The Friday Wrap — September 11, 2026: core CPI, and a $100 barrel
Archived exactly as it was sent. Prices, links and product details are as of September 11, 2026 and are not updated afterwards.
Four sessions lower, one back, and an index that finished within a percent of where it started. SPY closed at 764.29, off 0.77% on the week. The number is small and the week underneath it was not.
The Fed trade flipped
July frames what happened here, so start there. The committee held the target range at 3.50%–3.75% for a fifth consecutive meeting on a 9-3 vote, and all three dissenters wanted a quarter-point increase. That was the first time since September 2016 that three policymakers broke from the majority, and the direction they broke in has turned out to matter more than the count.
Then the data came in, starting with August payrolls on September 4 at 162,000 against a consensus near 53,000, with unemployment steady at 4.1%. Thursday's producer price index showed the annual rate quickening to 5.4% from 4.8%. Friday's consumer price index held the headline at 3.4% year over year, and the detail underneath it ran hotter: core prices rose 0.3% on the month against a 0.2% forecast.
Traders repriced, and by Friday's close Benzinga put the implied probability of an increase at Wednesday's meeting near 90%, up from roughly a coin flip when the week began. Trackers built on CME FedWatch were closer to 70% earlier in the session. The two readings disagree, and both describe a market that moved hard in one direction.
Bonds said it with more conviction, and the 30-year Treasury yield reached 5.35%, a level last seen in May 2007, while the 10-year touched 4.96% and the 2-year moved above 4.5%. A Treasury buyback on Thursday, meant to steady the long end, bought just $5.2 billion against a $6 billion cap.
Energy is doing the work
The inflation prints are an energy story wearing a monetary costume. Gasoline rose 3.9% in August and accounted for more than a third of the month's increase in consumer prices, running 27.4% above a year ago. Diesel is the sharper problem. AAA put the national average at a record $6.06 a gallon on Friday, 63% higher than a year earlier, and diesel is what sets the cost of moving freight, working farmland and running construction equipment. An energy shock reaches everything else through that line.
The supply side explains the price, and West Texas Intermediate pushed above $100, its highest since July, after another round of strikes around the Strait of Hormuz. Iran's Houthi allies struck energy facilities inside Saudi Arabia this week, Saudi crude production fell by roughly 1.9 million barrels a day in August, and tanker rates are at record highs. The International Energy Agency took a further 1.4 million barrels a day out of its 2026 global supply forecast on Friday.
Our own bars make that legible, because USO, the WTI proxy, gained 9.12% on the week and 15.86% month to date and now sits 41.95% above its 200-day average. The proxy fell 2.2% on Friday, which is most of the reason equities could bounce at all.
Households have noticed, and the University of Michigan's preliminary September sentiment reading sank to 47.8 from 51.7, missing expectations of 51.0 and marking the second-weakest print in a survey that dates to the 1950s. Year-ahead inflation expectations climbed to 4.6% from 4.0%. Survey director Joanne Hsu pointed at fuel costs and renewed trade friction.
Two beats, two lower stocks
Oracle and Adobe both reported Thursday after the close. Both beat. The tape took neither of them well.
Oracle's fiscal first quarter was, on headline numbers, the strongest AI-cloud print of the year. Revenue grew 30% to $19.3 billion, and adjusted earnings came in at $1.92 against the $1.67 the broker's own surprise record carried as the estimate. Cloud infrastructure revenue jumped 121% year over year, accelerating from 93% the prior quarter, and remaining performance obligations reached a record $664 billion on more than $30 billion of new AI infrastructure contracts signed in the quarter.
Then the quality. Non-GAAP gross margin fell to 61% from 68.7% a year earlier, short of the roughly 62% the street carried, per Morgan Stanley via MT Newswires. Capital expenditure ran $28.5 billion in the quarter against a fiscal 2027 guide of $90 billion to $95 billion, and free cash flow was about negative $5 billion. Growth at that scale is being bought, and the margin line is where the receipt shows up.
The stock is where the argument resolved, because Oracle popped roughly 6% after hours on Thursday and then closed Friday at 150.26, down 1.76% on the session and 5.37% on the week. It had traded as high as 170.70 on Monday. From that intraday high to Friday's close is a 12% round trip in four sessions, on a quarter the company beat.
Adobe's print rhymed, with earnings of $6.13 beating the $5.84 estimate and extending a run that covers every reported quarter in the broker's trailing eight-quarter window. Adobe rose 1.37% on Friday and still finished the week down 5.36%. Two large software companies beat on the same night. Both are worth less than they were on Monday.
The hardware side of the same trade was paid instead. Skyworks led the S&P 500 with a 28.1% week, its best since 2009, on the Apple foldable launch and a late-stage merger with Qorvo. Hewlett Packard Enterprise added 17.7%, Corning 15.0% on a multi-year fiber agreement with Verizon, and Dell 14.6%, including a 12% Friday after RBC initiated coverage at outperform.
