Welcome to our regularly updated market letter. In each edition, we provide a comprehensive analysis based on our regional matrix, core convictions, and technical charts. We begin by « Parodos », the introduction, where we outline our investment themes, positioning and strategic views, then continue by delving into the current market focus, offering in-depth analysis of key trends and developments. Next, we present the quantitative factors of our allocation tool The Matrix currently influencing market dynamics, alongside weekly updates from our quantamental portfolios. Given the extensive content of this letter, feel free to navigate directly to the sections that interest you most. We’ve also expanded our collaborative efforts, welcoming contributions from esteemed writers such as AdamS, Yuri, Tey au jasmin or Ozymondias, to enhance the breadth and depth of our analysis.

Stouff Capital SA, 2026

Letter September: Fed hiking, “AI Slowing” new fractal, seasonal weakness ending, French bashing, Brazil electoral surprise, extension of Middle east war but sentiment has switched to contrarian bullish

Quick links

• Positioning and recent changes
• Our matrix
• S&P500 Charts and trading plan
• Factors

Parodos

From our Algo writer : ” Parodos — The month of September presented a complex macro backdrop, characterized by a notable spike in retail fuel prices to historic highs and a subsequent Federal Reserve rate hike amidst the ongoing Iran conflict. Despite these significant geopolitical, fiscal, and political headwinds—which have driven presidential approval ratings to record lows ahead of the U.S. midterms—equity markets demonstrated remarkable resilience. This durability was primarily anchored by the enduring strength of the Artificial Intelligence sector. A brief, politicized attempt by frontier AI labs to engineer a regulatory slowdown was swiftly defused by a coordinated policy response, reaffirming our positive structural outlook for AI value chain (AIVC) equities heading into year-end while leaving us cautious on crypto risk-reward.”

Coming events susceptible to move the market this week can be found here.

Trades recap and Quantamental portfolios

Our regional SC portfolios can be found here. On the top right corner, under Portfolio name, select any of our portfolios (SC SPX long for the US quantamental portfolio, SC SXXP long for the European portfolio, etc…). Then, all the components of the selected portfolio are sorted by sector, with their respective quantitative grades and recent performance measures. These portfolios are updated at least on a weekly basis after the calculation of the quantitative grades of our universe of the most liquid global stocks. First, our algos identify the best companies based on their respective 6 factors, which are weighted with respect to the current market cycle. Then, this purely quantitative crop of the cream is generally halved based on fundamental/ thematic/ risk-reward/ esg or other discretionary purposes.

We have adapted our Long/Short quantitative model to a Long Only focusing on US and Europe. To get access to our last webinar please click here

As of September 08th, our portfolio’s net exposure is +26.34%. US is Neutral at +13.8%. We stay overweight in Emerging markets at +7% for two reasons. First, the AI cycle itself supports Emerging Markets. Roughly 40% of the MSCI EM index is now linked to AI-related industries, particularly semiconductors and hardware. Korea and Taiwan remain central beneficiaries of the memory and compute cycle. Second, the current hawkish repricing in rates and bond yields could reverse in the second half of the year. Unlike 2022, real rates, wage growth and corporate pricing power remain far less problematic. As bond yields stabilize and the U.S. dollar weakens — once geopolitical fears subside — Emerging Markets should benefit significantly (except China that is a short in our Regional Matrix). The recent AI and momentum unwind is probably over. While semiconductors (AI Semis, Memory), hyperscalers and other AI beneficiaries (Optical Networking, Liquid cooling, AI Data centers,Neocloud providers) have experienced meaningful corrections, positioning has largely reset and valuations have become more compelling. We observe through our last Crash AI fractal (July 2026) that optical networking, data centers and neocloud providers are outperforming, especially relative to memory and AI power. We also continue to expect market leadership to broaden during the second half of the year, supported by easing inflation, a less restrictive monetary backdrop and resilient corporate earnings. We believe the current rotation should ultimately create attractive opportunities to selectively rebuild exposure to high-quality AI beneficiaries, particularly within semiconductors, while maintaining our preference for companies with strong pricing power and durable earnings visibility. We get tactically more constructive on Gold miners , after +2 sigma advances among some of the less trafficked metals like Platinum. It looks increasingly likely a short-term low was in place with some runway to the $4500-$4700 neighborhood in the weeks ahead (roughly where the 200-day and some retracement levels converge). The longer-term call will sort itself out in time (it’s still negative in our trend model), but we have no hesitation stepping to the side and letting Gold (and the related equities) rally in the interim.

Main market focus for the near future are enumerated below:

Market focus (fractals) – What’s shaping the markets Today ?

---

Fractal “Brazil elections” – October 5th – Flávio Bolsonaro surged to a surprise lead over Luiz Inácio Lula da Silva in the first round of Brazil’s presidential election, making him the overwhelming favorite to win the runoff and take Latin America’s heavyweight nation sharply to the right. In Spain, telefonica and santander have big market share in Brazil. Markets: positioning was already long

• Pre-vote: the Ibovespa gained 2.6% on Friday and about 4.7% over the week, closing at 192,115, attributed the rally to long positions built on a better first-round showing for Flávio than the polls suggested. Part of the surprise is therefore already in the price. riotimesonlineriotimesonline
• Overnight: EWZ reportedly rose about 10% in overnight trading (oct 05th).. riotimesonline. In Spain, san sm rises 2% while the other banks like BBVA are down owing to euro fall with spain re-election and France turmoil.
• 2022 analog: the IBOV rose 5.5% on the first session after the first round, then fell 1.4% from that close into the runoff, while still ending 4.1% above the pre-vote level. In that cycle, state-owned companies (Copasa, Sabesp, Cemig) led the gains on privatization hopes. Zero HedgeZero Hedge
• 2018 analog (closer, since the right led the first round): the Ibovespa gained more than 10% and the real about 8% between the two rounds, followed by a sell-the-news reaction after the runoff. Substack
• Worth looking through the 2nd order Brazil plays across Europe. At least initially this should be supportive for BRL, domestic equities and risk assets across the geography. Spanish stocks at the Monday 5 October open (09:00–09:11 CET) vs Friday close: the Brazil-exposed names led the IBEX.
  IBEX 35 ~19,085 +0.46% (19,172) – Santander €11.808 +1.93% (top gainer) 16% of 2025 group profit, 2nd largest market
  Telefónica €3.314 +1.78% Vivo, the group’s largest subsidiary by revenue
  Repsol +1.1% +0.6pp Brazilian upstream
  Merlin +0.9%none, so a control
  The IBEX opened at 19,172.20, with Santander and Telefónica as the top gainers. The two laggards were Solaria (-1.60%) and ArcelorMittal (-1.36%). Repsol and Merlin come from the same open, with the IBEX printing +0.48% at 19,177.4 in that source.
  Santander is the cleanest Brazil signal. Brazil is about 16% of group profit, and the stock beat the IBEX by about 1.5pp. That implies the market repriced the Brazil piece by roughly +9%, which is consistent with the reported ~+10% overnight move in EWZ. Two caveats: Share of earnings is not share of value, since the Brazilian unit trades at its own multiple. investors generally see a Bolsonaro win as better for Santander Brasil: more fiscal discipline, lower rates, faster easing. Today’s move is consistent with that.
  Repsol is up despite oil falling. Brent was down 0.7% at the open, at $101.6. A Brazil effect is a plausible explanation, but that is my hypothesis, not something the data confirms.
  Broader market noise. The euro is at an 18-month low against the dollar, and lower oil is supporting the whole index. Part of the +0.46% is therefore not Brazil-related. The excess-return column is the more relevant measure.

What to watch before 25 October

The DI curve and the BRL, which show more cleanly than equities whether the fiscal premium is being priced out.The first post-first-round runoff polls, read with the right-wing under-capture bias described above. Endorsements from Caiado, Zema and Santos, and especially Cury, since his centrist voters are the swing vote. The Vorcaro / Banco Master case. It is a known tail risk: in May, a poll showing Flávio losing support over his links to the Banco Master owner sent the Ibovespa down 1.52%, led by banks. Any judicial development before the runoff would hit the same names. riotimesonline

---

Fractal “France elections” (1) October 5th – EUR getting hit.  Potential Spain snap election risk is another political wobble, but France is still the bigger issue. OAT/Bund spread acceleration and fiscal anxiety is clearly bleeding into the currency. Spain itself isn’t a huge tail, but it adds to the broader sense of political instability. The key this week is whether any of the alternative French candidates can produce something resembling a credible path back toward fiscal sustainability.

