X-Ray Analyses›Technology›Applied Materials, Inc.
X-RayNº 10918 November 2026
◆ Follow-upSecond X-Ray of Applied Materials, Inc.. Our previous, dated 15 Aug 2026 at $507.18, concluded “A Real Moat at a Cycle-Peak Price”. This report checks that call against what has happened since. Read it →
X-Ray Analysis◆The Buffett Lens
A

Applied Materials, Inc.

NASDAQ: AMAT·Semiconductors·United States·Explore AMAT live ↗

Applied Materials — the business beat, the multiple broke

◆ The Buffett LensIn August we called Applied Materials a wonderful business at a price with no cushion, and declined to name a number. Three months later the business has done everything asked of it: record revenue, a fourth quarter up 53%, another record year guided for 2027, and analysts' estimates higher than in August. The shares are down 31%. All of the fall is the multiple. That makes the price fairer; it does not yet make it a bargain for a business whose earnings have never been higher. This time we name the number. Closer, not yet: add near $280.
◆ Educational analysis & opinion — not investment advice. Figures as of 18 November 2026. See full disclaimer below.
7
Moat
7
Management & Capital
9
Financial Strength
9
Growth
5
Valuation
7.3
◆ The Scorecard · one-second read
"The business did its part. The market changed its mind about the price."
Q4 FY26 revenue +53%, EPS beat · FY27 consensus $18.41 → $19.29 · shares −31% · 18.1× FY27, 14.2× FY28, 27.7× trailing · export cliff moved to 10 Jan 2027 · $2bn buyback at ~$479
◆ Type · Cyclical compounder, a third cheaper on rising estimatesDividend · Grower — $0.53 a quarter, 0.6%, covered four times by cash★ Follow-up · August's 'no cushion' graded; two corrections, one new risk
Price & total return · live
The Tape, since we wrote
The gold pin marks the price we used. Daily closes, live.
—
Drag across the chart to measure
Live price history is momentarily unavailable · price at this X-Ray: $349.23.
Go deeper — the Trading Desk, 15 years of financials, DCF & peers for AMATOpen AMAT →
◆
Part
I

What we said in August — and what happened

Ten claims, graded in public, three months later

On 15 August we published Applied Materials at $507.18 under the title "A Real Moat at a Cycle-Peak Price". We said the business was genuinely wonderful and getting steadier, that the price required the boom to continue, the margins to keep rising, Washington and Beijing to stay their hands and the cycle not to turn, and that there was "no version of this price that contains a margin of safety". Unusually for us, we did not name a price at which to buy; we described both halves and left the weighing to the reader. The shares closed at $349.23 on 17 November. Here is the account, including two things we got wrong.

What we said on 15 AugWhat happened by 17 NovGrade
Q4 guided to $10.25bn, +51%Q4 revenue $10.41bn, +53%; adjusted EPS $4.16 against guidance of $4.02.✓ Right
'No margin of safety at any point in this price'The shares fell 31% in three months, to a low close of $319.25 on 5 November, while the business beat.✓ Right
'A lower multiple on a lower number' if the cycle turnsHalf of it happened. The multiple fell from about 27.5 to 18.1 times next year's earnings. The number did not: the fiscal-2027 estimate rose from about $18.41 to $19.29, and management guides another record year.◆ Half right
Affiliates Rule: a dated cliff on 9 November, ~$600mThe US Commerce Department extended the suspension to 10 January 2027, alongside the US–China trade truce. The cliff moved; it did not go away.◆ Too early
China rising in dollars, falling as a shareChina revenue rose 9% in fiscal 2026 to $9.33bn, but fell to 27% of sales from 30%, and to 25% in Q4. The finance chief expects less than 25% in 2027.✓ Right
Services growth of 22% 'flattered' by moving 200mm tools into AGSWe were wrong. Both releases say prior periods were restated. On a like-for-like basis AGS grew 20.2% in fiscal 2026. We missed a sentence in the release we were quoting.✗ Wrong
'A +245-day cash conversion cycle'Wrong label. 245 days was the operating cycle (inventory plus receivables days), not the cash conversion cycle, which subtracts the days taken to pay suppliers. The argument stands: free cash flow was 61% of net income in fiscal 2026.✗ Wrong label
Buybacks: 'management stopped buying as the price ran'In the fourth quarter Applied spent $2.0bn on its own shares, its largest quarter in more than eight years, when the shares averaged about $479. The praise was premature.◆ Overtaken
Insiders: sellers, no open-market buyerStill true. Small sales since August (the finance chief sold 7,500 shares at $479.25 on 25 August); no open-market purchase by any officer or director.✓ Right
No named successor to the CEOStill none. Gary Dickerson, chief executive since 2013, remains in post.✓ Unchanged

