The Friday Five: Issue #13

I hope you all had a great week. This is Friday Five, Issue #13, and this was the week the Federal Reserve raised interest rates and the bond market said thank you.

On Wednesday the Fed lifted its target range a quarter point, to 3.75 to 4.00 percent, the second leg of a tightening cycle the market spent all summer betting would be a cutting cycle. Short term yields went up, as you would expect. The 30 year Treasury yield went down, three basis points, to 5.33 percent.

That is not a contradiction. A central bank raising rates into an inflation scare is telling you it intends to win, and when the bond market believes that, it stops demanding extra compensation for lending out to thirty years. Short rates up and long rates down is the market saying the medicine is working.

The whole week follows from that. Anything whose value sits far in the future got cheaper to own, and anything held for its dividend got harder to justify. Technology rose 2.36 percent, Utilities fell 3.34 percent, and nothing about either group's business changed in five sessions.

One more thing worth holding onto. The S&P 500 rose 0.51 percent and the equal weight version of the same index fell 0.41 percent. The index went up while the average stock in it went down, and that has now happened three weeks running. Alright, let's get into it.

The Week in 60 Seconds

This Week in One Minute

The Federal Reserve raised rates on Wednesday and the 30 year Treasury yield fell anyway, which is the whole week in one sentence.

The Federal Open Market Committee lifted the target range a quarter point to 3.75 to 4.00 percent on September 16. The front end did what you would expect, with the 5 year yield climbing 13 basis points to 4.86 percent. The long end did the opposite: the 30 year fell 3 basis points to 5.33 percent. That combination is a market saying the Fed will succeed, and it repriced equities accordingly. Technology rose 2.36 percent and Health Care 1.65 percent, while Utilities fell 3.34 percent and Financials 1.78 percent. The S&P 500 gained 0.51 percent and the equal weight version lost 0.41 percent, so the index went up while the median stock went down. Volatility collapsed, with the VIX falling from 17.84 to 14.81.

The single thing to watch next week is the Treasury auction calendar. The government sells 2 year, 5 year and 7 year notes on three consecutive days starting Tuesday, and those auctions are the cleanest test of whether the long end stays where it landed.

The Week Ahead

WHAT I EXPECT · 63% · MODERATE

I expect the long duration bid to hold for another week. Specifically I expect technology to keep beating the equal weight market from the September 18 close to the September 25 close, because the mechanism behind this week's move is a discount rate effect rather than a mood: the Fed raised the front end and the 30 year yield still fell three basis points to 5.33 percent, which lowers the rate applied to cash flows that sit far out in time and is worth more to a software company than to a bank. That mechanism holds as long as the long end stays contained, and there are three Treasury auctions next week that test it directly. I put the odds of this view holding at 63 percent.

What would prove me wrong. The Technology sector fund XLK trails the equal weight S&P 500 fund RSP by more than 2.0 percentage points from the September 18 close to the September 25 close, measured as the percentage change in XLK minus the percentage change in RSP.

THE BIGGEST RISK · 20% · BY OCTOBER 2

Three of this week's five picks are the same bet on a falling long end, and a back up in long yields takes Generac, Zscaler and Natera down together.

Generac at 20 times forward earnings, Zscaler at 40 times, and Natera at 19.6 times sales with no profits are all long duration assets whose repricing this week tracked the 30 year yield falling to 5.33 percent and the VIX collapsing from 17.84 to 14.81. None of the three rerated because cash flows arrived sooner. The Treasury sells 2 year, 5 year and 7 year paper on three consecutive days next week into a market that has just absorbed a rate rise, and long run inflation expectations print on September 25.

What would confirm it. The 30 year Treasury yield closing above 5.45 percent, more than 12 basis points above its September 18 close of 5.33 percent, on or before October 2, 2026.

Last week’s call, scored.
In Issue No. 12 I said Energy would stop leading, and that the Energy sector would trail the S&P 500 rather than beat it from the September 11 close to the September 18 close. That call was falsified. I said I would be wrong if Energy beat the index by more than 1.0 point, and it reached 2.12 points on September 15. That is not a near miss, it is a miss by more than the width of the threshold itself. The uncomfortable part is that by Friday's close Energy had given all of it back and trailed the index by 0.93 points, so I was right about where the week ended and wrong about the path, and the call was already broken by then. This is the second falsified call in a row and both were the same mistake, declaring an energy supply premium finished before it was.