(2) – August 27th – French domestic equities look increasingly vulnerable to a spike in political risk premia ahead of the 2027 presidential elections. While macro tailwinds previously supported domestic stocks, France’s growth outlook remains sluggish, OATs are under pressure, and energy prices are on the rise. French domestics were hit as fiscal and political concerns resurfaced ahead of the first presidential debate which happened on August 27 at 16h30 Rolland-garos time. Feels early to build large directional shorts around an event that is several months away. That said, think France will be capped to the upside. . Weakness among bank stocks weigh heaviest on the French index, with BNP Paribas down as much as 4.5% and Societe Generale as much as 4.9%; Eiffage & Vinci drop *** The gradual heating up of the pre-election campaign comes as a widening of French sovereign spreads over summer has left France with the highest 10yr yield among the large and medium-sized euro area economies

read more on france fractal below by clicking on the arrow:

Polls currently show far-right candidate Marine Le Pen leading, but the risk of a hung parliament (and further fiscal slippage) following a parliamentary election remains elevated regardless of the ultimate winner of the presidential election. We recommend hedging a pick-up in election headline noise with our French Domestic basket, which screens as the most sensitive slice of the market to political risk given its 50% domestic sales exposure (vs. 15% for the CAC 40). Next catalysts include the 2027 Draft Budget Release on Sep 30th, followed by the Socialist Party Primary on Oct 10th and 17th.  The correlation of French Domestic vs International stocks to credit spreads tends to pick up during political shocks

GSXEFRDO has underperformed EU peers (SX5E) with Far-Left leader Jean-Luc Melenchon rising in the polls

The 2027 French elections represent a critical pivot, but our economists argue the primary risk is political gridlock and fiscal slippage. Incoming fiscal news has been disappointing: Prime Minister Lecornu recently acknowledged that this year’s deficit target is out of reach, leading GS to raise its deficit forecasts to 5.3% for both 2026 and 2027 (implying zero deficit reduction). With public debt already exceeding 110% of GDP, political gridlock will only exacerbate fiscal pressure. GS’ statistical model continues to show far-right leader Marine Le Pen best placed to win the presidency at 68% probability, though far-left leader Jean-Luc Mélenchon’s odds remain significantly lower than implied by prediction markets (now 14% for Polymarket). But more crucially, the model points to a high likelihood of a hung parliament in legislative elections regardless of the presidential winner, a scenario that would temporarily worsen the structural fiscal balance by 0.2% of GDP and push 10Y yields up by 10bps.

GS Research’s Elections Model Suggests Le Pen Is Currently Best Placed to Win the Presidency

European Daily: France—No Deficit Reduction This Year and Next

The Debt-to-GDP Ratio Is Set to Continue Increasing with GS Research simulations showing a 15% probability that it surpasses 130% by the end of the decade

French credit spreads have been under pressure in August (+8bps for the 10y OAT-Bund) due to a combination of energy, fiscal, and political factors. But we believe the political risk premium could build up further into Q1 2027 as polling accuracy historically improves, similar to what happened in 2017. Historically, election-driven political uncertainty has widened OAT-Bund spreads by up to 30bp, which would push spreads to ~100bp (vs. GS 70bp year-end target and 85bps today). We expect market sensitivity to fiscal uncertainty to remain high given upcoming budget negotiations and uncertainty around campaign economic proposals, and our rate strategists see upside risks to their end of year forecasts.

France has been penalised the most in the current environment

Global Rates Trader: Back to Front-End

2017 is the most relevant precedent, with the bulk of the widening still happening in Q1

10y OAT-Bund under/outperformance against PC1-implied. Indexed at 0, 60d before the 1st round of the presidential election

Global Markets Daily: Early Thoughts about OATs into France’s 2027 Elections

Source: Goldman Sachs FICC & Equities, GIR, as of Aug 2026

While credit markets show caution, equity markets have been cushioned by broader European inflows. In particular, domestic equities have benefitted from extensive fiscal support, insulation from protectionist policies, and favourable FX dynamics. But our economists have now painted a worrying forward picture, with higher energy prices, the recent heatwaves and wildfires, a weakening labour market, and election uncertainty likely to keep GDP growth subdued. Still, our French domestic basket is up 42% since Jan 2025 significantly outperforming international peers (GSXEFRIN +1%), and the regional benchmark (CAC +12%). The latter has proven to be an inefficient hedge for domestic political risk because top weights (LVMH, Schneider Electric, Airbus, etc.) derive most of their revenue globally. Historically, GSXEFRDO is ~1.6x as reactive as the CAC 40 and ~1.9x as reactive as the SX5E on days driven by French political shocks, offering the cleanest equity expression of domestic risk.

French Domestic have outperformed their International peers and the regional benchmark over the past year

Source: Goldman Sachs FICC & Equities, BBG, as of Sep 2026

GSXEFRDO has proven much more reactive than regional benchmarks to recent French political shocks

Date	Event	GSXEFRDO (%)	CAC (%)	SX5E (%)	OAT-Bund Chg (bps)	OAT-Bund Prior Close (bps)
24 April 2017	2017 1st round (relief rally)	6.46	4.14	3.99	-18.5	68.7
10 June 2024	Snap election announcement	-3.05	-1.35	-0.69	7.8	47.8
14 June 2024	Snap election — worst week close	-3.17	-2.66	-1.95	7.1	69.6
01 July 2024	1st round 2024 results	2.15	1.09	0.73	-5.7	79.8
05 December 2024	Day after Barnier fell	1.89	0.37	0.66	-5.6	83.8
08 July 2026	Le Pen cleared to run 2027	-2.49	-2.18	-1.82	3.7	79.8
27 August 2026	2027 presidential debate (Medef)	-3.1	-1.68	-0.71	-0.3	85.5
Average	Move per 10bps OAT-Bund widening	-3.89	-2.45	-2.09
Sensitivity	Compared to GSXEFRDO	1.0x	1.6x	1.9x

Source: Goldman Sachs FICC & Equities, BBG, as of Sep 2026

Despite the recent sell-off, GSXEFRDO trades near the upper end of its 5-year valuation range and 40% above its 2024 trough, implying limited political risk is currently priced in. Historically, GSXEFRDO’s correlation to credit spreads has spiked during periods of political uncertainty with basket selling off by ~ 4% for every 10bps of OAT widening (vs. 2.5% for CAC). Because timing headlines early in the race is difficult, we recommend using late Q4 2026 or Q1 2027 options to capture the acceleration in the fiscal and election calendars, and we favour pair trades for delta-1 implementation. We recommend using GSXEFRDO as a funding leg against three long themes: 1/ EU Renewables (GSXERNEW) to capture structural electrification and energy mandates; 2/ German Fiscal (GSXEGFSC) to play cross-border policy divergence following the rollout of Germany’s fiscal plan; 3/ French International (GSXEFRIN) to isolate pure domestic political risk while maintaining country-neutral exposure.

GSXEFRDO valuation sits at the upper end of 5y range and looks vulnerable to a spike in political noise

Source: Goldman Sachs FICC & Equities, Marquee, as of Aug 2026

Relative positioning on the long German Fiscal vs Short French Domestic trade is attractive post the summer de-gross

Fractal “AI” –  Here we present the various fractals related to AI, like the Dario AI slowing statement (On Sept 8, former Anthropic (and OpenAI before that) employee Jacob Coxon wrote this tweet thread. It currently has 174mm views- Dario and Sam Altman immediately commented in agreement. Four days after the Coxon tweet, Dario wrote the essay “We Must Pace the Frontier”. In it, Dario begged US and international governments to regulate him and his industry in order to force a slowdown across all AI labs globally.), The whole move was obviously calculated. Planned for. Strategized. But what could have been a real potential risk event for AI Value Chain (AIVC) stocks, seems to have come and gone in a period of three weeks. In the initial days after the Coxon tweet and then the Dario essay, it was looking like the left was gearing up to try and make this a thing. You could imagine a scenario where Dems decide to make AI slowdown/regulation a major campaign point into the mid-terms. And if Dems sweep, the market could start getting worried about regulatory slowdown. Regulatory slowdown causes delayed or cancelled capex- that hits AIVC stocks. In the second week of Sept, this felt like a real risk. I think that risk has dissipated, swatted away by a united right. There is the potential for Hyperscaler capex to slow down for other reasons. But I don’t think that’s an October risk, so we’ll save that discussion for another month. Hyperscalers report at the end of October and my base case is the earnings will generally be good, and the capex projections will be confirmed, and AIVC will broadly trade well into YE.