The uncomfortable grade is not one of the wrong ones; those were mistakes of reading, now corrected. It is the half-right one. We expected that if the price fell it would be because the cycle turned and earnings fell with it. The opposite happened: earnings estimates rose and the price fell anyway. The market did not wait for the cycle; it simply decided to pay less for the same earnings. That matters for what we say now. A fall in the multiple with the business intact is the kind of fall an owner can use. But it also means the earnings the price now rests on are still record, boom-time earnings, and nothing in the last three months has tested what happens when the boom pauses.

And one more thing an honest grader must record: in August we declined to name a number. That was defensible, but it left readers without the one thing a value-investing report owes them. This time we name one.

Our August price → today
$507 → $349
−31.1% in three months
From the 30 June record
−51.7%
$723.00 → $349.23
Fiscal-2027 EPS estimate
$18.41 → $19.29
up 4.8% while the price fell
◆
Part
II

★★ The business beat; the multiple broke

Why the shares fell a third while the forecasts went up

The business beat; the multiple brokeAMAT daily close, $, 1 Apr – 17 Nov 2026.30040050060070012345678AprMayJunJulAugSepOctNov130 Jun · record close, $72321 Jul · AI-overbuild report, −10%315 Aug · our X-Ray, $507413 Oct · 2030 targets; −6.7%530 Oct · Intel −29%; AMAT −7.5%65 Nov · post-midterm crash, −13.6%712 Nov · record Q4; guide up817 Nov · $349.23Price, forecast and multiple, X-Ray to todayFY2027 consensus EPS (Yahoo) and the multiple paid for it.Share priceAug $507.18now $349.23 (-31%)FY2027 EPS estimateAug $18.41now $19.29 (+5%)Price / FY2027 EPSAug 27.5×now 18.1× (-34%)★ The estimate rose 5%; the price fell 31%. The whole fall is a lower multiple.Sources: our price history; Yahoo Finance analysis (17 Nov 2026); press reports for the causes.

The chart tells the story in two lines that went in opposite directions. Between our report and today, the market's estimate of what Applied will earn in fiscal 2027 rose by about 5%. The price fell 31%. Nothing in Applied's own news explains that fall; every large move came from outside the company.

DateWhat happenedAMAT that day
30 JunRecord close after the equipment makers' shares more than doubled in the second quarter.$723.00
1 JulA report that Meta might rent out spare AI computing capacity is read as a sign of over-building (CNBC).−10.0%
27 JulA report that a state-backed Shanghai group is mass-producing immersion lithography tools (The Information, via Investing.com).−3.6%
30 JulLam Research reports a record quarter and guides higher.+15.0%
13 OctApplied sets 2030 targets at its investor breakfast (more than 2 points of market share gained, ~54% gross margin, ~40% operating margin); the shares fall on profit-taking.−6.7%
22 OctLam says its next quarter would have been higher without supply-chain constraints.−8.2%
30 OctIntel falls 29% on a weak fourth-quarter guide.−7.5%
5 NovThe semiconductor index has its worst day in years after the US midterm elections, on fears that a new Congress will slow data-centre construction; no Applied news.−13.6%
12–13 NovRecord fourth quarter and a higher first-quarter guide; at least ten analysts then cut their targets, mostly citing valuation.−3.6%, −2.7%
17 NovOur price, the day NVIDIA reported after the close.$349.23

Two lessons for an owner. The first is that the equipment makers are now traded as one basket with the AI chip designers, and the basket moves on news about AI spending, politics and other companies' quarters. On 5 November Applied fell 13.6% on a day when nothing happened to Applied. An owner who understands the business can use days like that; an owner who reads the price as a verdict on the business will be shaken out by them.

