In partnership with
Last week the economy shed 23,000 jobs, and the S&P 500 closed at a record 7,757.64. Both of those sentences are true, and the second one is not insane. Wall Street was not applauding the layoffs. It was applauding what the layoffs did to the odds of the Fed making things worse on purpose.
Which brings us to the number nobody put in a headline. Fed funds futures still give the September meeting roughly a 40% chance of a rate hike. Not a cut. A hike. The best week since April was a celebration that a coin flip had been downgraded to a 6-in-10.
Wednesday at 7:30 AM CT, July CPI decides whether that math survives the weekend it was born on.
(Where we left off: the full autopsy on last week's party is here, and we flagged how wildly Wall Street was guessing the morning before the number landed.)
Personal best. Conditions apply.
What You Need to Know in Under 60 Seconds
July CPI is due Wednesday at 7:30 AM CT. Consensus is 3.4% year over year, down a tick from June's 3.5%.
June's headline fell 0.4% in a single month. That was gasoline collapsing 9.7%, not inflation dying of natural causes.
Core CPI is seen at 2.5%, below the headline rate. When the core is cooler than the top line, the inflation problem is an energy problem, and July is the month oil went vertical.
September hike odds sit near 40% after Friday's jobs miss. A week earlier, the market had a hike as the likelier outcome.
PPI and jobless claims Thursday, 7:30 AM CT, wedged between two Fed speakers 25 minutes apart.
Retail sales Friday, 7:30 AM CT. This is where you find out whether the consumer noticed any of it.
Treasury sells 3-year, 10-year, and 30-year paper Tuesday through Thursday. The 10-year price four and a half hours after CPI is a rough day to be the person taking bids.
Gold just had its best week since January, up 7.25%. The VIX closed at 14.90. Those two markets cannot both be reading the same newspaper.
The Week at a Glance
Everything below happens on a clock. Everything above it is opinion.
Day | Data (CT) | Earnings | Wildcard |
|---|---|---|---|
Mon Aug 10 | 9:00 a.m. CB Employment Trends; 10:30 a.m. 3- and 6-month bill auctions | MNDY (pre); ACHR, RKLB (post) | Nothing on this list can hurt you. Savour it. |
Tue Aug 11 | 5:00a NFIB small business optimism (97.8 est); 9:00a existing home sales; 10:00a NY Fed Q2 household debt; 12:00p 3-year note auction | SE, ONON, TME (pre); CRWV, SMCI, CAVA (post) | CoreWeave is the AI capex bill wearing a ticker. |
Wed Aug 12 | 5:00a OPEC monthly report; 7:30a JULY CPI; 9:30a EIA crude stocks; 12:00p 10-year note auction; 1:00p July budget balance | CSCO (post) | The rest of the week is a footnote to one line. |
Thu Aug 13 | 7:15a Fed's Hammack speaks; 7:30a July PPI + jobless claims; 7:40a Fed's Barkin speaks; 12:00p 30-year bond auction | JD, GDS (pre); AMAT, NU (post) | A voting hawk takes a microphone 15 minutes before a print she has not seen. |
Fri Aug 14 | 7:30a July retail sales; 9:00a Michigan sentiment (prelim); 12:00p Baker Hughes rig count | Quiet | Michigan one-year inflation expectations still start with a 4. |
June's Inflation Report Was Written by a Gas Pump
June's CPI fell 0.4% in a month, and almost none of that was the Fed's doing. Energy dropped 5.7%, with gasoline alone down 9.7%, and that single line dragged the whole index into the green. Strip it out and the picture was considerably less flattering, which is why the celebration in June always looked a bit like a man bragging about his weight after taking off his boots.
July is when the boots go back on. The math here is almost insultingly simple: for headline CPI to rise this month, gasoline does not need to spike. It only needs to stop falling at June's ridiculous pace. Economists have July at +0.1% month over month and 3.4% year over year, which is the polite way of saying they expect the discount to expire.
The tell is in the core, the version that strips out food and fuel because both bounce around like a toddler. Core is seen at 2.5% year over year, a full point below headline. In a normal cycle core runs hotter and stickier. When the headline is the one running away, your inflation problem is sitting in a fuel tank, and July is precisely the month crude went vertical on the Gulf.