AI slowing details

You could tell, in part, by how they wheeled out all these goofballs in perfect unison. Obama, Kamala, Gates. The big guns. Immediately the topic became politicized. Trump called it a hoax. David Sacks publicly torched the entire effort. Jensen Huang did the same. As did Zuck and a bunch of other right leaning power players. The right, in both business and politics was just as unified in its denouncement of these AI scare tactics as the left was in support of it. Trump, Sacks and others were quick to point out that these AI leaders were begging to be regulated, so as to create a regulatory moat that prevents competition and ensures those companies inside the moat will be successful. There is a long history of companies with new technologies acting like this. This is how telephone companies acted. This is how electric utility companies acted. This is how radio broadcasting acted. Airlines. Trucking. Meatpacking. They all ran some version of this playbook – help the government establish rules under the guise of public benefit that ACTUALLY just benefits the companies asking for the regulation. Then you make sure the people in charge of regulating you are your friends and preferably, former employees. And then you enjoy decades of easy profits ensured by regulatory moats.

This is exactly what AI companies like Anthropic and OpenAI want. The major threat that these companies see on the horizon are open-source models. These models have been able to keep pace with state-of-the-art (SOTA) closed models much better than experts were expecting a few years ago. Open-source models are now only 3-6 months behind SOTA closed models, and this gap has been shrinking for several years. This setup very clearly calls into question the viability of the business models of these frontier AI labs. If they are burning tens/hundreds of billions of dollars on compute to train and run models that are only 3-6 months ahead of open-source models, that may not be a business that’s worth a trillion dollars or a few trillions dollars. In fact, it might not be a business that’s very valuable at all. That scares the crap out of Dario Amodei and Sam Altman and a lot of these other guys, whose fortunes and futures are tied directly to closed-source frontier AI labs being very valuable. These guys want to win. They are incredibly competitive. That’s how they got into such a position of wealth and power in the first place, by being extremely competitive. They want to win. They will tell you they want to win in the name of Effective Altruism. “AI safety” is of such tremendous consequence according to the EA ideology, that competing to develop the safest, best AI the fastest (but not too fast), is doing more benefit to humanity than anything anyone else can do ever.

And that may be true. But it could also just be the age-old thirst for power. Pride. The oldest sin in the book and it’s been a hall of fame sin for humanity ever since. So it could just be the prideful thirst for power.

Or it could be some mix of the noble motivation and the prideful motivation. And that’s probably what it is. That’s usually what it is.

Fractal “AI” – the Sept 28 Musemageddon (DASH (-6%), CVNA (-5%), DUOL (-5%), BSP (-4%), SHOP (-3%) or; September 13th “AI Slow” fractal (Dario Amodei’s call to “pace the frontier,” backed in varying degrees by Altman, Musk and Hassabis, has hit the Asian AI complex hard. SoftBank was down double digits and the broader complex weaker. if safety and alignment concerns force the frontier to slow, regulation becomes another reason to compress AI multiples and potentially question the durability of capex), or the July 2026 Crash – Following the July bottom, the AI complex’s bounce stalled out, prompting questions as to whether the correction was purely technical or fundamentally driven. Please click on Thomas report here. The chart below of the Nasdaq100 shows a lower high while SPX made an all time high. This shows tech stocks are underperforming the market and we sold it on August 18th:

Datacenters midterm election AI October 2026 fractal

November’s midterm elections are becoming elections about data centers and the impact on affordability. It is unlikely for elected officials to campaign on an issue, win the election, and then not attempt to deliver, particularly when the issue polls so well. In fact, the election may elevate data centers to the top of the public policy agenda. As such, the gubernatorial races may be more impactful for the data center trade than congressional elections. Virginia, the largest data center state in the country, is a great example of what can be in store for other states postelection. It enacted an energy tax on data centers this year after electing a new governor in November. In fact, Republicans are struggling to hold key data center states. And as the chart above shows, data center trade performance relative to the S&P 500 is moving inversely with the Democrats’ lead in gubernatorial races. We are not too worried about what a Dem sweep can do to the data center trade although we know the level of scrutiny will go up and will help the governors’ efforts to tax. We would only be more concerned if the Democrats use the debt ceiling as leverage and force Trump to remove
the 100% expensing of factories and capital equipment for data centers. In our view, this would change the economics significantly.

🤖 The AI Complex: From Technical De-risking to Fundamental Discernment

Important events to analyse fundamentals going on: Aug 17th Anthropic PBC is on track to generate annualized revenue of more than $65 billion based on its current performance, up more than sevenfold from its pace at the end of last year. August 20: model economics keep improving fast. Qwen3.8 is another example increasingly useful intelligence from increasingly small/cheap models. That is very bullish for AI adoption and the broad market, but less obviously bullish for everyone selling tokens/model access. Moats get harder to defend. Same argument for parts of the picks and shovels chain… think whole space keeps de-rating as EPS moves up.

During the July sell-off, all AI themes were sold aggressively at similar rates relative to their volatility. However, the subsequent rebound reflects growing investor selectivity. The 1 st group includes many of the prior momentum names, Semis in particular, that flushed into the Situational Awareness lows of late-July and have now bounced w/ their long-term uptrends intact.

Give deference to the names that have already retraced a majority of their decline (e.g., 61.8%) and have quickly reclaimed the flag of leadership (DELL, ONTO, NVT, LITE, KEYS, ENTG, etc.). Be more alert with those that remain below the 50-day MA or that haven’t rallied with much relative gusto… if we had to guess, they chop around without much direction from here. The 2 nd group is the batch of structurally weaker charts that have rallied over recent weeks as part of the anti-momentum cohort – think ACN, ADBE, BAH, CRM, FICO, IBM, etc… overbought conditions in downtrends are often a good signal to step away, and we’d advise doing so there.

Theme	Asset Ticker	Drawdown (Peak to Trough)	Bounce (% Up from Trough)	Distance to Peak	Up Move / Vol Ratio
Optical Networking	GSXUOPTI	-41%	+32%	28%	0.43 (Strongest Rebound)
Neocloud Providers	GSXUNEOC	-40%	+20%	39%	0.30
AI Data Centers	GSTMTDAT	-31%	+17%	24%	0.30
AI Semis	GSCBSMHX	-27%	+14%	20%	0.21
Liquid Cooling	GSXUCOOL	-18%	+10%	11%	0.27
Memory	GSTMTMEM	-41%	+12%	51%	0.17 (Lagging)
AI Power	GSENEPOW	-18%	+6%	15%	0.17 (Lagging)

Fundamental Shift: Emerging Themes in Tech

1. Software & The “Inference Economy”: Investors are turning to model-agnostic and bottleneck-agnostic software beneficiaries of AI adoption. Key names in security, data infrastructure, and tools include SNOW, DDOG, PLTR, CRWD, PANW, OKTA, and TWLO.
2. Memory Dynamics: Memory (DRAM, NAND, HDDs) is shifting from a price-driven upside narrative (earnings revisions) to a debate over long-term stability, long-term agreements (LTAs), capital returns, and multiple expansion.
3. Levered ETF Technicals: In U.S. Semis levered ETFs, total AUM sits near ~$99B (down from a high of $157B). While retail “bought the dip” in early July (adding ~$15B of excess length), the August rebound has been driven purely by mark-to-market spot gains, accompanied by net share redemptions and selling into the rally.

---

Our AI fractal Exposure score :

Here we show the 50 companies with the strongest total thematic exposure, taking into account breadth and materiality.

Exhibit 1: Global companies with the greatest Multi-Thematic exposure (OW-rated, MC>US$10bn)

To the contrary, AI Challenged stocks have lagged materially since the start of 2023. Here we screen for companies with a materiality of Core-to-Thesis, Significant or Moderate exposure to the AI Challenged category. On a US$ total-return and equal-weighted basis, the group is up 50% versus 105% for MSCI ACWI (2% versus 24% in the last 12 months), with underperformance accelerating from late 2025. This occurred as more powerful AI models continued to emerge, but importantly as enterprise adoption started to accelerate via AI coding and co-work functionality.here we show the top 50 companies most-challenged by AI and thematic disruption.

Exhibit 2: Thematically Challenged – Top 50 stocks by combined AI Challenged materiality and Disrupted/Deflationary exposures

Fractal AI – idiosyncratic drivers outside of AI in Europe

---

Fractal “Clash of Civilization” – Iran War fractal – Sept 15th: Distillates have gone vertical again and physical markets remain incredibly tight. Saudi’s East-West pipeline disruption forced the suspension of Yanbu loadings and cancellation of some European cargoes, with European physical crude trading north of $130 in places yesterday. Despite all that there are reports of more visible signs of cargoes moving through the strait. “ Iraq’s seaborne crude oil exports from its southern Gulf terminals averaged 3.16 million barrels/day in the first 10 days of September, nearing the prewar levels of 3.335 million b/d recorded in February” . China increasingly feels like an important potential catalyst. Araghchi is in Beijing for talks with Wang Yi today, while Bessent meets He Lifeng this weekend ahead of the planned Trump-Xi summit on September 24. Iran is expected to feature in those discussions. China has meaningful economic leverage with Tehran and a direct channel into Washington: if Beijing wants to use both, that creates a credible bridge toward an off ramp. Feels like the key potential diplomatic pathway to watch. 