The second is less comfortable. The market's mood changed before the numbers did. That is often how cyclical tops look from the inside: the business is still beating, the guidance is still rising, and the share price has already started to price the turn. We do not know whether this is one of those times. We do know that the people buying at $723 in June were paying for a future that looked certain, and that it looks a good deal less certain now, at a third lower, with the same numbers.

◆
Part
III

The record year

Fiscal 2026 taken apart, and the first quarter of 2027 guided higher still

The boom, quarter by quarterRevenue, $ billions. Q1 FY2027 is the midpoint of guidance ($10.6–11.6bn).7.10Q2 257.30Q36.80Q47.01Q1 267.91Q29.12Q310.41Q411.10Q1 27eChina, falling as a shareChina's share of Applied's revenue.35%Q3 FY2530%FY202528%Q3 FY2627%FY202625%Q4 FY26<25%2027 (CFO)★ Q4 revenue +53% with China down to a quarter: growth came from Taiwan, Korea, the US.2027: the CFO expects China 'probably less than 25%' (third-party transcript). Sources: Applied releases, 10-Q.

Applied reported its fiscal fourth quarter, to 25 October, on 12 November. Revenue was $10.41bn, up 53%; the machines business grew 64% to $8.0bn, services 26% to $1.9bn. Gross margin was 50.6% on the company's adjusted measure and operating margin 35.3%. Adjusted earnings per share were $4.16, against guidance of $4.02. Reported EPS was lower, at $3.40, because of a $653m unrealised loss on Applied's strategic investments; the release does not say which holding.

For the year, revenue was $34.45bn (+21%), reported EPS $12.62 (+46%) and free cash flow $6.20bn (+9%). For the first quarter of fiscal 2027, a 14-week quarter, Applied guided revenue of $11.1bn ± $0.5bn and adjusted EPS of $4.35 ± $0.20. Every line of the guide is a record.

Q4 FY2026Q4 FY2025FY2026FY2025
Revenue$10.41bn (+53%)$6.80bn$34.45bn (+21%)$28.37bn
Semiconductor Systems$8.00bn (+64%)$4.88bn$26.14bn$21.44bn
Applied Global Services$1.90bn (+26%)$1.51bn$6.90bn (+20%)$5.74bn
Gross margin (adjusted)50.6%48.1%50.1%48.8%
Operating margin (adjusted)35.3%28.6%33.1%30.2%
EPS reported / adjusted$3.40 / $4.16$2.38 / $2.17$12.62 / $12.89$8.66 / $9.42
Free cash flow$2.62bn—$6.20bn$5.70bn

Where the growth came from. Foundry and logic were 58% of machine sales in the quarter, memory chips for computers (DRAM) 33% and flash memory 9%; a year earlier DRAM was 28%. The buyers are building capacity for AI: the most advanced logic chips, high-bandwidth memory, and the packaging that stacks them together, which management says will grow about 90% this calendar year. China fell to a quarter of revenue; Taiwan, Korea and the United States made up the rest of the growth.

The cash, honestly. Free cash flow was 61% of net income in fiscal 2026. That is normal for an equipment maker in a boom, because it builds machines before it is paid for them and lends its customers time to pay: receivables rose 73% in a year against revenue up 53%, to about 79 days of sales. We said in August that "cash runs below earnings in a boom". It still does. When the boom slows, the reverse happens, and cash comes in faster than profit; that is the one consolation of the cycle.

◆
Part
IV

★ Is 2027 the peak?