What could go wrong: a hot print, and Friday's rate-cut fantasy gets repossessed before lunch. A cold one, and everybody decides the Fed is finished, which is exactly the assumption that got dissented against three times in July.
June weighed in with its shoes off, its coat on the floor, and its keys on the counter, and Wall Street wrote the number down as a personal best. Wednesday, it steps back onto the scale, fully dressed, with two jerry cans in its hands. Nothing about the body changed. Only the honesty of the measurement.
The 40% Nobody Puts in the Headline
The Fed's own committee is not having the conversation the market thinks it's having. On July 29, under Chair Kevin Warsh, the FOMC held the target range at 3.50% to 3.75% on a 9-3 vote. All three dissenters wanted to go the other way. They voted to raise. There has been no serious cut caucus at that table for months, and the market's "dovish repricing" last week was not a move toward easing. It was a hike in odds, falling from likelier-than-not to roughly 40%.
Then Thursday morning happens, and it is a beautifully stupid piece of scheduling. Cleveland's Hammack, one of those three hike dissenters and a voting member, speaks at 7:15 AM CT. July PPI prints at 7:30. Richmond's Tom Barkin speaks at 7:40. So the hawk with a vote talks 15 minutes before the data, and the man without one talks 10 minutes after it. One of them is guessing, and the other is reacting, and the tape will treat both like scripture.
Worth keeping straight, because it changes the weight of every quote you'll see: Barkin is an alternate this year and does not vote. Hammack does.
What could go wrong: PPI runs warm and Hammack's pre-print caution reads as prophecy, which is how a 40% becomes a 55% by Friday's open.
The market spent last week celebrating that it probably won't get hit. That is a genuinely lower bar than "getting paid," and somewhere between the two is where portfolios quietly go to be disappointed. Next FOMC is September 16. You have five weeks of data to sit through before anyone at that table has to be honest.
Four small dots and one enormous red one. That is the week.
Gold's Best Week Since January, and a Fear Gauge Taking a Nap
Two markets watched the same week and filed opposite reports. Gold, via GLD, ran +7.25% Monday to Friday, its strongest weekly gain since January, with spot finishing near a seven-week high. The VIX, Wall Street's smoke detector, closed the week at 14.90, down 6.8% and comfortably in the range that signals nobody is buying protection.
Neither of those can describe the same reality. A 7% week in gold is people quietly buying the thing that pays no interest, does nothing useful, and only wins when currencies and central banks disappoint. A 14.90 handle on the VIX is people writing insurance because they cannot imagine the fire. The gold buyers and the volatility sellers are, structurally, betting against each other.
Meanwhile, the rest of the tape was uncomplicated: SPY +3.51%, QQQ +5.09%, IWM +3.56%, DIA +2.92%, with the 10-year yield easing to about 4.65% and oil, via USO, down 8.66% as the Gulf headlines cooled. Everything went up together, which is lovely, and which tells you approximately nothing about what happens when one number lands sideways.
What could go wrong: if Wednesday's print is hot, gold, and the VIX both get to be right at once, and that is not a fun way to find out who was hedged.
When the insurance market and the panic-metal market disagree by this much, one of them is being paid to be wrong. You do not have to guess which. You just have to notice that a 14.90 VIX makes protection historically cheap the week before the year's loudest data point, and that "cheap" and "unnecessary" are not synonyms.
Three Report Cards for the AI Build-Out
The mid-week earnings are all the same story told from three seats. Tuesday after the close, CoreWeave (CRWV) reports with analysts penciling in a loss of $1.67 a share. Wednesday after the close, Cisco (CSCO) at $0.99. Thursday after the close, Applied Materials (AMAT) at $3.36. Renting the computer, wiring the computer, and building the machines that make the chips that do the computing.
That order matters. CoreWeave tells you what capacity costs to carry. Cisco tells you whether the networking spend behind the data centers is real or a slide in somebody's deck. Applied Materials tells you what the semiconductor industry is committing to eighteen months from now, because equipment orders are a promise you cannot quietly cancel.