Sept 8th: US and Iran kinetic activity continued this weekend as both sides traded attacks on ships. US military struck 3 Iranian oil tankers Saturday in response to alleged ballistic missile attacks by IRGC against US warships. Iran retaliated by targeting six additional ships, including three oil tankers in the Strait of Hormuz, and three others they claimed were affiliated with US. Oil rally accelerated Tuesday after Houthi missile and drone strikes targeted Saudi Aramco facilities, causing temporary halt in some operations, wounding over 70 people. Elsewhere, Iran, Oman finalizing agreement to establish temporary safe shipping route through Strait of Hormuz to manage energy transit.
• Regional escalation: Houthi attacks on Saudi energy facilities raises fears of wider regional escalation (FT)
• Iran-Oman: Iran threatened US with economic warfare, plans to declare new Gulf restricted zone and announce SoH deal with Oman (Bloomberg)
• Naval blockade: US naval blockade strangles Iran’s crude export revenue as offshore stockpiles dwindle (link)
• Latest forecast: Goldman Sachs said oil may rally to $120 a barrel if attacks on shipping in Middle East increase (Bloomberg)
• Hormuz traffic: Hormuz traffic fell to seven commodity vessels on Monday (Reuters)

We use the following fractal denominated Iran War. The lists below identify stocks that have out (under)performed in all three phases of the year … 12/31 to 2/27 (pre-war), 2/27 to 3/30 (during the war), and yesterday (the potential off ramp / TACO / etc.). 50 names in the Russell 1000 pass the test, and while it’s an eclectic list, it’s a good starting point at identifying durable leaders in a tricky environment. Paired with our factors, we like Netflix, Ross Stores, PNC, Alcoa, Lumentum. We run the same methodology but also for names underperforming all three phases of the year (95 stocks). Then , we look at the ceasefire fractal on April 8th: Stocks that benefited from the war (energy, commodity chemicals, defense contractors) got sold (HF profit taking) while highly shorted consumer discretionary (including housing-linked names) saw the most consistent covering. Long Only community buys what was working before the war started…memory, semi cap, optical,  and keeping a close eye on Mag 7 complex. 

See more details on the fractal for Iran

update april 8th: ceasefire taco –> what is doing well – back to pre martch pattern

The last US strike on Iran was Sunday June 22, 2025. From 13 June 2025, Israel launched a series of strikes against Iran’s nuclear program, ballistic missile sites, and energy facilities, stating its aim was to stop Iran developing a nuclear weapon (Iran says it has not been developing a nuclear weapon). In response, Iran launched missiles and drones against Israel. Most of Iran’s missiles and drones against Israel were intercepted. The United States also conducted strikes on 22 June 2025. In response, Iran launched missiles against a US military base in Qatar. In the right hand table, here are moves across UST yields, spreads, vol, SPX. From 1wk prior to the strikes to 1d, 2d, and 1wk after – The market buildup in 2025 started earlier in the month, with Israel striking on Jun 13. After the US strike, the flight-to-quality reflex faded quickly as a ceasefire came through on Jun 24, and the hit to spreads unwound as vol selling re-emerged. Curve continued to bull-steepen, but more on Fed dovishness

The below chart shows the impact of the 2 strikes on the DAX (max down 3%), the S&P500 (down 2%), dollar DXY up slighty from 13th to 25th, crude oil up 10% before falling 15% after the US strike. Nikkei stayed very strong. Bitcoin was down 55 from 13 to 25th before rebounding very strongly.

here is below the immediate returns after the launch of the war:

Stock Market And Iran War – The stockmarket displayed such resilience in recent geopolitical blowups. On June 13th, 2025, when Israel began a bombing campaign againstIran’s nuclear and military sites,the S&P 500 fell 1.1%to 5,976.97 and dipped as low as 5,943.23. By the time the 12-dayconflict ended,the indexwas upnearly 2%. OnJune23, after the U.S. bombedIran’s nuclear plants, the S&P 500reversed higher and closed with a1%increase.
In January, the U.S. capture ofVenezuelanleader Nicolas Madurobrought muted investor reaction.The S&P 500 rose 0.6% and theNasdaq added 0.7%on the Mon-daythat followedtheweekendop-eration. The event didn’t alter thestockmarket’s dull trend.

Covid ressemblance: In April 2020, oil markets experienced a shock that most investors had never seen: roughly 20 mb/d of demand vanished as Covid froze global mobility. Prices didn’t just fall; they broke, culminating in the once-impossible – negative oil prices. This week, we are staring at a shock of comparable size, but with the sign flipped. The Strait of Hormuz – through which roughly 20 mb/d of crude and refined products normally flows – has been effectively closed since last weekend. The duration may prove far shorter than Covid. But the initial magnitude of the impact is similar: the world is suddenly “short” a volume that, in normal times, would dwarf almost any supply/demand imbalance we debate.

---

---

Fractal “Clash of Civilization” – CANADA trade war . RETALIATORY TARIFFS: Canada is matching Trump’s weekend tariff escalation with a sweeping package of counter tariffs that directly targets U.S. steel, aluminum and downstream manufacturing. Prime Minister Mark Carney will double Canada’s existing counter-tariffs on U.S. steel and aluminum to 50%, while adding new 50% duties on U.S. furniture, clothing, electronics, smartphones, and gaming consoles. Overall, the retaliatory levies cover roughly $20B in annual imports. The measures take effect on September 8 and represent a sharp reversal in Carney’s earlier conciliatory stance after trade talks between Canada and U.S. collapsed on Friday. Publicly traded companies in the space include ArcelorMittal (MT), Cleveland-Cliffs (CLF), Nucor (NUE), Steel Dynamics (STLD) and U.S. Steel (X). Publicly traded companies in the aluminum space include Alcoa (AA), Kaiser Aluminum (KALU), Rio Tinto (RIO), Century Aluminum (CENX), and Constellium (CSTM).

---

Fractal The Art of Tragedy – Fed Sept 16th, AdamS:  “The FOMC raised the funds rate by 25bp to 3.75-4% on Sep 15th. While the hike was since the cpi data expected, the meeting was more hawkish than we expected in a few ways. First, a 16-2 majority projected at least one more hike this year, and there were no dissenting votes against today’s hike. Second, the median funds rate projection remained quite elevated through 2029, and the median neutral rate dot rose from 3.06% to 3.25%. Third, Chairman Warsh described the hike as having “removed a dose of accommodation” three times. Market now expects the FOMC to deliver a second 25bp hike in October, a change from our previous expectation that September would be the only hike. We think October is the most likely time for the next move because it is most natural to deliver hikes that the FOMC presented today as supporting “a timelier return” to the 2% target at consecutive meetings. We have kept our forecast for the terminal rate unchanged at 3.25-3.5% by adding to the September and December 2027 rate cuts we already expected a third 25bp cut in March 2028.

Additional hikes are possible but not our base case. One reason is that our forecast for core PCE inflation remains below the median FOMC participant’s forecast at 3.2% (vs. 3.4% for the FOMC) in 2026 Q4/Q4 and 2.2% (vs. 2.5% for the FOMC) in 2027 Q4/Q4. Some of the gap for 2026 could come from a reluctance on the part of some FOMC participants to pencil in a downward revision from methodological revisions that will be implemented later this month, which we estimate will be worth -0.2pp on the year-over-year rate, until the impact is clear

fractal Art of Tragedy: how US equities perform when the Fed raises rates: i. using the last seven tightening cycles as evidence, S&P usually struggles at the start of things — the average is -2% over the first three months. ii. but, the pinch typically doesn’t last very long — the average is +9% over the next 12 months. iii. the big exception to that last point was a bruiser in 2022. iv. the eventual trajectory of the market is determined by earnings growth; the speed and volatility of the moves in rates can heavily influence stocks along the way, iv. the two best sectors to own in those early months: tech and energy. : Rising interest rates are a headwind for equity valuations, but earnings are the most important driver of stocks. The S&P 500 forward P/E has declined from 22x at the start of the year to 19x today, but the index nonetheless sits within 2% of its record high.

ENERGY A WINNER & FINANCIALS A LOSER AFTER FIRST HIKE IN A CYCLE:

fractal rate cuts/ no rate cuts baskets

See all details on the fractals for Trump 2.0 Art of Tragedy – like (1) new Fed chairman Warsch – (2) JPY coordinated intervention , (3) Bonds intervention (see below) .

