This is the first Sunday Recap, the weekly email. Every issue is published here as well, so the archive stays open whether you subscribe or not.
A note from our founder
I keep seeing the same argument: earnings are growing, the forward P/E is falling, so the stock must be a bargain. But a cheaper multiple does not tell us the selling is finished. A stock can keep making lower lows while looking cheaper at every step. What I want to see alongside the valuation is evidence of recovery: a higher low, reclaimed levels and progress that holds. Undervaluation can build the case for owning a business. It does not prove the stock has turned around.
Chart of the week: cheaper does not mean recovered
Adobe is the clearest example. From its October 2025 closing peak to October 2, the stock fell 34%, even as its trailing adjusted EPS rose 16.7%. Its P/E fell from 17.8x at that 28 October 2025 peak to 10.1x, and price remained below its 50- and 200-day averages.
Qualcomm tells a different story. It fell 41.2% from its 29 May closing peak to the 31 July low, then rebounded 25.2% into October 2. Its multiple moved with it: 21.1x at the May peak, 13.0x at the July low, 16.3x on October 2. That is compression first and repair second, which is not the same shape as Adobe's. The October 2 close was above both averages despite weaker recent earnings, and that is evidence of price repair rather than a guarantee the recovery holds.
Both P/Es use the latest four reported quarters available at each date. This is trailing adjusted earnings, separate from the forward estimates in the opening note; company adjustments and fiscal calendars still differ.
Read Earnings vs. Price: Recovery Watch
Where the price checks stand
Every level we published on 2 October, and what the market did with it since. An intraday touch is separated from a close, because they are different events.
| Name | What we marked | Through the 2 October close |
|---|---|---|
| $QCOM | 186.37, the first level overhead | Reached. Traded to 188.95, closed back below at 184.87. The run of lower highs that chart described ended there. |
| $AVGO | 367.42, the 200-day, and the 362-372 band | Not reached. High of 357.35, closed 355.14, still under both. |
| $ASTS | the 84-90 heaviest-traded band | Not reached. High of 59.54. |
| BTC | the 88,000-92,000 band | Not reached. High of 87,219, and still no trade above 88,000 since 29 January. |
| $NFLX | the 200-day at 84.20 | Not reached. High of 67.87, and still no close above that average since April. |
| $SMCI | 44.58 as the highest since 9 June | Held. The settled high was 44.585; it closed 43.69. |
| $SOFI | the 50-day at 17.47 and the 17.38-17.80 weekly band | Not reached. High of 16.25, still underneath. |
We publish every level we marked, including the ones the market never came near. These are not all the same kind of mark: some sat overhead, and $SMCI's was a high already made, so they are listed rather than scored.
One correction belongs here too. That $SOFI note also carried a weekly 200-period average near 20.40, taken from a circulated chart. Recomputed from the weekly closes it is 13.08, and the chart on the site was rebuilt without it.
These are dated observations, not a trading record. A level being reached does not establish a recovery, and an average moves as the window rolls.
How last week's earnings ended
Two names on our pages reported Thursday, in different ways. Accenture reported before the open, so Thursday was its reaction day and Friday was the second. Nike reported after Thursday's close, so Friday was its first regular session on the news.
| Name | Reported | Reaction | Where Friday left it |
|---|---|---|---|
| $ACN | Thursday, pre-market | +15.78% Thursday | -6.31% Friday. Still +8.47% against the pre-print close, so a little over half the move held. |
| $NKE | Thursday, after the close | -3.64% Friday, its first session | Traded down to 31.97, a 52-week low, before closing 33.87. |
Accenture ended 12.6% under its post-print high of 227.63. Nike beat and fell anyway: by Alpha Vantage's consensus it was the thirteenth straight quarter it has beaten on EPS, with the last miss reported in June 2023, and the stock still traded as much as 9.1% below Thursday's close before recovering into the bell.
Neither outcome says the result was misread. It says the result and the share-price response are separate questions, which is the same distinction the chart above is making.
This week's stock reviews
- $META: Q2 free cash flow did not cover that quarter's dividend. Watch cash generation alongside infrastructure spending and repurchases.
- $DPZ: Negative book equity makes conventional book-value screens misleading. Does the second low above the first hold?
- $MU: Customer contract liabilities reached $12.90 billion. Customer prepayments carry obligations; they are not already-earned revenue.
- $MAT: A reported takeover approach drove an 18.7% rise. The implied offer value is not an agreed transaction.
- $DKS: Acquired Foot Locker sales lifted revenue while operating margins weakened. Can integration turn that added scale into profit?
- $APP: Reported earnings are growing, but the cheaper forecast multiple has not confirmed a stock-price turnaround.
- $ASTS: Satellite deployment is advancing. Revenue delivery, capital spending and recovery above the daily averages remain separate checks.
Two comparisons worth opening
Uber vs. Lyft: Lyft has the lower forecast multiple and stronger current daily-average position. Uber has stronger reported group-level margin and cash evidence. Acquired growth and different business scopes complicate the comparison.
Amazon vs. Alphabet: Similar forecast multiples, different cloud operations and cash bills. Both stocks ended Friday below their 50-day averages and above their 200-day averages.
The week's movers
Friday 25 September close to Friday 2 October close, across the 167 companies on our watchlist above $10 billion.
Up: $IBRX +21.9%, $CCL +15.8%, $LITE +15.3%, $SNPS +15.1%, $COHR +13.9%.
Down: $FICO -23.4%, $CBRS -19.5%, $APP -13.7%, $MDB -12.8%, $TEM -9.9%.
A week reads differently from a day. $FICO rose 11.7% on our Thursday board and still finished the week down 23.4%, so the rebound came after the damage rather than instead of it. $APP appears here the same week we reviewed it, which is the question that review was built around.
Two things about this list. It is our watchlist, not the market, and that watchlist grows during the week: names are often added because they moved, so read it as what we were watching rather than a screen of a fixed universe. And a week's return says what happened, not why.
The week ahead: October 5-9
Eight dates are listed on our earnings page against company announcements. All times below are Eastern.
- Tuesday, October 6: $RPM reports before the open; $LW releases at approximately 8am. $STZ reports after the close, with its call Wednesday at 8am.
- Wednesday, October 7: $APLD reports after the close, with a 5pm call. $LEVI also has a 5pm call.
- Thursday, October 8: $PEP releases around 6am, with analyst questions at 8:15am. $TLRY reports before the open, with an 8:30am call.
- Friday, October 9: $DAL's September-quarter results call is at 10am.
Applied Digital is our AI-infrastructure report to watch for execution and financing commentary. A result and the stock's response answer different questions; a headline beat alone need not repair a weak chart.
Calendar, company sources and estimate caveats
What we'll revisit next Sunday
- Does Adobe begin repairing its chart, and does Qualcomm hold its recovered ranges?
- Does AppLovin's price behavior start agreeing with its reported earnings growth?
- Do next week's earnings reports change the operating outlook, and how do the stocks respond?
We'll check back on these questions, including the ones that remain unresolved.
Free tools and the weekend board
The covered-call and cash-secured-put calculators remain free.
The Weekend Board tracks perpetual contracts while cash equities are closed. Check its timestamp and venue: these are not official stock prices, and contract prices can differ from the underlying markets.
We use TradingView for chart research. If you upgrade through our affiliate link, we earn a commission at no extra cost to you.