The question the price turns on, and what can and cannot be known

Everything in Applied's valuation now depends on one question: is the boom in chip-equipment spending close to its top, or is it a step to a permanently higher level? Management's answer on 12 November was unambiguous. "We expect 2027 to be another record year for the market and for Applied," the chief executive said, according to transcripts of the call. Spending on wafer-fab equipment would be "up more than 30%", and the most advanced logic and DRAM lines are running "at full utilization". The constraint is not demand but cleanroom space and parts: "Our Q1 number would be slightly higher if everything were perfectly available." Applied plans to double its output of systems from fourth-quarter levels by 2028.

Others say the same. ASML said in October that its planned supply for 2027 and 2028 "will not fully meet the demand" of its customers. Tokyo Electron, as reported, sees chip-tool spending above $210bn in 2027. SK hynix said memory demand would exceed supply until 2030. Lam Research's September quarter grew 54.5%, the same pace as Applied's.

The bull caseThe bear case
Contracted, not hoped for. Customers give Applied rolling eight-quarter forecasts; none has delayed deliveries. More than 20 fab ramps and expansions are tracked for 2027.Every cycle looked contracted at the top. In fiscal 2019 Applied's revenue fell 13% with no recession, two years after customers had said the same things.
Structural shifts. New transistor designs, new DRAM structures and chip stacking add steps per wafer, and Applied claims a larger share of each.The buyers are few and borrowing. Two customers were 34% of revenue in the first nine months; the AI spending they serve is increasingly financed with debt, and politics has entered the data-centre business.
Services soften the fall. AGS is $6.9bn a year, growing 20%, on an installed base that has to be maintained whatever the year.China is shrinking and policy is a cliff. China falls below 25% in 2027; the Affiliates Rule returns on 10 January unless extended again.

What cannot be known. Nobody can tell you when this cycle turns; anyone who claims to is guessing with confidence. What we can say is that every equipment cycle in Applied's history has turned, that the turn has always come while customers were still saying demand exceeded supply, and that the price of the shares today assumes the turn is mild and late. That is a reasonable assumption. It is not a margin of safety.

◆
Part
V

The moat, rechecked

Still breadth, not monopoly — and one correction in its favour

The claim (August)The evidence nowVerdict
Breadth across ~1,000 process stepsMachines business 'on track to grow more than 40%' in calendar 2026, against industry growth the CFO calls 'slightly lower'. If it holds, Applied regains share it lost in 2025.✓ Intact, likely stronger
Pricing for valueAdjusted gross margin 50.6% in Q4, guided ~50.8% for Q1, despite hiring about 2,600 people in a quarter and supply bottlenecks.✓ Intact
Services as ballastAGS +20% like-for-like in fiscal 2026 (our August doubt was wrong), operating margin up from 26.9% to 29.5%. Target above $12bn by 2030.✓ Stronger than we said
Switching costs in yieldHybrid bonding: Applied says it serves about 80% of the steps. New packaging and process-control lines (+50% this year) widen the footprint.✓ Intact
China's toolmakers taking the mature nodesNaura's growth reportedly slowed to about 1% in its third quarter; Applied claims share gains in China in 2026 at 28nm. Local-content estimates range from about 21% to 35% depending on the definition.◆ Slower than feared

We keep the moat at 7. The evidence of the last three months is, if anything, supportive: share likely regained in 2026, margins held through a ramp, services stronger than we credited. We would move it to 8 if the outgrowth holds through the first quarter of fiscal 2027 and the 10-K confirms it. What we do not change is the description. Applied's moat protects its share and its prices; it does not protect its volume. In a downturn, Applied keeps its customers and loses their budgets.