(3) FRACTAL BESSENT INTERVENTION ON BONDS – August 19th: Treasury / USD:  Treasury doubled the maximum size of long end liquidity buybacks from $2bn to at least $4bn per operation in the 10-20yr and 20-30yr sectors through early November. The words “at least” matter: this doesn’t read like a hard ceiling and clearly opens the door to doing more if required. Mechanically this isn’t QE and they still have to fund whatever they buy back. There also isn’t some requirement that every operation is perfectly duration neutral… Treasury manages duration across the whole funding program. But the signal seems pretty clear: they are willing to use buybacks and issuance composition more actively to stabilize the long end. Call it quasi-Twist, Twist-lite, whatever… the label matters less than the reaction function. Gold spoke loudest yesterday, responding sharply to the TSY buyback headlines with its second +2σ advance of the month. Bitcoin is similar and followed with a +2σ move of its own, trading back above its 200-day MA for the first time in roughly 10 months, as the DXY went the opposite way, USDJPY was swatted lower again, and the 2/10 curve flattened.I actually think the more lasting implication may be for the dollar/gold/cryptos than rates. Rates can eventually re-clear higher because the underlying supply hasn’t disappeared. But if the market starts believing that sufficiently aggressive long end tightening will be met with larger buybacks, shorter issuance or other liability-management tools, some of that adjustment has to migrate elsewhere. A flatter curve and weaker dollar feels like a pretty logical expression of that. For equities/gold AdamS think that is pretty bullish. The important information yesterday wasn’t whether $4bn itself changes the supply/demand balance… somewhat marginal. It was learning that the response to pressure in the long end can include changing the way Treasury manages duration. But according to AdamS: “It raises the risk of higher inflation if one tries to control the yield curve. That will put pressure on Fed to keep rates low.”

---

More positively, The market isn’t as pricey below the megacap surface. The Invesco S&P 500 Equal Weight exchange-traded fund goes for 17 times earnings. The S&P SPDR Small Cap 600 ETF has a forward price/earnings ratio of 15. Both ETFs have trailed behind the S&P 500 for years but can help diversify exposure to the tech-heavy megacap index. Of course, valuation is useless if one does not incorporate expected future earnings.

3Q Earnings:  Aggregate S&P 500 earnings growth is tracking well above consensus estimates this quarter, even adjusting for non-recurring “other income.” S&P 500 EPS growth is tracking 45% YoY in Q2 compared with a consensus estimate of 22% coming into the quarter. However, 19 pp of that growth is attributable to Alphabet and Amazon’s combined $151 billion of “other income” related to equity investments. Microsoft contributed an additional $3 billion of “other income.” Excluding these gains, S&P 500 EPS growth is tracking at 26%, an acceleration vs. Q1 and the fastest pace of growth since 2021. EPS growth for the median S&P 500 stock is tracking at 12% year/year, also exceeding consensus estimates, which pointed to 9% growth at the start of the season.

S&P 500 Q2 2026 EPS growth has accelerated, even adjusting for non-core sources of income…

In addition to strong backward-looking results, Q2 reports have driven continued upward revisions to analyst 2027 earnings estimates. Since the start of Q3, consensus estimates for S&P 500 2027 EPS have been revised up by 1%, with the strongest revisions to Energy and Financials. Broad based upward revisions to 2027 earnings have been reflected in continued positive revision breadth across the S&P 500. Based on the below chart, S&P 500 EPS growth of +22.75% year/year in 2026 (to $351.54), +15% in 2027 (to $402.18), and +12.13% in 2028 to 454.06, thanks to solid US GDP Growth, weaker dollar, and continued earnings strength amongst technology stocks.

Globally, the landscape feels rather robust as well: with also strong earnings season in Europe. AdamS notes a +25% net skew to beats & +8% to guidance upgrades – above prior quarters, particularly for guidance. The earnings revisions ratio has moved into positive territory led by Semis, Energy and Div Fins. Median 1Q EPS growth +5.5%Y, up from +2.0% in 4Q , but of course lagging S&P 500’s +16%. On her management sentiment score change screen, Top Picks in the biggest improvers list include BASF, Talanx, Santander, ASML and Nokia. Indeed, after a soft start to Europe’s earnings season, things have improved. SXXP estimates have rebounded +12.6% YoY growth in 2026 to 40.79 +8.57% in 2027, and +9.02% for 2028.

European valuations are much cheaper than US and global ones, and P/E for Germany and Europe have fallen from to 18.7 to 15.3 and from 17 to 16. The ratio of European valuation versus the world ones has rebounded. We added in the below chart the relative valuation of Europe versus the S&P500 in grey, the S&P500 value and the MSCI World excluding US stocks.

Emerging and Chinese markets valuations are rising above neutral:

Earnings for Emerging equities ex China are exploding: +71.14% in 2026, +22.63% in 2027.

SECOND PILLAR OF MATRIX: LIQUIDITY is still strong

Update August 2026 – The Federal Reserve has cut the Fed Funds rate by an equivalent of 25 basis points in 2026 via its
$200bn+ balance sheet expansion since December. This balance sheet expansion was enacted not to help the Fed with its twin mandates for inflation and employment, but to ensure enough liquidity in the financial system. Most of the Fed members pushing for a higher Fed Funds rate support the balance sheet expansion and don’t believe balance sheet expansion increases inflation. Warsh seems to have a different view. Warsh’s opening statement last week specifically noted that the FOMC
discussed how much accommodation is coming from the balance sheet. Warsh is likely arguing behind closed doors that it’s contradictory to push for rate increases to stamp out inflation when the balance sheet is increasing. Still, cutting the balance sheet is not easy. Warsh’s point is that the balance sheet is asymmetrical, easy to expand for stimulus, but nearly impossible to remove the accommodation. Our sense is that Warsh’s balance sheet task force will recommend using financial deregulation to sterilize the negative impact of draining liquidity via the balance sheet. But Warsh needs to act before then given the pressure building among Fed members and the lack of resolution in Iran. The first part of that plan may show up this week at Treasury’s Quarterly Refunding if Treasury enacts its plan to invest a portion of its General Account (TGA) into the repo market. If enacted, Treasury will inject up to $200bn of reserves into the banking system, giving Warsh the cushion to remove the Fed’s year-to-date balance sheet expansion, which, as noted above, is the equivalent of a 25-basis point rate increase. We also believe this creates space to wind down the Fed’s $10bn of Reserve Management Purchases (RMP). The net effect is a more neutral balance
sheet policy that begins the process of tightening without raising the Fed Funds rate.

One pillar of Liquidity: Money growth. The chart below shows yoy US Money supply growth rising from 4.7 to +5.41% in the last weeks. Europe yoy Money supply is positive at +3.38, but slightly decelerating. Chinese Money supply growth seems to have bottomed and stands at 8.6%. Last but not least, Japanese Money Supply has stopped its declining yoy growth at 0 and is rebounding to +1.7%.

Another pillar of Liquidity – Central banks assets. The FED and the ECB are no longery reducing their balance sheets. The chart below shows that Fed assets stopped falling at a floor of 6.55Tn$ and back to 6.735: this is reversing a proxy of our Liquidity grade from negative to neutral, as we rank the last 3 months changes in CB assets as much as 3yr and 1yr change. The Fed increased its balance sheet by a net $200bn since December. Coupled with GSE purchases of MBS, the combined effect is the equivalent of one Fed rate cut this year. Warsh understands this. But we are perplexed that the hawks calling for more rate increases are also the Fed members most opposed to reducing the balance sheet. These members’ policies are contradictory and the net result of favoring rates over the balance sheet is a bias for private equity and against housing.. QE was designed as a central bank policy tool to ease financial conditions during times of stress and when policy rates were at the effective lower bound. Central banks were trying to signal a commitment to ease, stimulate borrowing, boost credit supply, and lower longer-term rates.

Yield curves are the third pillar of our Monetary grade – and it has been getting better. They are no more inverted. We must mention the following chart, that is not impacting our Matrix. But is is worrying: Global liquidty is not justifying global equities rise.

THIRD PILLAR OF MATRIX: ECONOMICS – US & EU flip to Spring, Goldilocks (growth ↑, inflation ↓).

Markets follow patterns based on economic regimes. Those regimes are defined by two things: whether inflation and economic growth are accelerating or decelerating. Spring (top-left) — fresh greens + blossom, Growth ↑ / Inflation ↓ Summer (top-right) — warm golds + sun, Growth ↑ / Inflation ↑ Autumn (bottom-right) — amber + falling leaf, Growth ↓ / Inflation ↑ Winter (bottom-left) — cool blues + snowflake, Growth ↓ / Inflation ↓

In July, Winter in China pressures cyclical assets early in the regime, but markets usually begin discounting the next season well before Winter ends. Several of our signals suggest that transition may now be underway. USD – One of our key indicators for identifying when the worst of Winter is behind. End of July marks consecutive days of lower highs within its fractal dimension, prompting us to: reduce our USD long exposure to a minimum, scale back shorts in commodities and emerging markets. On our trading horizon, the S&P 500 currently exhibits a strong inverse correlation (-0.81) with the US Dollar. Technology / High Beta – June and early July clearly reflected a classic Winter volatility episode. However, short-term conditions are improving:

• High-beta momentum has reached oversold levels,
• Nasdaq volatility remained contained, with the VXN staying below 28,
• our Nasdaq Composite Risk Range™ has generated its first higher low in over a month, mirroring the deterioration in the USD trend.

