Global Market Intelligence 2H September: Getting Closer
Series: Global Market Intelligence

2H September: Getting Closer

By
Scott Rubner

September 18, 2026

I am currently on a multi-country global roadshow, and the biggest change I have noticed is how quickly sentiment around AI has turned negative.

That does not mean we think the September weakness is finished. The supply/demand setup into month-end remains unfavorable, the technical backdrop is still working against equities, and we continue to think equities can trade lower over the next two weeks.

But the setup is beginning to change. Positioning has been reduced, sentiment has deteriorated quickly, and we are increasingly comfortable using further weakness into month-end to add back to core longs.

We will be back shortly with our full Q4 playbook. For now, the message is simple: cautious into the end of September, increasingly constructive on what comes next.

1. Sentiment Has Flipped

A. The Selloff Is Broader Than It Looks

In the September Setup we argued that the tailwinds were fading and that we would rather reduce exposure into the September window than chase the market higher. That is how the month has traded. The S&P 500 is down 1.8% month-to-date and 3.2% from the August 13 high, but the headline index has understated the damage underneath.

Nine of eleven sectors are lower month-to-date. Only Energy and Communication Services, 14% of the index between them, are higher. Since the August high, Information Technology, Industrials and Consumer Discretionary account for the entire decline in index points, while Energy and Communication Services have added roughly 40 points against them. This has been a meaningful drawdown under the surface, but importantly, it has remained rotational rather than disorderly.

How the S&P 500 Went from All-Time Highs to -3.2%
Sector Contribution to the S&P 500 Return, August 14 – September 16

Sector contribution to the S&P 500 return August 14 to September 16

Source: Bloomberg, S&P Global, as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
S&P 500 Under the Surface
Sector Weight Performance Month-to-Date and Since All-Time High (August 13)

S&P 500 sector weight performance month to date and since all-time high

Source: Bloomberg, S&P Global, as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

B. Concentration Has Cushioned the Index

The concentration of the index has helped cushion the headline drawdown. The 10 largest S&P 500 constituents are now roughly 40% of the index, and the largest weights have generally held up better than the average stock. That has allowed the S&P 500 to look relatively contained even as weakness has spread across a much larger share of the market.

S&P 500 – The Top 10 are 40% by Weight
Combined Index Weight of the 10 Largest S&P 500 Companies, Since 1995

Combined index weight of the 10 largest S&P 500 companies since 1995

Source: Bloomberg, S&P Global, as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
The Mag 7 Has Masked a Much Broader Selloff
Mag7, SPX, SPW, NDX, RTY, SOX Performance Month-to-Date

Mag7 SPX SPW NDX RTY SOX performance month to date

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

C. The Vol Market Looks Different This Time

The price action under the surface has been weak, but it has not looked disorderly. Unlike the July selloff, the current weakness has not been accompanied by the same single-stock volatility or positioning stress. Instead, the clearest repricing has occurred at the index level, where investors are finally paying up for protection.

SPX 1-month normalized put/call skew is now in the 62nd percentile versus the past year, up from roughly the 5th percentile at the start of September. NDX is similarly in the 70th percentile, while RUT has moved into the 83rd percentile.

This is a meaningful change from the start of the month. When we published the September Setup, broad-based index hedges were among the cheapest they had been in years. They are not anymore.

Index Skew is Starting to React
1-Month Normalized Put/Call Skew, 1-Year Lookback

One month normalized put call skew one year lookback

Source: Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

Importantly, that demand for protection has remained concentrated at the index level. Sector skew has only slightly repriced from the extreme lows, with Technology in just the 40th percentile, Industrials 31st, Consumer Staples 27th and Consumer Discretionary 12th. Outright sector vol also remains relatively subdued.

In other words, investors are paying up to hedge the market, but we are not seeing the same scramble for protection underneath it. Implied correlation is reflecting that shift as well: 1-month implied correlation has risen to roughly 14.5%, while 3-month is now around 13% – both the highest levels since early June.