Market Risk Score

What supports the market. Volatility collapsed, with the VIX falling from 17.84 to 14.81, its lowest reading in months. The long end of the curve fell despite a rate rise, which is the bond market voting confidence in the inflation path. Credit showed no stress. Technology and Health Care, together roughly 45 percent of the index, both rose.

What works against it. The Fed is tightening into a supply shock rather than into demand strength, and crude is still above 99 dollars a barrel. Financials fell 1.78 percent on a flatter curve, which is the market telling you lending margins compress from here. Utilities fell 3.34 percent. And a VIX at 14.81 in the week of a surprise rate rise is not calm earned by good news, it is calm bought cheaply.

The read: this is a neutral tape whose main hazard is its own composure. Nothing in the data says sell, and nothing in the price says anyone is being paid to hedge.

Breadth. Cap weight beat equal weight by 0.92 percentage points this week, after 1.14 and 0.87 in the prior two. Three straight weeks of the index beating the median stock is a steady narrowing rather than a new one, and at under one point a week it does not yet clear the bar where I would rebuild the issue around it.

Sector Rotation Update

Into: Technology (+2.36%), Health Care (+1.65%), Consumer Staples (-0.35%, best of the decliners).

The mechanism this week is unusually clean, and it is not a mood. The Fed raised the front end and the long end fell, so the rate used to discount distant cash flows came down while the rate you earn on cash went up. That combination is worth the most to companies whose profits arrive late and the least to companies that are bought as bond substitutes. Technology is the former and Utilities are the latter, which is why they finished 5.7 points apart. Financials are a third case: banks do not mind high rates, they mind a narrow gap between short and long rates, and that gap just narrowed.

Do flows confirm it? Partly. Volume was real in the names that moved, with Zscaler at 2.7 times average on Monday and Generac at 6.7 times on Thursday. But Friday was a quarterly expiration session, when four classes of derivatives settle at once, so its volume and some of its closing prices carry positioning noise rather than conviction. I would not read much into any single Friday print this week, my own picks included.

Does it last? This rotation runs off the rate path rather than a headline, which normally makes it durable, and the condition that ends it is testable: if long yields back up, the mechanism reverses and the leadership goes with it. My own record argues for humility, since betting on this week's leader to keep leading has been the least reliable call I make. Four picks sit in sectors that rose or held; Marathon sits in an Energy sector that fell, which its bear case addresses.

The Five

Three of these five are the same bet, and you should know that before you read them. Generac, Zscaler and Natera are all long duration assets: a generator order book that does not deliver until 2027, a software company that does not yet report a profit, and a diagnostics company trading at roughly 19.6 times sales with a net loss. All three repriced this week in the same conditions, a falling long end and a collapsing VIX, and none of the three did so because cash arrived sooner.

The uncomfortable reading is that this is one position wearing three tickers, and holding all five does not diversify it. Diversification comes from what a holding is exposed to, not from how many sector labels it spans. The honest distinction is that Generac's re-rating has a signed delivery schedule underneath it while the other two rest on a narrative and a dataset. If the long end backs up I would expect all three to fall together, and Generac to be the one that recovers, because a purchase order survives a change in sentiment and a multiple does not. That is why it is the Stock of the Week, and why the risk I named is about the 30 year yield rather than any one company.

Generac Holdings (GNRC)

Industrials / Electrical Equipment

+14.19% this week

The catalyst.
Generac signed a transaction agreement with Amazon on September 16 and filed it the same day. Initial deliveries of data center backup generators are expected to total 2.4 billion dollars in 2027 and 2028, scaling to 8 billion dollars of aggregate payments. Amazon took a warrant for 1,693,745 shares at 200.9266 dollars, 307,954 of which vested immediately. The stock rose 18.3 percent the next session on 6.7 times average volume. U.S. Securities and Exchange Commission, September 16, 2026.