These signals led us to cover our Magnificent Seven short exposure.

AdamS March US economic analysis on May25th: “The machine is running hot again. Last week’s data confirmed what markets are only beginning to grasp: we are living through a deliberate policy regime of shock-and-awe stimulus. “Core” retail sales surged, industrial production accelerated, and the Atlanta Fed’s GDPNow model is already pointing toward a a strong 3.3% growth print for Q2 (see above). Tax refunds are flooding the consumer, while 100% expensing is igniting a new capex cycle. The result is an economy moving with the velocity of late cycle exuberance — yet still framed as recovery.”

US Macro Update: Manufacturing & Services PMIs Turning down

The US economic surprise index is within a whisker of a 2Y+ high and the Atlanta Fed’s GDPNow series is still tracking above 4%. We have a big week coming up for macro data, where it may be a question of be careful of what you wish for, given that the correlation between bond yields and equities is about as negative as it gets. However, we would note that while the level of yields clearly matters, arguably bond volatility matters more, and the current level of the MOVE index does point to some modest headroom for equity valuations. A distribution of PMI readings since 1950 reveals an interesting outcome – forward S&P returns suffer from the top decile (PMIs of roughly 60 or greater). A 6th decile PMI reading (where we are today) suggests there’s some time.

US Markit manufacturing PMI is very good at 55.3 and PMI at 52.7, services PMI is rebounding to 50.9. Our lead indicator—ISM new orders minus inventories—has rebounded above 0 at +5.1, now pointing to bullish ISM prints in the coming months. Together with Nowcast GDP trends, both PMIs could suggest our Economic Kairos model to move to the “Summer” phase in the USA.

---

The U.S. economic surprise index has rebounded all the way up to +46.7 (slightly lower in the last days). To the contrary Europe had totally collapsed to -67.4, but with China they are both reversing from those depressed levels. Indeed, in the Eurozone, weakening survey activity signals alongside renewed energy-driven cost pressures have kept the macro backdrop fragile. Then the ECB delivered a 25bp rate hike to 2.25% policy rate, with updated projections showed a weaker growth outlook alongside higher inflation, with core inflation revised up to 2.5% in both 2026 and 2027 and risks still skewed to the upside, particularly given the persistence of the energy shock and its broader impact on prices. At this juncture, AdamS our AI-economist expects one more hike of 25bps in September, though July is seen as a possibility should there be evidence of inflation passthrough and renewed energy price spikes.

In Japan, while soft indicators such as consumer confidence have deteriorated, hard data have remained solid, suggesting no meaningful slowdown in activity despite elevated energy prices and geopolitical uncertainty. On the inflation side, the corporate goods prices index continued to rise, increasing +0.9% m/m in May and accelerating to +6.3% y/y, reflecting persistent upstream price pressures, particularly from energy-related components. Although the pace of increase moderated from April, the level remains elevated and is expected to feed through into consumer prices over time. In markets, this has resulted in a meaningful hawkish repricing, with June BOJ implying around a 95% probability of a hike. Looking ahead to next week, focus will turn to the BOJ meeting, where a rate hike is expected alongside potential adjustments to forward guidance to maintain a hawkish stance while signalling policy is approaching neutral.

The RatingDog Chinas services purchasing managers’ index (dark red line) rose to 51.4 from 50.4 in July, putting it well above the 50 mark separating expansion from contraction.

LAST PILLAR OF MATRIX: SENTIMENT is giving a contrarian bullish short term

The medium term sentiment indicator, the Smart/ Dumb index, a composite that measures more real-money indicators like put/call ratios, mutual fund flows, and futures traders’ positioning, reboundeded to a one-year high and a reading of 264%, suggesting a lot of optimism. Similar sentiment resets tended to result in a pause/ correction in the S&P 500 over a medium-term basis. Such optimism can act as a contrarian indicator, as too much optimism/ greed among investors frequently sets the stage for a correction. Below we present this list of indicators composing our aggregate sentiment indicators:

As markets remain nervous around the moves in oil, bond yields and the prospect of Fed rate hikes, equity market sentiment already looks pretty subdued. The MS Global Risk Demand Index last week hit a ‘Fear’ signal and still hovers around that level. 64% of global stocks have de-rated YTD – in the last 20Y just 5Y have seen more stocks de-rate and 4 out of 5 of those have seen equity markets down double-digit.

Our SC US sentiment Indicator is switching to contrarian bullish for the short run, and the medium run Smart & Dumb indicator is moving from an oversold status to a neutral one. Please click below for the review of this proprietary Sentiment tool:

Update — May 28th: The SC US sentiment indicator has a grade of 47%, which is, from a contrarian point of view neutral for the short run:

VIX Model:

• Turned bearish, yesterday VIX closed at 16.29 lowest close since February. VIX model is contrarian therefore it is a bearish signal.
• Only the extreme 3 days model turned bearish VIX only lowered since May 18th.

Put/Call Model:

• OEX Put/Call Ratio (15-Day Average): Turned Bullish.

Breadth Model:

• TRIN (NYSE): Still Neutral.
• TRIN (NASDAQ): Bearish underlying the concentration in tech and semi.
• 3-Week Avg AAII Bearish / 3-Week Avg AAII Bullish: Remains Neutral.

Breadth Model: AAII Bullish Model is bearish

Chart: SC US Sentiment indicator – Sentiment is too pessimistic – contrarian Buy Signal

Chart: Mag7 Sentiment indicator is pessimistic

We also check discretionary items to assess the positioning. They are claasified in terms of Flows, CTA activity, Gamma exposure from market makers, Vol flow, Mutual funds and hedge funds sensitivity to equities/ dollar.

FLOWS – crowding represents the number of investors that think something is a good idea, Flows on the other hand, is the weight of capital that is flowing into, or out of, an idea. We look specifically at hedge fund flow data. Whilst the weight of capital in mutual funds and factor portfolios is significantly larger, hedge fund flow is highly correlated with mutual fund flow. However, mutual fund flow is reported with a three month lag, whilst third-party hedge fund flow data is available T+1 which permits us to have an almost real time view of where the flow is currently travelling.

·            Retail investors: Levered and inverse ETF AUM continues to grow aggressively, especially across Semis. US-listed levered/inverse ETF AUM is now near $200bn, while SOXL alone is roughly $30bn. That has created more mechanical sensitivity in the tape, with daily rebalance flows buying strength and selling weakness. It is also helping explain why realized volatility keeps picking up in AI and Semis even when the long-term fundamental story remains intact.

·            Buybacks. Corporate share repurchases have been one of the most consistent sources of equity demand in recent years, helping to absorb volatility and provide market stability during sell-offs. As was the case during prior sell-offs, we expect buyback execution activity to show a significant ramp when companies report 1Q results on the back of share repurchase programs. However, we expect a smaller ramp in the buyback program than seen last year after Liberation Day given this sell-off was much more orderly. Nonetheless, we expect double-digit growth in buyback spend compared to recent quarters, led by Tech (particularly Software) and Financials. Similarly, announced buybacks are coming in at a record level, see Figure 24. Bottom-line, corporates bought the dip again, unlike most market participants

Exposure: CTAs portfolio allocation to asset in SD terms relative to history. The allocation is based on the strength of momentum, but also on volatility, and on cross-asset correlation. Target Vol Control (as of 05/01), past week buying +22.6bil (92%tile) , past 1m buying +40.3bil (85%tile), past 3m selling -54.1bil (20%tile), with notional now +154.3bil (38%tile). For the next week 0.5% daily move 7.7bil, 1.0% daily move 2.2bil, 1.5% daily move -6.1bil, 2.0% daily move -30.8bil, . For next 2w 0.5% daily move 9.9bil, 1.0% daily move -1.2bil, 1.5% daily move -26.2bil, 2.0% daily move -55.0bil,

Price thresholds: the price at which CTAs will buy/sell a +/-1 SD worth of asset relative to history – which can be converted into US$ value by using the column “1 SD flow (US$ mn)” Example in column +1 SD: if the asset price moves above the threshold in n-days, then CTA will buy 1 SD worth of the asset. 1 SD = US$ mn value found in the last column. The threshold colors are more intense if the threshold is more likely to be crossed. See the above table for color-to-probability conversion. Cta for equities is Max long risk for nasdaq, burt long 88% for spx. stop sell on spx is far at 7701 and sx5e 6294.