The Hedge Bid Is Concentrated at the Index Level (This is New)
1-Month Normalized Put/Call Skew by Sector, Percentile on a 1-Year Lookback

One month normalized put call skew by sector percentile

Source: Citadel Securities, Global Market Intelligence, as of September 15, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
Implied Correlation Is Starting to Reprice
SPX 1M Implied Correlation, 1-Year Lookback

SPX one month implied correlation one year lookback

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 15, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

The speculative upside bid has also cooled: the share of S&P 500 constituents with 1-month call IV above ATM has fallen from 60% at the June peak to 45%, and the share with skew fully inverted (calls priced over puts) has halved from 24% in mid-August to 13%.

The VIXEQ/VIX spread tells the same story. After reaching record highs during the summer, the gap has continued to normalize as single-stock stress has faded. That does not remove the near-term downside risk, but it does suggest the market is entering this window with materially less embedded single-name leverage than it had in July.

VIXEQ vs. VIX
1-Year Lookback

VIXEQ versus VIX one year lookback

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
Leveraged ETF Assets Under Management
Indexed to 100 at the June Peak, Since January 2025

Leveraged ETF assets under management indexed to June peak

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

D. Retail Participation Has Faded (Expected)

Retail activity has slowed so far this month, which is seasonally typical. At the start of September we highlighted that this is historically the month when retail accumulation fades and dip-buying slows. That has been the case again this year: average daily gross notional is tracking roughly 10% below its 1-year average, while average daily net notional spent in cash equities has fallen below its 1-year average for the first time since April.

Retail Cash Equities – Net Notional
Average Daily Net Notional (Indexed to 1Y Average) Per Month, 1-Year Lookback

Retail cash equities net notional indexed to one year average per month

Source: Citadel Securities, Global Market Intelligence, as of September 15, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

Retail has not capitulated in AI names, it has disengaged. Average daily gross notional in semis flow on our retail platform is tracking 46% below the June peak, while opening bullish options premium is down 45%. Directionally, retail remains 3% better to buy semis through our Call/Put Direction Ratio. This is a participation story, not a liquidation story.

Retail Semiconductor Flow Has Slowed
Average Daily Net Notional and Average Daily Opening Bullish Premium, Per Month, Trailing 1-Year

Retail semiconductor average daily net notional and opening bullish premium

Source: Citadel Securities, Global Market Intelligence, as of September 15, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

2. Into Month-End, Supply Still Beats Demand

Sentiment has deteriorated sharply, but the near-term technical backdrop is still working against equities. For the remainder of September, several important sources of demand are either fading or already deployed, while the potential sources of supply are increasing.

A. Systematics Still Have Exposure to Sell

Systematics: there is still potential supply here. Vol-targeting exposure is still elevated, with the modeled 10% vol-target strategy at roughly 86% exposure, the highest since March.

SPX Exposure of Vol Targeting with 10% Risk
1-Year Lookback

SPX exposure of vol targeting with 10 percent risk

Source: Bloomberg as compiled by Citadel Securities as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

CTAs are still long US equities, albeit less long: our US Equity total z-score has fallen from +2.4 at the end of August to +1.1 today. That is a reduction, not a full unwind, leaving CTAs with room to sell further on additional weakness into quarter-end. At the same time, positioning in US Treasuries remains heavily short across the curve.

US Equity Total – CTA Aggregate Positioning
Z-Score of Net Positioning, Since January 2024

US equity total CTA aggregate positioning z-score

Source: Citadel Securities as of September 15, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

B. Quarter-End Creates a Cross-Asset Rebalance

The quarter-end rebalance also starts from an unfavorable cross-asset move. The S&P 500 is still up roughly 1% in Q3, while bonds are down 2.2%, increasing the potential need for pensions to sell equities and buy fixed income into quarter-end.