How knowable is the next ten years?
3/5 — the product is explainable; the cycle and the politics are not. We are confident that in 2036 chips will need more deposition, etch and inspection steps than today, and that Applied will sell tools for most of them. We cannot know which years will be booms, how far China will close its doors to American tools, or how long the AI build lasts.
What you must believe to buy at $349.23
  • That fiscal 2027 and 2028 arrive roughly as forecast (about $19.30 and $24.60 a share), and that any slowdown after them is shallow, so that earnings in 2030 are around $28 a share.
  • That the market then still pays about 18 times earnings for an equipment maker, near the top of its pre-2024 range, rather than the 10–15 times it pays at the bottom of a cycle.
  • That the export rules returning on 10 January cost only what management has already sized (about $600m a year), and that the suspended denial order in the BIS settlement is never reinstated.
◆
Part
VI

The competition

Everyone in the business is booming; the question is who gains share

RivalWhere it meets AppliedThe latestRead
Lam ResearchEtch and deposition, above all in memory.September quarter $8.22bn, +54.5%; December guide $8.6bn; China 35% of revenue.◆ Matching pace
Tokyo ElectronDeposition, coating and etch.Forecasts operating profit up 87% this fiscal year; sees chip-tool spending above $210bn in 2027 (third-party summary). China 30–35% of sales.◆ Matching pace
KLAInspection and measurement, where Applied is small.June quarter $3.66bn; December guide $4.7bn (third-party).◆ Different pond
ASMLLithography (no overlap); a read-across on demand.Raised 2026 guidance to €44–45bn; supply for 2027–28 'will not fully meet' demand.✓ Confirms the boom
Naura, AMEC (China)Mature-node deposition and etch inside China.Growth reportedly slowing; localisation figures disputed.◆ The long threat

The pattern. In August the worry was that Lam and Tokyo Electron were taking share from Applied. The latest quarter shows no new gap: Lam grew 54.5%, Applied 53%, and Applied says its machines will outgrow the market in calendar 2026. What the comparison shows instead is how correlated these businesses are. They rise and fall together, on the same customers' budgets. Owning two of them is not diversification; it is a larger bet on the same cycle.

Across our board. We wrote last week that ASML's fall was "the scare we kept the powder for" and that TSMC, below our level, should be bought with both hands. Applied is different from both. ASML has a monopoly and TSMC a lead no rival can close; Applied has breadth, which is a fine moat but a less certain one. And on trailing earnings Applied is still dearer than its own history, which neither of the others is.

◆
Part
VII

Washington and China

A cliff postponed, and a risk we did not report in August

The Affiliates Rule. In September 2025 the US Commerce Department's Bureau of Industry and Security extended its export restrictions to companies at least half-owned by listed Chinese entities, which Applied estimated would remove about $600m of fiscal-2026 revenue. The rule was suspended in November 2025 under the US–China trade arrangement, and in August we called the expiry of that suspension, on 9 November 2026, a dated cliff. On the eve of that date BIS extended the suspension to 10 January 2027 (Federal Register 2026-22793), in step with the trade truce. Applied did not discuss it on its results call three days later. The cliff has moved by 62 days; its size and its politics are unchanged.

A risk we missed. In February 2026 Applied paid $252.5m to settle BIS charges over shipments of ion implanters that reached China's SMIC. We reported the fine. We did not report what Applied's 10-Q of 20 August spells out: the settlement includes a denial order that is suspended, and will be waived after three years if Applied completes the required audits on time. If the suspension were lifted, Applied says it would be prohibited from exporting certain of its products outside the United States. The chance is small; the consequence would be grave. It belongs in any owner's list of risks, and now it is in ours.

Policy itemStatus (17 Nov 2026)What would change it
Affiliates RuleSuspended to 10 Jan 2027Expiry or extension alongside the US–China truce
China share of revenue25% in Q4; <25% guided for 2027New controls on mature-node tools or servicing
BIS settlementPaid; denial order suspended for three yearsFailure of the required audits
US data-centre politicsPost-midterm fears of slower permittingActual moratoriums in large data-centre states
◆
Part
VIII

Management & capital allocation

A $2bn buyback near the top, small insider sales, and still no successor

G
Gary Dickerson · President & CEO since 2013
Led Applied through two cycles and the shift to selling 'solutions' across process steps. Sold about $105m of shares at $590–736 between April and July 2026 (reported in August). Still no named successor after thirteen years.
B
Brice Hill · SVP & Chief Financial Officer
Has described the policy, begun about three years ago, of charging for the value Applied delivers, which took gross margin above 50%. Sold 7,500 shares at $479.25 on 25 August 2026; holds about 129,000 including unvested units.
A
Akash Palkhiwala · Director since 27 August 2026
Qualcomm's finance chief (since 2019) and chief operating officer (since 2024); joined the board and its audit committee.