Why it made the five.
Generac has spent years trying to stop being a company whose revenue depends on storms, and this is that transition in one document. The initial delivery figure alone is more than half of what the whole company sold last year, and it is contracted rather than forecast. It also lands in the physically constrained part of the artificial intelligence build out: data centers need backup power before anything else, and that has to be manufactured.

The case against it.
The customer is now a shareholder, so the reward for delivering is partly paid in dilution. Revenue does not start until 2027, leaving five quarters in which nothing is confirmed by results. One buyer anchors the growth case and holds the leverage that comes with it, and a quiet hurricane season means the residential business carries nothing while you wait.

What to watch next.
Third quarter results are expected in late October, though Generac has not yet announced the date. What matters is not earnings but whether management quantifies large megawatt capacity for 2027 and lifts the data center outlook. If the Amazon volume merely fills capacity other commercial customers would have taken, the deal is less additive than the price assumes.

The takeaway.
A weather business became a contracted infrastructure business in one filing, and the stock still trades near 20 times forward earnings, 30 percent below its own high.

Marathon Petroleum Corporation (MPC)

Energy / Refining and Marketing

+8.27% this week

The catalyst.
Crude eased while refined product margins did not. West Texas Intermediate fell about 0.5 percent on the week to 99.53 dollars as Saudi Arabia signalled it could restore roughly half the damaged East to West pipeline capacity within days, while diesel crack spreads held at record levels. Marathon closed at an all time high of 424.89 dollars on September 18, a sixth consecutive weekly gain. CNBC, September 17, 2026.

Why it made the five.
A refiner does not sell oil, it sells the gap between oil and the fuel made from it, which is why this is the one pick that improves when crude falls. Marathon's realized margin ran at 36.33 dollars a barrel last quarter against 17.58 a year earlier, and the reason is supply: drone strikes have removed more than 2.8 million barrels a day of Russian refining capacity.

The case against it.
This is the most extended name in the issue. It closed at 99.3 percent of its 52 week high, above the average analyst target of 370.17 dollars, after six straight up weeks. It trades at 7.5 times forward earnings against 14.7 times trailing, and that gap is not cheapness, it is a windfall extrapolated. Energy also fell 0.95 percent this week, so the pick fights its own sector.

What to watch next.
Third quarter results are confirmed for November 3, where the realized margin per barrel is the whole report. Before that, track the Saudi pipeline repair: full operations are flagged within roughly six weeks, and cracks should start narrowing before that work finishes rather than after.

The takeaway.
Marathon is earning genuinely extraordinary margins for genuinely temporary reasons, and its forward multiple quietly assumes otherwise.

The Kroger Co. (KR)

Consumer Staples / Food Retail

+5.36% this week

The catalyst.
Kroger reported on September 11 with identical sales excluding fuel up 0.2 percent and adjusted earnings of 1.09 dollars a share. It cut full year identical sales guidance to 0.2 to 0.8 percent from 1.0 to 2.0 percent, citing pharmacy headwinds and cautious consumer spending, while reaffirming adjusted earnings of 5.10 to 5.30 dollars and operating profit of 5.0 to 5.2 billion dollars. Digital sales rose 20 percent, retail media 24 percent. The Kroger Co., September 11, 2026.

Why it made the five.
A company lowered what it expects to sell, kept what it expects to earn, and rose 5 percent. That tells you what this market pays for, and it is margin control rather than growth. Two businesses do the work: digital, growing 20 percent, and retail media, growing 24 percent, both with far better economics than groceries. At 11.4 times forward earnings and a 2.6 percent dividend, almost none of that is priced.

The case against it.
Holding profit while cutting sales works until costs move, and this is a business whose costs are moving: diesel is a delivery charge on every item in the store, and it rose 24.1 percent in a month in the August producer price data. Kroger named cautious consumer spending itself, which margin discipline cannot fix. The stock touched 62.25 dollars on Tuesday and closed the week at 60.00.

What to watch next.
Third quarter results are expected in early December. The test is whether identical sales land in the upper half of the reduced range while gross margin holds. A second cut would mark the reaffirmed profit guidance as optimism, and the fade off Tuesday's high suggests some readers got there already.

