Pinned Tweet
It’s not in the news
you saw the headline. so did everyone else.
a handshake moves oil.
oil moves rates.
rates move the dollar.
the dollar moves everything.
then the crowd shows up with an opinion.
it’s not in the opinion. it’s in the positioning.
СhaseTrust - geopolitics, macro and btc options in one terminal.
source: chasetrust.org
dollar is beating gold?
• Oil 145
• 10y yield 124
• copper 121
• s&p 112
• dollar 104
• gold 95
• btc 70
there’s a war on and oil is 45 points above where it started.
$GOLD just printed a 13-month low - it’s down about 7% in a month, and copper is down about 7.5% over the same month.
war hedge doesn’t fall in step with an industrial metal.
something trading the price of money does and the price of money is going up.
the 10y went above 5.3% in october - the dollar is at a 19-month high.
bitcoin:native got it worst - around $65k, down roughly 20% in 30 days and almost half off its high.
stocks held out the longest.
they’re still 12 points above the line, but on november 5 the s&p had its worst day since april 2025.
the war shows up in oil. everything else is trading rates.
which makes december 9 the date.
- hike is fully priced?
- seems yes
• fedwatch has 25bp at 95% and 50bp at 5%
• 50 is the fear, since waller said it may be needed
our view
clean 25 is a positive driver for markets - it removes the 50 tail and gives everyone a known number.
5
105
End of Huang Era.
Next up: $NVDA needs an accountant
so the board looked around the room and handed the company to the CFO.
Colette Kress has been signing off on $NVDA numbers since 2013. cool.
let's look at what those 13 years look like on paper.
THE PAPER TRAIL
• 2022
SEC fines @nvidia $5.5M
2 quarters in a row they "forgot" to mention that gaming revenue was really crypto miners. no admission, no denial, nobody charged personally
• 2024
Supreme Court tosses Nvidia's appeal, so the investor lawsuit over those same crypto sales is still alive. company and Huang are the defendants. a lawsuit, not a verdict
• 2026 filings
$279B owed to suppliers, up to $108.5B in guarantees to cloud partners, around $50B parked in AI labs that turn around and buy Nvidia chips
• customers now take 60 days to pay instead of 45
"WE SEE IT DIFFERENTLY"
that's Kress in August, word for word: "we know some will call this circular financing. We see it differently."
Burry sees it the old way.
his take this week: Huang leaving is not the problem, the problem is whose money is buying the chips.
to be fair to her: WSJ says Kress was the one pushing back on those guarantees and cut the OpenAI commitment in half. Huang called her a "reporting CFO" for it.
BEFORE NVIDIA
#Microsoft, then #Cisco
2010 to 2013 she ran finance for the unit that handled Cisco's products and supply chain. same years Cisco restructured and cut 6500 people.
SO
nobody has proven the sales are inflated. what we do have is a company lending money to its own customers, and the person who signed all of it now sits in the big chair.
earnings on the 17th and forget revenue, look at receivables, guarantees and inventory.
the jacket era is over.
the cable turned into a card, the question didn't change.
*just an observation, not financial advice.
5
163
BREAKING: $NVDA looking for a new accountant who will be involved in inflating sales figures.
Sell-side +43 when press -56 on US equities = the widest gap since July.
It’s mostly a lag.
Every 8000 target was written before the AI selloff, and hitting it now takes +6% in 7 weeks.
Watch for the first bank to cut it.
5
69
AI sector correction is already priced in, and the consensus has already acknowledged it.
The analyst view on technology shifted over the week from bullish to disagreement, and the tape is consistent with this: AI sector lost 8.23% over the week.
Across US equities broadly, banks retain a bullish-leaning overweight view, while the broad market declined over the week.
In favor of the banks here is Friday's rebound combined with short speculative positioning in index futures, against them is the inverted volatility curve, which does not typically fade within a single session.
Our reading diverges from prices in the assessment of credit:
High-yield spread widened to 348 bp with a weekly widening of 22 bp, rare by historical standards, and the investment-grade segment also moved off its tight levels.
The base is low, the domain is calm, there are no signs of a default cycle, but credit responds to equities with a lag, and its restraint this week is not yet an argument for fading.
4
76
CHASETRUST
ChaseTrust is an ecosystem of research products covering markets, macroeconomics and geopolitics.
Its common principle is verifiability:
• an assessment of the market is published with its confidence level
• the date of its last revision and the age of its data
• hypotheses that fail testing are published alongside those that hold.
SPECTRUM estimates from $BTC options data where dealer hedging tends to dampen price moves and where it tends to amplify them, and combines this with an assessment of the macro regime across seven domains. It is intended for traders who already have a method of their own and do not need entry signals.
• chasetrust.org
SPECTRUM DEPOT is a private research library in Telegram: press, journals, bank and institutional research, market data.
• @spectrum_depot_bot
Further products are in development and will be announced in this channel.
Pricing: dash.chasetrust.org/en/tarif…
Research: chasetrust.substack.com
Telegram: t.me/chasetrustglobal
Russian channel: t.me/chasetrustchannel
X: teapawt.net/chasetrustorg
YouTube: youtube.com/@chasetrust
Threads: threads.com/@chasetrustorg
Instagram: instagram.com/chasetrustorg
Company profile: crunchbase.com/organization/…
Analytical services, not investment advice.
🤖 Made with AI
4
109
$VIXY 👀
2/7 stress gauges are almost at the line, and this week shows why.
volatility is at 82, 2 points from strain.
most of that is oil. brent was near $118 last friday and around $95 by wednesday, down almost 9% in one day and the lowest since early september.
after the midterms the white house said there are no immediate plans to strike iran, and traders started pricing an end to the war.
gulf exports are still only about 3/4 of pre-war levels though, so it can swing back just as fast. ovx sits at 92 of 100.
funding is at 77, four points off. this is the quieter problem.
SOFR has been sitting at 3.95%, right under the 4.00% top of the fed’s range, and sofr-effr is in its 96th percentile, so repo is running hot vs fed funds.
no relief from the fed either. it held at 3.75-4.00% on oct 28, two members wanted a hike, and markets price about 85% odds of one on dec 9.
liquidity is the one gauge saying relax, at 55. its most strained piece is the chicago fed nfci at 64 of 100.
nothing’s broken yet. but oil just moved 20% in a week and repo is pinned near the ceiling.
close enough to watch.
6
122
Our Scenario Board Has a Clear Leader Right Now: funding stress and reserve scarcity, at 40%
the rest trail well behind.
duration repricing through real rates at 30%, energy shock and a stagflation fork at 14%, calm drift with sticky inflation at 8%, carry unwind out of japan at 8%.
here’s how the top one plays out.
bank reserves slip past the point where banks start hoarding them, the treasury’s account is full, and there’s no buffer left. the secured rate breaks away from the policy rate, repo leverage gets squeezed, and the selling spills into treasuries. it doesn’t stop until the fed steps in with its standing facilities or stops draining liquidity. until then, everything is risk-off.
the odd part: market pressure on this scenario is only 28/100.
it’s the path we think is most likely, and the market isn’t pricing it yet.
what we’re watching:
- the secured rate staying above the policy rate outside calendar dates
- reserves falling below the comfortable zone
- more use of the discount window and emergency facilities
- rates vol flipping into crisis mode
- swap lines getting activated
if it happens, the playbook is to cut leverage early, sit in cash and short-dated treasuries, buy protection before the fed steps in, and rotate back into risk after.
40% is an estimate, not a measurement. it’s still the one we’re watching closest.
5
92

