Rounding out the week: Sea Limited (SE), On Holding (ONON) and Tencent Music (TME) Tuesday morning, Super Micro (SMCI) and Cava (CAVA) Tuesday night, then JD.com (JD) and Nu Holdings (NU) Thursday. It is a real slate, but it is not the week's boss.
What could go wrong: AMAT lands Thursday night into a tape already jumpy from two inflation prints, and a soft equipment outlook does more damage in that mood than it would on a quiet Tuesday.
The AI trade has spent a year being graded on ambition. This week it gets graded on invoices. Three companies, three points in the same supply chain, and if all three say demand is fine, the story holds. If the guy selling the machines hedges, everybody upstream of him has a problem they cannot spin.
Wednesday, 7:30 AM CT.
Geopolitical Corner
Ranked by how much of your portfolio it can actually reach.
Hormuz: an agreement Iran says exists and says is not enough. Iranian officials said Friday they had reached an arrangement with Oman for a temporary shipping route through the Strait, with reported terms that ban American and Israeli vessels and fine violators up to 20% of the cargo value. By Saturday, Tehran's stated position was that the deal alone would not reopen the waterway and that it wanted US concessions first. Trade intelligence firm Kpler reported Hormuz traffic down 33% on Friday compared with the day before. The UAE separately said one of its ships was hit by an airstrike. Market handle: oil, and through oil, Wednesday's CPI. USO fell 8.66% last week on the diplomatic noise. If the talks visibly break, that discount comes back with interest, and it shows up in an inflation print.
OPEC's monthly report, Wednesday, 5:00 AM CT. The cartel agreed in principle on August 2 to lift September quotas by roughly 188,000 barrels a day, then pause for the fourth quarter. Market handle: crude, energy equities. The demand forecast in this report is what to read, not the headline. It also lands two and a half hours before CPI, so if it moves crude, it moves the mood into the print.
The calendar is empty, and that is its own risk. No G7, no foreign central bank decisions, no scheduled elections that trade. Congress is away on its August recess, and Jackson Hole is still weeks out. Market handle: liquidity. August books are thin, desks are half-staffed, and thin books do not absorb surprises; they amplify them. A quiet calendar is not the same thing as a safe one.
The One Thing
July CPI. Wednesday, August 12, 7:30 AM CT. Bureau of Labor Statistics.
Consensus is 3.4% year over year and +0.1% on the month, with core at 2.5%. Everything else this week is commentary on this line.
If it comes in at or below consensus, the September hike stays a minority view, the 40% drifts lower, and last week's record close keeps its receipt. Gold probably keeps working anyway, because gold has stopped asking permission.
If it runs hot, and remember it only has to stop falling to do that: the hike conversation reopens with Hammack holding a live microphone the very next morning, the 10-year auction at noon Wednesday gets genuinely awkward, and a 14.90 VIX turns out to have been an invitation.
The honest posture is the boring one. Do not be a hero in front of a print that has this much room on both sides of it. Whatever conviction you have at 7:29 AM Wednesday, you will be able to buy the same idea at 7:35 with actual information attached. The only thing you lose by waiting is the story you were going to tell at dinner.
Sunday night. Nobody on the scale yet.
Follow us on X and Instagram for daily market updates
Markets don't wait, and neither do we. Follow Tracking the Trade for fast takes, the numbers that matter, and the occasional meme that hits a little too close to your portfolio.
Instagram: @trackingthetrade
Tracking the Trade is written for entertainment and general education. It is not investment advice, it is not a recommendation to buy or sell anything, and it is emphatically not a substitute for a professional who is legally obligated to care what happens to your money. We are none of those things. Every number here came from a data pull dated the weekend of August 8-9, 2026, and markets have a documented habit of changing their minds before you finish reading. Do your own research, size your own positions, and please stop taking financial advice from a newsletter that opened with a joke about a bathroom scale.
If you work in fintech or finance, you already have too many tabs open and not enough time.
Fintech Takes is the free newsletter senior leaders actually read. Each week, I break down the trends, deals, and regulatory moves shaping the industry — and explain why they matter — in plain English.
No filler, no PR spin, and no “insights” you already saw on LinkedIn eight times this week. Just clear analysis and the occasional bad joke to make it go down easier.
Get context you can actually use. Subscribe free and see what’s coming before everyone else.