The top 100 US pension plans are approximately 112% funded, their highest funding levels since 2001. Strong funding levels continue to incentivize plans to de-glide and immunize portfolios, creating the potential for mechanical equity selling and fixed income buying into quarter-end.

CTAs Are Very Short Duration
CTA Aggregate Positioning in US Treasury Futures, Z-Score of Net Positioning, Since January 2025

CTA aggregate positioning in US Treasury futures z-score

Source: Citadel Securities as of September 15, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
Pension Funding – Highest Since 2001
Milliman 100 Pension Funding, 25-Year Lookback

Milliman 100 pension funding 25 year lookback

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

C. Triple-Witching Expiry Potentially Removes Another Support

Quarterly options expiry is another major technical event into month-end. Approximately $7T of US equity options exposure expires this Friday, representing roughly 25% of total US options exposure.

That creates another potential reset in the market’s technical backdrop. As these positions expire or roll forward, the positioning that has helped dampen realized moves can change materially, potentially leaving the market more sensitive to underlying flows afterward.

September Quarterly Expiry – Roughly $7T of Options Notional Set to Expire
September 18, 2026 Expiry Open Interest

September 18 2026 expiry open interest

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
September Quarterly Expiry – Second Largest Option’s Expiry on Record
Notional Open Interest by Expiry

Notional open interest by expiry

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

D. The Corporate Bid Goes Dark

Buybacks are heading into blackout. 10% of S&P 500 weight is in a pre-earnings blackout today. By September 30 that rises to 61%, and the window does not reopen for the majority of the index until November 1. One of the market’s largest structural buyers is stepping aside during exactly the window when the calendar is weakest.

Projected Buyback Window (% of Index Weight)
August 10 – October 26, 2026

Projected buyback window percent of index weight

Source: Bloomberg, S&P Global, as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

3. September Is Not Done Yet

The calendar is now entering the exact window we highlighted earlier this month. Historically, September weakness has been concentrated in the back half of the month, and the pattern has been even more pronounced during midterm years. We continue to think the path into month-end is lower.

Historically, the average midterm-year path continues to weaken from here into quarter-end. Since 1930, the average path has declined roughly 1.1% between now and September 30, before recovering through October and accelerating into and beyond Election Day. From the September 30 low, the average midterm year has gained 5.6% into year-end.

S&P 500 2H Performance – Midterm Years
Average 1928-2025, Indexed to July 1

S&P 500 second half performance in midterm years

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
S&P 500 Half-Month Seasonality – Into and Out of the September Window
Average Return and Hit Rate, 1928-2025

S&P 500 half-month seasonality into and out of the September window

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of August 30, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.
S&P 500 Half-Month Seasonality – Midterm Years
Average Return and Hit Rate, Midterm Years from 1928-2025

S&P 500 half-month seasonality midterm years

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of August 30, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

4. The Setup Changes After September

Where the setup gets more interesting is after quarter-end. The near-term technical headwinds begin to fade in October, and the starting point for positioning looks increasingly different than it did earlier this summer.

A lot of the excess we wanted to see come out of the AI trade has now come out. Semis vol has round-tripped, leveraged semis AUM is roughly half its June peak, retail participation has fallen sharply, and many equipment and infrastructure names are 30% to 55% below their highs. At the same time, the conversation on the road has flipped remarkably quickly from euphoria to fear.

Three months ago, the risk was that everyone was in the same trade. Increasingly, the risk is that everyone has moved to the same side of the conversation.

This matters for the index. Technology and Communication Services are nearly half of the S&P 500, and it is within Tech where positioning and leverage have cleaned up the most. If that complex catches a bid again, it does not take much to pull the index higher. And if AI leadership broadens again into earnings, the rally can extend well beyond the names that led the first leg.

S&P 500 Q4 Performance by Election Year Type
Average 1928-2025, Indexed to September 30

S&P 500 Q4 performance by election year type

Source: Bloomberg as compiled by Citadel Securities, Global Market Intelligence, as of September 16, 2026. Figures are for illustrative purposes only. Past performance figures do not guarantee future results.