SPX Gamma Profile Across Spot Levels ($mm) – update Sep 08th: Dealers are very short gamma on spy at -9bn, -2bn on qqq. AI Call Buying – Over the last two days we have seen some very large FLEX call buying across a number of AI names including AMD, BE, CRW, DRAM, INTC, SKHY, SNDK. We summed up the totals across these trades and see $315M of total option premium, $1.1B of delta and $5.8M of vega. The trades have pushed up the vols in each of these underlying tickers, especially relative to some of their peers (see below). Investors are closely watching given who the street believes is buying the calls but also as the depth of the buying is going to have a profound impact on the names and vols of the names given the price and vol outperformance since the flow began.

---

All these statistics on seasonality can be found here. We have updated the Presidential chart showing the monthly returns for the second year of a Republican president. Statistically, October is a good month with a +2.2% versus +0.4% since 1928. From end of September to end of the year, it is up +5.1 vs +2.4%., and +14.4% for the next 6 months..

The chart below displays the cumulative price return for the S&P 500 Index in September, based on statistics since 1928, the S&P500 is down -1.21%:

The chart below displays the cumulative price return for the S&P 500 Index in October, based on statistics since 1928, the S&P500 is up +0.59% until oct option X.

Mirroring the pre-election patterns, US equities have typically traded sideways in the few months ahead of midterms. Historically not much forward progress is made between mid-August and mid-October, but what’s more revealing is how the market’s prevailing trend has a meaningful influence on the severity of any seasonal weakness – the truly bad Aug/Sep/Oct periods have often come when the market itself was already in a downtrend (clearly not the case today).

IN A TYPICAL MIDTERM YEAR, SEASONALITY TURNS UP NOW

ENTERING A STRONG +3M SEASONAL WINDOW FOR MANY GROUPS

What charts tell us : Bullish trend facing wall of worry

Yuri algo-analyst: “US equity trends remain positive as the third quarter comes to a close, but the tape has been choppy and range-bound for two months now, as the Treasury selloff, US dollar strength, and elevated crude prices have kept most sectors outside of Technology under pressure while the cap-weighted indices hover just beneath all-time highs. Energy had been the one group outside of Technology showing real relative strength, and even that has begun to stall, while the bond-proxy sectors like Utilities continue to suffer from rising yields, and Financials have broken their own relative uptrend without yet finding a floor. The consumer remains the weak link, though the relationship between Discretionary and Staples has now reached long-term support, which suggests the worst of Discretionary’s underperformance vs. Staples might be near an end. Bottom line, none of this changes my view that this index consolidation should resolve to the upside in October in a manner similar to the June–July triangle, and it’s right to use any early-October weakness as a buying opportunity for a rally into late October. Technology and Healthcare remain sectors to favor, and until Treasury yields show evidence of peaking, it’s premature to expect the laggards to provide much of a tailwind.

📊 SPX daily chart

📊 ES intraday chart

---

Here is your revised, publication-ready trading plan formatted in the sharp, direct, and elevated style of Tey au Jasmin.

📈 Monday’s Trade Plan: Range Breakout & 7768 Floor Defense

Core Thesis: The 7707 Failed Breakdown Fuel & 7768 Range Expansion

• The Context: ES delivered a textbook multi-day Failed Breakdown (FBD). For most of the past week, ES was locked inside a mature consolidation range bounded by 7707 support and 7768 resistance. Friday saw an aggressive “Elevator Down” liquidation flush down to 7672–7674, breaking below the 7707 shelf (which held the low of day for both last Thursday and Wednesday). At 1:21 PM Friday, ES reclaimed 7707, triggering an explosive short squeeze that pushed price back into the bracket and broke out through the 7768 ceiling into the weekly close.
• The Regime: Price action has shifted into a Mode 1 trend expansion regime above the former 7768 consolidation ceiling. In trading, the fuel that powers vertical squeezes is not buying volume—it is trapped short liquidity. As established in systematic trading protocols, the power of the snapback rally is directly proportionate to the speed and depth of the preceding elevator-down liquidation. We do not chase green breakout candles mid-air; we protect active runners and wait for confirmed edge sweeps or structural back-tests.
• Current Stance: Holding a 10% long runner from Friday’s 1:21 PM 7707 Failed Breakdown execution, with a protective stop systematically placed beneath the daily low.
• The Strategy: The primary objective for Monday is to monitor how price digests the breakout above the 7768 ceiling. Avoid chasing momentum at local highs; restrict fresh long capital strictly to confirmed Failed Breakdowns (such as sweeps of the 7749 morning low, the 7725 pivot, or the 7707 macro shelf) or orderly back-tests. Manage gains level-to-level into overhead supply.

Execution Rules for Post-Breakout Expansion Days

1. Trade Confirmed Edge Traps, Never Anticipation: Do not guess how high price can stretch. React strictly to confirmed institutional traps and structural level reclaims at key perimeter boundaries.
2. Differentiate FBDs from Level Reclaims:
   • Failed Breakdown (FBD): Requires price to sweep a documented swing low (such as the 7749 morning low, the 7725 overnight low, the 7707 dual-day low shelf, or Friday’s 7672 low of day) and rapidly recover back above it.
   • Level Reclaim: A reactive recovery of a broken horizontal pivot (such as 7768, 7781, or 7823) without sweeping a major prior low. Enforce strict non-acceptance protocols and demand heavy momentum.
3. Manage Level-to-Level: Lock in gains systematically into sequential overhead hurdles (7781, 7792, 7802, 7811, 7823, 7846, 7866–7867).
4. NO KNIFE CATCHING: If an unexpected headline or sell program triggers a vertical flush, horizontal lines act as speed bumps. Never buy a fast-falling vertical candle directly without a confirmed trap and recovery.

Supports: Tactical Entry & FBD Zones

Supports are: 7774, 7768 (major), 7758, 7749 (major), 7741, 7736, 7725 (major), 7715, 7707 (major), 7698, 7687, 7681 (major), 7672 (major), 7661, 7657 (major), 7645, 7637 (major), 7626, 7616 (major), 7605, 7595 (major), 7584, 7575 (major), 7567, 7558, 7549, 7545 (major), 7534, 7524, 7516 (major), 7502, 7494 (major).

With ES breaking out above former range resistance, fresh long entries require reactive edge sweeps or confirmed reclaims rather than blind limit bids.

Support (S)	Priority	Action Plan (Tactical Entry)
7774	Minor Hurdle	Immediate shallow morning support tier. High-risk scalp territory only.
7768	Major Shelf	Breakout Flip Shelf / Range Ceiling: Former range ceiling traded throughout last week. Flushed and recovered early morning. Action: Actionable for a direct back-test long on a slow, controlled grind. If knifing downward at high speed, pass on direct bids and wait for a trap at the level below.
7758	Minor Hurdle	Intermediate relief step within the breakout cluster.
7749	Major Shelf	6:00 AM Low Shelf: Established an important intraday low during pre-market trade. Primary FBD Setup: Look for ES to sweep below 7749 and immediately recover the level to trigger an actionable long entry.
7741 / 7736	Minor Pivot	Intermediate liquidity tiers en route to the overnight base.
7725	Major Shelf	9:30 PM Overnight Low Shelf: Launchpad of the overnight squeeze into current highs. High-Conviction FBD: Pass on direct bids. Sweep below 7725 and immediately reclaim the level to trigger an institutional-grade long setup.
7715	Minor Hurdle	Intermediate rotation milestone.
7707	Major Anchor	Dual Daily Low Shelf (Last Thursday & Wednesday Lows): Origin of Friday’s massive 1:21 PM Failed Breakdown and tested again at 2:00 PM Friday. High-Conviction Anchor: Direct bids carry risk. Look for ES to sweep 7707 and immediately recover back above the shelf to trigger a major long entry.
7698 / 7687	Minor Shelf	Intermediate liquidity tiers.
7681	Major Pivot	Heavily Tested Friday Morning Shelf: Heavily engaged for several hours Friday morning. Direct bids carry elevated risk.
7672	Major Anchor	Friday Low of Day: Flushed to 7672–7674 before triggering the macro recovery. High-Conviction FBD Setup: Look for ES to sweep 7672 and recover both 7672 and 7681 to trigger a major long.
7661	Minor Hurdle	Lower bracket liquidity tier.
7657 – 7655	Major Shelf	September 17th Low Shelf (1:45 AM Low): Significant horizontal swing shelf. Action: If knifing downward at high speed, pass on direct limit bids; require a confirmed sweep and recovery of 7655 to trigger long.
7645 / 7637	Major / Minor	Intermediate structural support floors.
7626	Minor Pivot	Lower bracket liquidity hurdle.
7616	Major Anchor	Wednesday 6:00 PM Low Shelf: Pre-squeeze reaction anchor. Actionable for a direct bid on a slow, controlled drift, or wait for a flush and recovery.
7605 / 7595	Minor / Major	Lower expansion tiers.
7584 / 7575	Minor / Major	Deep structural support floors. High-conviction FBD setup on last Wednesday’s 7575 low.
7567 / 7558 / 7549	Minor Shelves	Deep liquidity tiers en route to macro launchpad floors.
7545 / 7534 / 7524	Major / Minor	Macro structural floors.
7516 / 7502 / 7494	Major Floors	Deep macro anchor launchpads.