The takeaway.
The margin lever still has travel in it, and the reason Kroger needs the lever is a customer spending less.

Zscaler, Inc. (ZS)

Technology / Cybersecurity

+20.69% this week

The catalyst.
Security software repriced as a group on September 14 after weekend essays from frontier artificial intelligence lab chief executives warned that autonomous agent swarms could cause damage running to hundreds of billions of dollars. The market's reasoning was that a worse threat environment lifts security budgets whatever else happens. Zscaler rose 16.5 percent that session on 2.7 times average volume, with CrowdStrike and Palo Alto Networks moving alongside it. 24/7 Wall St., September 14, 2026.

Why it made the five.
Of the names that moved together on Monday, this is the one with something underneath the story. Zscaler reported on September 3 with revenue of 898 million dollars against an 877 million estimate, up 25 percent, and guided fiscal 2027 revenue to 3.91 to 3.94 billion dollars. The sector got a narrative; this company already had numbers. It also held its gain into Friday while CrowdStrike and Palo Alto each gave back roughly 3 percent.

The case against it.
Nothing about the business changed on September 14. A 16.5 percent move on an essay is multiple expansion, and that reverses without warning: my ledger records a software re-rating in Issue No. 10 that I named as a risk and that handed itself back within two weeks. The stock trades at 40 times forward earnings and 9.6 times sales, and still loses money, with the net loss widening 52 percent.

What to watch next.
First quarter fiscal 2027 results are expected in late November, where billings rather than revenue settles it. Revenue reflects contracts already signed; billings reflect what customers committed to while the threat narrative was loudest. Billings accelerating past the 25 percent growth rate would turn a story into a fact.

The takeaway.
Even after a 20 percent week Zscaler sits 41 percent below its 52 week high, which makes this a partial recovery rather than a stock running away from itself.

Natera, Inc. (NTRA)

Health Care / Diagnostics

+13.26% this week

The catalyst.
At the World Conference on Lung Cancer, Natera presented the largest molecular residual disease dataset ever shown in the disease: 1,129 patients with resected stage 1 to 3 non small cell lung cancer, followed from 2016 to 2025. Patients testing positive shortly after surgery carried roughly five times the risk of recurrence or death during surveillance. The stock closed at a record 369.39 dollars on September 18 on 2.7 times average volume. Natera, September 18, 2026.

Why it made the five.
Evidence at this scale is how a test moves from useful to expected. The clinical problem is concrete: after surgery a physician must decide whether to give chemotherapy to a patient who may already be cured, and a blood test separating five times risk from baseline changes that conversation. Revenue is already growing 37.8 percent to 2.71 billion dollars, so the commercial engine exists.

The case against it.
This is the widest gap between evidence and price in the issue, which is why it goes last. Natera is unprofitable, with a net loss of 192.35 million dollars, and trades near 19.6 times sales at 99.9 percent of its 52 week high. Prognostic data is not a reimbursement decision: showing a test predicts recurrence is easier than showing that acting on it improves survival, and payers fund the second.

What to watch next.
Third quarter results are expected in early November, where sequential Signatera volume converts data into revenue; it held near a record 34,000 unit pace earlier this year. Watch too for any coverage decision extending reimbursement from prognosis into treatment selection. It has no scheduled date and is the event that would re-rate the business rather than the story.

The takeaway.
Natera holds the strongest dataset in its field and a price that already assumes the people who write the cheques will agree.

Stock of the Week

Generac Holdings (GNRC). This is the only pick this week whose catalyst is contractual rather than atmospheric. Generac did not benefit from a mood, an upgrade, or a conference presentation: it signed an agreement that puts roughly 2.4 billion dollars of scheduled deliveries against a company that did 4.44 billion dollars of revenue in the last twelve months, and the market confirmed it with 6.7 times normal volume. It is also the cheapest large move of the five at about 20 times forward earnings and the only one not sitting near its own high. The dilution attached to the deal is real and named in the bear case, but a purchase order is a harder thing to revise than a narrative.

What Could Change My Mind?