The buyback, regraded. In August we praised management for cutting buybacks to $440m in a quarter while the shares ran, calling it "the behaviour of people who understand what a share is worth". In the next quarter Applied spent $2.0bn, its largest in more than eight years. The exact average price will be in the 10-K in December; the shares averaged about $479 in the quarter, a third above today's price. Over the whole of fiscal 2026 Applied returned $4.75bn, about 77% of free cash flow, below its own target of 80–100%. A company that buys more when its shares have fallen 35% from a peak is not being foolish. But it is not the price discipline we described, and an owner should know that our praise was premature.

Insiders. Since August, three small sales: the finance chief (7,500 shares at $479), a director (1,000 at $465) and the controller (725 at $545). No officer or director has bought shares in the open market since at least April. Succession remains unaddressed: Mr Dickerson has run Applied since 2013, and the company has named no successor.

Besi. On 26 October Applied agreed to raise its stake in Besi, the Dutch maker of hybrid-bonding tools, to about 20%, without seeking a board seat. It is a sensible way to stay close to a technology Applied says it serves 80% of; it also means Applied's reported earnings will now move with Besi's share price, as the $653m unrealised loss in the fourth quarter suggests.

Who owns it
Institutions
Index and long-only funds
Vanguard and other index holders lead the register; our feed's ownership list carries duplicate and mislabelled Vanguard rows (Part IX).
★ Insiders
Small sellers
Three sales since August, the largest $3.6m; no open-market buy since at least April 2026.
Analysts
Mean target ~$562
S&P Global data: 40 analysts, average $561.89, range $358–900. At least ten cut targets on 13 November, mostly on valuation.
The company
$10.8bn of buyback room
After $2.0bn in Q4; the policy is to return 80–100% of free cash flow over time.
◆
Part
IX

The numbers

A fortress balance sheet, a boom-time working capital, and a screen with six faults

At 25 October Applied held $9.3bn of cash and short-term investments and $4.8bn of longer-term investments, including its strategic stakes, against $6.5bn of debt: net cash of about $7.6bn counting the long-term investments, $2.8bn without them. Operating profit covered interest about 40 times. Customer prepayments rose 40% to $3.6bn, a sign of how tight supply is. The balance sheet is not a risk in any downturn we can imagine; its strength is why we score financial strength at 9.

The working capital is the part to watch. Receivables rose 73% in a year to $9.0bn, about 79 days of sales against 69 a year earlier; inventories rose 16% to $6.9bn. In a boom, an equipment maker lends its customers time. The 10-K in December will show whether that lending is ordinary timing or something more generous.

★ Six faults in our feed — read these before trusting a screen
What the screen showsWhat is wrong, and the number to use
Two Q1 FY2026 rows (operating income $2,096m and $1,831m)A duplicate quarter. The right one ended 25 January 2026, with operating income of <b>$1,831m</b>; the full year minus the other three quarters confirms it.
Payables $6,696m, accrued expenses $0; cash conversion cycle 99 days$6,696m is accounts payable <i>and</i> accrued expenses together. Splitting them (Q3: $2.39bn payables) gives a much longer cash conversion cycle; the 99 days on the screen is too short.
Segment revenue for FY2025: $37.4bn; no FY2026 rowStale rows mixed with current ones. Actual FY2025 (restated): Semiconductor Systems <b>$21.44bn</b>, AGS $5.74bn, other $1.19bn; FY2026: $26.14bn, $6.90bn, $1.40bn.
Dividend summary empty; the 10 September declaration missingFour payments in the last twelve months (<b>$1.98</b>); the next $0.53 is payable 10 December to holders on 19 November.
Analysts' mean target $643Stale, from before the cuts of 13 November. S&P Global's average is <b>$561.89</b> (40 analysts), low $358.
Desk note: 'cash conversion 106%'; 'leadership in metrology'Operating cash flow was <b>88%</b> of net income in fiscal 2026 (free cash flow 61%). Applied is a small player in metrology, where KLA leads.
◆
Part
X