We will save the full Q4 playbook for the next note, but several pieces are beginning to line up:

  • Seasonality flips. In midterm years, Q4 has averaged +5.6% from September 30 versus +2.9% across all years, with the average midterm path turning higher almost immediately after quarter-end.
  • Buybacks come back. More than half of S&P 500 weight is back in an open window by November 1 and nearly all of it by November 8, alongside fresh Q3 buyback authorizations.
  • Positioning is cleaner. Leverage has come out most aggressively in the same parts of the market where sentiment has deteriorated the most.
  • Retail seasonality improves. September has historically been the seasonal low point for retail accumulation.
  • Earnings return. Q3 reporting begins in mid-October following a Q2 season that delivered roughly 33% EPS growth and the steepest positive revision path since at least 2000.

We are comfortable using further weakness into month-end to add to core longs.

GMI Bottom Line

Our tactical view remains unchanged: we think equities have more downside into month-end. Buybacks are moving into blackout, $7T of options exposure rolls off Friday, systematic positioning remains a potential source of supply, quarter-end rebalancing is unfavorable, and we are entering the weakest part of the midterm-year calendar.

But importantly, we are becoming more constructive, not less, as that weakness develops.

  • Now: AI sentiment has turned sharply negative and positioning has been reduced.
  • Through September: supply/demand and seasonality still point to downside.
  • Into October / Q4: those same September resets begin to work in the other direction, first in Tech and potentially across the broader market into earnings and year-end.

We would use further weakness into month-end to add to core longs. We will be back shortly with our full Q4 playbook and why we think the setup improves materially once we get through September.

Copyright © Citadel Enterprise Americas LLC or one of its affiliates. All rights reserved.

Legal Entities Disseminating this Material: This material is disseminated in the United Kingdom by Citadel Securities (Europe) Limited (“CDGE”) authorized and regulated by the Financial Conduct Authority (“FCA”) (Registered company number: 05462867); in the European Union by Citadel Securities GCS (Ireland) Limited (“CSGI”) and its Paris Branch authorized and regulated by the Central Bank of Ireland (“CBI”) (Registration Number: C173437); in Hong Kong by Citadel Securities (Hong Kong) Limited (“CDHK”) licensed by the Securities and Futures Commission of Hong Kong (“SFC”), in Japan by Citadel Securities Japan Co., Ltd (“CSJC”) registered as a Type 1 financial instruments business operator with the Japan Financial Services Agency (“JFSA”); and in the United States of America by Citadel Securities LLC (“CDRG”) registered with the Securities Exchange Commission (“SEC”), Financial Industry Regulatory Authority (“FINRA”), and Securities Investor Protection Corporation (“SIPC”), Citadel Securities Institutional LLC (“CSIN”) registered with the SEC, FINRA, and SIPC, or Citadel Securities Swap Dealer LLC (“CSSD”) registered with the SEC, Commodities Futures Trading Commission (“CFTC”), and National Futures Association (“NFA”). Unless governing law permits otherwise, you must contact a Citadel Securities entity in your home jurisdiction if you want to use our services in effecting a transaction in any financial instruments or securities, including derivatives.

FOR INSTITUTIONAL USE ONLY; FOR PROFESSIONAL CLIENTS AND ELIGIBLE COUNTERPARTIES ONLY. This material is not intended as and does not constitute investment research. Contents of this material will be strictly limited to non-specific, generic information (i.e. macro events/topics) and are not subject to the Markets in Financial Instruments directive (MiFID II) or FINRA research rules. This material does not constitute an offer, solicitation, invitation, or inducement to purchase, acquire, subscribe to, provide, or sell any financial instrument or otherwise engage in investment activity.  Please see additional important disclosures, including disclosures that may be relevant to your country of residence or business at www.citadelsecurities.com/GlobalSalesTrading.

https://www.citadelsecurities.com/privacy/