A long end that backs up. My entire read rests on the 30 year yield staying near 5.33 percent. Three Treasury auctions land next week and long run inflation expectations print on September 25. If the 30 year pushes through 5.45 percent, the duration trade reverses, three of my five picks fall together, and the sector rotation I described unwinds with them.

A second energy leg. Crude is still above 99 dollars. Saudi restoration is a plan, not a completed repair, and I have now been wrong twice in a row in the same direction by assuming an energy supply premium was finished. If crude turns back up, Marathon's margin story improves and everything else in this issue gets harder.

Where my own inputs disagree. The rotation read says lean on the leaders; my measured record says leaders are the worst thing I bet on, with sectors I labelled Leading going on to beat the market in 1 of 8 non flat weekly readings. I have taken the mechanism over the base rate because the mechanism is a rate path rather than momentum, but price that as the judgment call it is. One more caveat on my own scoring: across matured picks, a score above 85 has not produced better one month returns than a score below 75. Read the reasoning rather than the number.

What would make me more constructive. Breadth turning. If the equal weight index starts beating the cap weighted index while the long end stays contained, the rally broadens and the main structural worry in this issue goes away.

Scoreboard

This is the running record, losers included. Two numbers matter and they say different things.

The issue week column is not a forecast that came true. Names are screened into the issue partly because of how they moved that week, so they beat the index that week by construction. It cannot come out any other way. It is shown because it is the number readers saw, not because it measures skill.

The number that measures anything is what happened after publication, from the Friday close a reader could actually have acted on. Across 60 picks over 12 issues, one week after publication the average pick returned 0.32 percent against 0.41 percent for the S&P 500, so 0.09 points behind, with 62 percent finishing higher and 53 percent beating the market. At one month, where 40 picks have matured, the average pick is up 4.35 percent against 1.65 percent, 2.70 points ahead, with 63 percent beating it. These picks have not beaten the market in week one and have beaten it by a fair margin by week four.

ISSUE

PICK

ISSUE WEEK

SINCE PUBLISHED

VS SPY

No. 12

Academy Sports (ASO)

+24.12%

-12.77%

-12.43

No. 12

DigitalOcean (DOCN)

+19.79%

+5.85%

+6.19

No. 12

Signet Jewelers (SIG)

+19.13%

+0.12%

+0.46

No. 12

Cloudflare (NET)

+9.37%

+5.57%

+5.91

No. 12

HCA Healthcare (HCA)

+2.93%

+0.18%

+0.52

No. 11

Deere (DE)

+11.52%

-2.57%

-1.80

No. 11

GitLab (GTLB)

+10.04%

-5.70%

-4.93

No. 11

CF Industries (CF)

+9.63%

-0.21%

+0.56

No. 11

Dell Technologies (DELL)

+9.34%

+8.23%

+9.00

No. 11

Best Buy (BBY)

+4.70%

+0.59%

+1.36

No. 10

Okta (OKTA)

+28.88%

+2.63%

+2.52

No. 10

Salesforce (CRM)

+22.69%

+1.26%

+1.15

No. 10

Veeva Systems (VEEV)

+12.61%

-0.58%

-0.69

No. 10

Cleveland-Cliffs (CLF)

+10.15%

+7.57%

+7.46

No. 10

Dollar General (DG)

+4.40%

+8.40%

+8.29

The "since published" column is measured from the publication Friday close to one week later, so its benchmark differs from the issue week figure quoted elsewhere. Academy Sports is the loss worth naming: I flagged the squeezed United States household as the biggest risk in Issue No. 12, and Academy Sports is the pick that took it, falling 12.77 percent in the week after publication. The risk call was right and the pick still lost money, which is the distinction between seeing a hazard and avoiding it.

Terms in This Issue

A Possible Next Step

I'm testing interest in a future Edge Report with entry zones, invalidation levels, sizing tiers, and exit frameworks. It does not exist yet and nothing is for sale. If that is something you would read, tell me and I'll build it.

What's going on, everybody?

The Week in 60 Seconds

This Week in One Minute

The Week Ahead

Market Risk Score

Sector Rotation Update

The Five

Stock of the Week

What Could Change My Mind?

Scoreboard

Terms in This Issue

A Possible Next Step