The dividend

Small, safe and growing 15% a year — tested

Applied pays $0.53 a quarter, $2.12 a year, a yield of 0.6% at today's price (0.4% in August). It raised the dividend by 15% in March 2026, after +15% in 2025 and +25% in 2024, and has raised it for nine years running. The next raise is customarily declared in March. Nobody owns Applied for its dividend; it matters as a signal of how management thinks about the cash.

★ Proving the dividend — six tests
TestEvidenceResult
1. Cover on free cash flowFiscal 2026: $1.57bn of dividends against $6.20bn of free cash flow, 25% (about four times covered).Pass
2. The trend of that coverDividends took 24% of free cash flow in fiscal 2025 and 25% in 2026, and about a sixth of earnings.Stable
3. From operations or from paper?From operations; Applied has net cash and has not borrowed to pay it.Pass
4. Balance-sheet roomNet cash ~$7.6bn including long-term investments; interest covered ~40 times.Pass
5. What would force a cutFree cash flow would have to fall by about three-quarters and stay there. Even in the 2019 downturn Applied's cash flow covered the dividend several times.Very unlikely
6. The growth rate and its direction+25% (2024), +15% (2025), +15% (2026). Steady, and well within the cash.Strong
◆
Part
XI

★ Valuation

What a cycle is worth, and why the price is closer but not yet there

Reported EPS, FY2026
$12.62
27.7×. A record year; still above every pre-2024 annual high multiple (17–23×).
FY2027 consensus EPS ◆
$19.29
18.1×, from about 27.5× in August. Another record year assumed.
Owner earnings per share
~$10.4
Operating cash flow less depreciation as a proxy for upkeep capital spending: ~34×, from ~58× in August.
Free cash flow per share
$7.81
44.7×, a 2.2% yield, depressed by boom-time working capital.
Where the multiple sits in its own historyP/E range within each fiscal year (low to high, on that year's reported EPS).10×20×30×40×50×20171820182120192020201720212220222220231920243020252720265727.7× now18.1× FY27eTrough years in red.★ On last year's earnings the shares are still above every pre-2024 annual high (17–23×).Each year's actual EPS, so a hindsight multiple. Sources: our price history; Applied's 10-Ks via our feed.

The chart is the most important picture in this report. On last year's earnings, Applied still trades above the highest multiple the market paid in any year from 2017 to 2023. At the bottom of the last two cycles, in 2019 and 2023, it traded at 10 to 11 times. On next year's estimate it trades at 18 times, near the top of that old range. The fall has brought the price from extraordinary to merely full.

Our method. As with Texas Instruments last week, we estimate earnings per share in 2030 (Applied's fiscal year), apply an exit multiple, discount 4.1 years at 10%, and add the dividends received on the way (about $8 a share in present value).

CaseFY2030 revenueOperating marginEPS 2030MultipleValue today
The cycle turns$36bn30%$12.3615×~$135
A normal cycle ◆$50bn36%$20.8818×~$265
The company's plan$65bn40%$30.5721×~$445

The cases. "The cycle turns" has 2030 revenue back near this year's level after a downturn, with margins of a normal year and the market paying 15 times. "A normal cycle" lets the boom of 2027 run and then pause, so 2030 revenue is about $50bn, roughly fiscal 2027's expected level, at a 36% operating margin and 18 times. "The company's plan" takes the targets set on 13 October at face value: share gained, a 40% operating margin, and revenue about 9% a year above the 2027 level by 2030, at 21 times.

