is a sample observation, not a whole-market volatility estimate.

Campaign monitoring thresholds are desk-authored, using Friday's levels. They are not the classifier's official transition rules:

• Advance confirmed: SPY closes above $779.37 and QQQ above $748.65 for two consecutive settled sessions, with VIX below 16 on both.
• Base deterioration: SPY closes below $756.86 and RSP below $217.12 for two consecutive settled sessions.
• Longer-trend failure: a settled SPY close below $712.15 requires a fresh climate review.
• Review after the September 11 close. These references remain fixed for this campaign review.

Month versus day

The AI-provider cohort deserves the first read: its median rose 3.91% Friday and 10.82% over seven days, while remaining 1.29% lower over 30 days. Eight of ten names rose Friday, but only three of nine with a 200-day average closed above it. CBRS lacks that history. This is an early recovery with incomplete long-trend confirmation.

The rotation ranking places crypto and hard assets first on the month. Mixed constituent coverage prevents treating that ranking as an independently verified, uniform Friday comparison. NAND provides the clearest supported example of strength across both horizons.

The durability sample contains 85 jagged names among 291 durable names: 29.21%. Every member passes Friday date and soundness checks. This measures drawdown concentration within that sample; it does not establish a historical record or prove positioning is crowded. The earlier study's 27.6% used a distinct sample.

Rates, credit and the next releases

Borrowing costs remain demanding, while the supplied credit measures do not show an abrupt funding shock.

The ten-year yield rose from 4.63% on August 5, while the high-yield spread narrowed from 2.75%. TLT's price declined 0.95% over 30 days and HYG's declined 0.45%; price returns do not substitute for a credit-spread diagnosis. USO rose 23.57% over that window, keeping energy inflation relevant.

BLS reported 162,000 additional August payroll jobs and 4.1% unemployment. July CPI increased 0.1% month over month, seasonally adjusted. Cleveland Fed's September 4 model estimates August headline CPI at 0.36% month over month and core CPI at 0.20%; these are estimates. August CPI is scheduled for September 11, followed by the September 15–16 FOMC meeting.

Three observable risks

These proposed review conditions are not forecasts or official code thresholds. Their implications for the provisional themes are analytical inferences.

The U.S. Maritime Administration advisory is a standing shipping-risk warning, not evidence of a new weekend attack, a current closure, or lost transit volume.

Sources

• NYSE calendar, BLS employment report, BLS CPI report, Cleveland Fed nowcast, Federal Reserve calendar, Maritime Administration advisory: verified September 6; actuals, estimates and scheduled events distinguished above.

• Frozen numeric receipt: selected canonical-reader summaries, regime classifications, lane aggregates, durability counts, FRED observations and verified primary-web receipts, with source hashes. Prices/returns are the September 4 settled snapshot; FRED observations carry their own dates. Benchmarks use price, change7d, change30d, change1y, vsSma200; fixed confirmation levels use pivotPrice, sma50, sma200. Lane medians use rotation rankings[].medianChange30d/medianChange1d; neocloud medians/counts were independently computed in Python.

• Earlier Lane Tape: research/issues/2026-09-05-market-brief.md, supplied through a reviewed excerpt. Some medians disagree with the frozen rotation snapshot; this does not establish a new session. Memory membership differs (12 in rotation, 11 in regimes/earlier brief), so its aggregate median is omitted in favor of the named NAND subset. The older utility description also conflicts with its positive annual return and is not repeated.

• Scope: full regime transition code was not supplied to the analyst; only recorded criteria are described as official. Broad momentum totals contain mixed crypto/proxy dates. FRED series have different endpoints; the September 4 curve spread is not recomputed from September 3 yields. Fresh macro-news acquisition supplied leads, not proof of hypothetical events.

• Canonical price authority: research/market-engine/data/summaries/market-pulse.json, research/market-engine/data/summaries/focus.json and research/market-engine/data/summaries/neoclouds.json, September4 settled; frozen selected rows and dates in the sibling receipts.