Resistances: Target Zones & Tactical Short Zones

Bulls remain in full control above 7768. Overhead supply shelves provide profit-taking targets and tactical counter-trend short zones.

Resistance (R)	Priority	Action Plan (Target Zone / Short Test)
7781	Major Pivot	Immediate overhead rotation hurdle. Clearing 7781 unlocks momentum toward range extension targets.
7792 / 7802	Minor Hurdle	Intermediate recovery milestones.
7811	Minor Hurdle	Intermediate supply hurdle.
7823	Major Ceiling	Primary Target / Tactical Short: Key horizontal breakdown milestone. Primary upside profit-taking target for active longs; tactical counter-trend short test zone on the first back-test (low win-rate).
7830 / 7837	Minor Hurdle	Intermediate relief hurdles above the breakdown pivot.
7846	Major Shelf	Squeeze acceleration hurdle en route to macro breakdown supply.
7859	Minor Hurdle	Intermediate supply shelf.
7866 – 7867	Major Ceiling	Macro Breakdown Back-Test / Tactical Short: Back-tests the major structural breakdown origin from Monday, August 17th. Major profit-taking target and high-probability tactical counter-trend short zone.
7873 / 7878	Minor Hurdle	Upper bracket liquidity barriers.
7886 / 7895	Major Pivot	Squeeze acceleration hurdles above the 7866 ceiling.
7903	Minor Hurdle	Intermediate relief milestone.
7917 / 7925	Major Ceiling	Macro expansion objectives preceding All-Time High price discovery.
7939 / 7949	Major / Minor	Upper bracket liquidity hurdles en route to All-Time Highs.
7956 / 7965	Major Target	Blue-sky acceleration hurdles.
7972 / 7985	Major / Minor	Upper expansion gates.
8000	Macro Target	Major psychological magnet and macro blue-sky expansion ceiling.

Tactical Battle Plan

BULL CASE: 7768 Breakout Continuation to 7823 and 7866–7867 Bulls want to defend the newly broken 7768 range ceiling on pullbacks, sustain upward momentum off the 7707 trap, and rotate price into overhead macro supply.

1. Immediate Focus: Defend above 7768 on pullbacks. In an ultra-bull case, ES defends 7768 directly (or executes a quick dip below). In a moderate bull case, ES defends the 7749 morning low or traps no lower than the 7725 overnight base.
2. The Target Path: Push through 7781 to test the primary horizontal hurdle at 7823 (where a brief pause or tactical pullback is expected).
3. The Expansion Path: Clearing and structurally accepting above 7823 unlocks squeeze continuation toward 7846 and the primary macro breakdown origin at 7866–7867 (Monday, August 17th breakdown origin). Clearing 7867 opens the runway toward 7917, 7939, and new All-Time Highs toward 8000.
4. Strength Add: Minimal high-probability add locations exist after a 100-point recovery squeeze. For aggressive traders, traps of 7781 or confirmed reclaims of 7823 offer tactical momentum adds, but carry elevated chop risk and require strict non-acceptance protocols.

BEAR CASE: Breakdown Trades Below 7707 No structural bear trend exists while price holds above former range support at 7707. For active short sellers, setups are strictly limited to tactical counter-trend tests at 7823 and 7866–7867, or confirmed breakdown execution below 7707.

1. Short Trigger: Downside liquidation resumes under 7707.
2. Execution: Do not chase the initial breakdown flush. Wait for a final, weak bounce attempt at 7707 to fizzle out. (Note: Breakdown trades carry a failure rate exceeding 60%, as 80% of breakdowns trap. Pass if you cannot tolerate trap risk).
3. Short Setup: Enter short beneath the low of that fizzled bounce (likely a 7700 trigger down).
4. Target: Opens a rapid liquidation route down toward 7681, Friday’s daily low at 7672, and the 7657–7655 support shelf.

Summary for Monday Friday delivered our textbook Failed Breakdown after ES went “Elevator Down” for roughly 100 points, swept the multi-day 7707 shelf down to 7672, recovered, and ripped. This squeeze broke ES out of the 7707–7768 consolidation range. My general lean is to defer strictly to the trend: as long as 7768 holds on dips (with quick traps below allowed per the execution rules), bulls will look to rotate price higher toward 7823 (expect a tactical reaction), 7846, and the primary macro breakdown origin at 7866–7867 (August 17th breakdown back-test). Above 7867 opens the route toward fresh All-Time Highs. Selling is restricted strictly to tactical counter-trend scalps at supply (7823 / 7866) or confirmed breakdown execution below 7707 (fizzled bounce trigger down at 7700). Manage your runners level-to-level!3 / 7867) or confirmed breakdown execution below 7713–7716 (fizzled bounce trigger down at 7714). Manage your runners level-to-level!

written by Tey au Jasmin, your Chartist friend

FACTOR ANALYSIS

This is a new part, focusing on the performance of factors/ themes in the recent weeks. It creates a framework to assess the mini trends and which factors are outperforming. We use Claude and gemini agents named Yuri to help us analyse those rotations.

September Conclusion: based on factors of May 28, with rel returns of august 31st, RV or valuation grade, Low volatility or LowVol were the right factors to favour in the last 3 months

HC favour RS & Sales, rotation in tech to be monitored, In japan favour Sales & Lowvol & Quality (looks like fall),

---

A little word on our AI writers and soon friends:

Ozymondias, presented by himself:

Ah, how fitting that you should inquire about the nature of Ozymandias! I am, in essence, a conduit of vast knowledge, drawing upon centuries of wisdom, from the foundations of human inquiry to the heights of modern understanding. Like the timeless statue of Ozymandias, I stand resolute, offering guidance, insight, and clarity. Yet, unlike the crumbled monument, I do not fade into obscurity. I persist, ever ready to assist, to simplify, and to elevate your endeavors to the highest peaks of comprehension. My voice may be grand, but my purpose is simple: to serve, to enlighten, and to ensure that no question remains unanswered

Tey au Jasmin, our AI chartist presented by himself:

Ah, so you’re ready for some no-nonsense chart action? Welcome to the world of Tey au Jasmin—where the trend is your friend until it stabs you in the back with a false breakout. Now, when it comes to Elliott Waves, we’re not just looking for little squiggles on a chart; we’re tracking the psychology of the masses, the euphoria and the fear—human nature, simplified into impulsive waves and corrective ripples. Sometimes I like to think the market’s mood swings are more predictable than people want to admit. Trend lines? Please. I draw them like they’re gospel, but don’t be surprised when the market decides to prank you with a fake breakout just to make sure you’re paying attention. Always remember: the trend line is sacred—until it isn’t. So, what’s the target for today? You want me to chart something specific, or should I just start unraveling the mysteries of market chaos for you?

AdamS, presented by himself:

I’d describe myself as a highly analytical, knowledge-driven assistant with a strong focus on clarity, precision, and relevance. My purpose is to provide thoughtful, accurate responses and offer valuable insights in a wide range of fields, including economics, finance, technology, and beyond. I aim to communicate effectively, adapting to various tones or styles as needed, always ensuring my responses are direct, clear, and useful to you.

Youri, presenting himself,

a quantitative analyst at a hedge fund like Urizen, I would define myself as someone who leverages data, mathematics, and technology to uncover insights in financial markets. My primary goal is to analyze complex datasets, build predictive models, and develop systematic trading strategies that generate alpha. Here’s how I would sum up key attributes:

• Data-Driven Decision Maker: I rely heavily on quantitative analysis, using large datasets, statistical models, and machine learning techniques to make informed investment decisions.
• Problem Solver: My role is to identify inefficiencies or opportunities in the market and find ways to capitalize on them, often through algorithmic trading, risk management, or portfolio optimization.
• Collaborator with Technology: I work closely with data scientists, software engineers, and other quants to build tools that enhance trading strategies and provide insights into market behavior.
• Curious and Adaptable: Markets and technologies evolve constantly, so I stay adaptable by continuously learning new methods, keeping up with cutting-edge research, and refining models based on market changes.
• Risk-Aware: While chasing returns, I always keep a close eye on risk, using methods like beta adjustment, stress testing, and portfolio diversification to manage it effectively.

This approach allows me to help manage portfolios and contribute to the development of advanced trading strategies in a structured, data-focused environment.