What the tape did with it
SPY spent the week under its short averages, with price 0.19% below the 10-day and 0.34% below the 20-day while holding 0.75% above the 50-day and 7.34% above the 200-day. The gap between the 20-day and the 50-day narrowed to 1.09% from 1.23%. The weekly frame is still intact, with price 0.52% above the 10-week average and 0.21% above the 8-week, but the weekly MACD histogram has rolled over to -0.36. Price repaired faster than momentum did.
Friday's candle earns a look, because SPY gained 0.85% and printed a small body with an upper tail running 60% of the range, closing below its open. The bounce happened and it was sold into.
The divergence is the honest part of the week. QQQ lost 0.57% while IWM lost 2.41%, and small caps closed at 288.89, below their 10-, 20- and 50-day averages, with daily RSI at 38.7 and stochastics at 12.9 and 7.8. That close sits 0.60% above the 60-day low of 287.18. QQQ is within a tenth of a percent of both its 10- and 20-day. One index is still arguing about direction and the other has already answered.
Month to date reads SPY -0.36%, QQQ -0.26% and IWM -1.71%, a spread of 1.45 points between best and worst. Zoom out and the ranking inverts, because for the year IWM leads at +17.85%, with QQQ at +16.65% and SPY at +12.68%. The index that had the worst week and the worst month is still the one that has paid most in 2026.
Cross-asset, in ETF proxies: TLT fell 1.63% and IEF 1.34%, which fits the yield move above. GLD lost 1.97% on the week despite an inflation scare, which is not what the narrative would predict. UUP, our dollar proxy, finished flat at -0.04%; a market pricing a near-certain increase did not bid the dollar, and the disagreement goes in the letter unexplained. VIXY is a futures ETF and not spot VIX, and it gained 1.76% on the week while falling 4.74% on Friday.
Open interest, from the OCC file as of September 9, had the September 11 expiry concentrated at 770 on the call side with 24,005 contracts, and at 760 on the put side with 45,532, with 755 puts at 35,224. The September 14 book is thinner and further out, with 785 and 800 calls against 750 and 755 puts. Those are contract counts and nothing more, and because we hold no greeks we will not call it gamma.
The two-red-week analog the model tracks is not in effect this week.
What this sets up
Good news stopped paying. A record AI-cloud quarter gave back 12% from Monday's high, two beats finished the week lower, and the index held together mostly because crude backed off on Friday. Rates are the common cause, which puts the whole argument on Wednesday's calendar.
That day carries both halves at once, with August retail sales at 8:30 ET and the FOMC decision at 2:00 ET, the press conference behind it. Jobless claims follow on Thursday.
The levels are where this gets settled, and SPY's 10- and 20-day averages sit at 765.72 and 766.88, directly overhead, with the 50-day at 758.62 the first structural support beneath. Under that, the 20-day traded range bottoms at 756.64 and the 60-day at 716.58. IWM's question is narrower and has one number: 287.18 is the 60-day low, and Friday closed 0.60% above it. QQQ sits on its 10- and 20-day and resolves whichever way it leaves them.
Which of those gives is not ours to say. Where they are is.
On the board
Coil closed the week in day mode with the macro layer at risk-off and defensive, which stands longs down and treats the weakest-names list as the two-way read. The SPX book reads names-on and chop-capped, with SPY and six of eleven sectors open and 262 names fully qualified. The QQQ book reads names-on but below its 50-day, with three of six sectors open and 39 names qualified. The crypto book has its gate open and both sleeves in, with BTC 9.4% above its 200-day and 33 of 81 names on the RH-tradable panel above their own.
What we do not carry: dark-pool prints, closing-auction imbalance, spot VIX and true dealer greeks. We would rather show a thinner sheet than a fabricated one.
Read the board at coil.trade/scanner · own the engine, $79 one-time, at coil.trade/buy/engine.
The Friday Wrap is impersonal market research. It is not investment advice, it is not a recommendation to buy or sell any security, and it does not take your circumstances into account.
Sources — prices, ranges, moving averages and weekly frames are our own Alpaca SIP bars through Friday's close. Single-stock sessions, closes and earnings surprise records are from the broker's own data. Open interest is OCC, as of September 9, 2026. Macro and company reporting: Benzinga, MT Newswires, CNBC, Kiplinger, Al Jazeera, U.S. News, AAA and the University of Michigan Surveys of Consumers.
The Friday Wrap is impersonal market commentary. It is not investment advice, not a recommendation, and not tailored to anyone’s circumstances. It contains no forecasts, targets, or trades. Prices come from our own market-data bars; every external claim names its source in the letter.