The middle case moves to about $220 at 15 times and $305 at 21 times; to $232 if 2030 revenue is $44bn and $303 if it is $58bn; to $241 at a 33% margin and $291 at 40%. What is priced in? At $349.23 and 18 times earnings in 2030, Applied must earn about $28 a share that year: roughly analysts' 2028 estimate grown a little further, with no downturn in between. The price is now at the consensus path. In August it was well above it.

Margin of safety. Applied is an excellent business in a violent industry: the "uncertainty present" tier, which asks for a large discount to a central value. We will not insist on a cycle-bottom price for a business this good; that would mean waiting for 10–11 times trough earnings, which may come but cannot be planned for. We would add near $280: about 14.5 times next year's estimate and 22 times this year's earnings, a little above our central value, and 12% below the low of the 5 November crash. If it comes with the boom still intact, begin; if it comes because the boom has ended, wait for the 10-K and the first quarter of 2027 before deciding.

The cycle turns ($135), a normal cycle ($265; we would add near $280), the price ($349), the company's plan ($445) — Closer, not yet
$135
$265
$349
$445
$100$480
★ Closer, not yet — add near $280. The price is about 32% above our central value of ~$265 and below the company's own plan (~$445). Analysts' average target: ~$562. → Interactive valuation on the company page
◆
Part
XII

Risks, lawsuits & controversies

Verified 18 November 2026 against the Q3 FY2026 10-Q, the Q4 release and the Federal Register

Record earnings in a cyclicalAffiliates Rule returns 10 Jan 2027Suspended BIS denial orderTwo customers ~34% of revenueChina below 25% and fallingUS data-centre politicsReceivables +73% in a yearBesi stake moves reported EPSNo named CEO successor

Legal and regulatory. Applied's 10-Q of 20 August 2026 discloses the February 2026 settlement with the Commerce Department's Bureau of Industry and Security ($252.5m) over ion-implanter shipments that reached SMIC. The settlement requires internal audits, training and reporting, and includes a denial order that is suspended and will be waived after three years if the audits are completed; if it were activated, Applied would be barred from exporting certain products outside the United States. The Justice Department and SEC inquiries were previously closed without action. We found no securities class action filed in 2026. The fiscal-2026 10-K, due in mid-December, may update these disclosures.

Export controls. The Affiliates Rule, which extends export restrictions to companies majority-owned by listed Chinese entities, has been suspended until 10 January 2027 (Federal Register 2026-22793). Applied previously estimated its effect at about $600m of annual revenue.

★ How we will know — six dated tests
TestTodayWhat would change our mind
Fiscal-2026 10-K, mid-DecemberQ4 buyback price unknownQ4 buybacks concentrated near the October lows would restore our view of the discipline; customer concentration above 35% would raise the risk.
Affiliates Rule, 10 Jan 2027SuspendedA further extension removes the nearest cliff; expiry would cost about $600m a year and test the 2027 guide.
Q1 FY2027 results, mid-FebruaryGuide $11.1bn ± $0.5bnRevenue at or above $11.1bn with a higher second-quarter guide would support the normal-cycle case; a lower guide would say the peak is near.
Customers' 2027 capex plansRecord budgetsA cut by TSMC, Samsung or a large memory maker would arrive in Applied's orders a quarter later.
Receivables days~79Above 90 would suggest Applied is financing its customers' boom.
The price$349.23Near $280 with the business intact: begin. Near $280 because orders fell: wait one quarter.
◆ PART XIII · To our shareholders
The Letter ⓘ

In August I wrote that Applied Materials was a wonderful business at a price with no cushion, and I declined to tell you what I would pay for it. Three months later the shares are a third cheaper, and I owe you both an account and a number.

The account first. The business did everything I said it might. It reported the best quarter in its history, revenue up fifty-three per cent, and guided the next one higher still. Analysts now expect it to earn more next year than they did in August. Its services business, which I suspected of being flattered by an accounting change, was not; I misread the release and I have corrected it. Its moat, if anything, looks a little wider. And the shares fell thirty-one per cent. Nothing in the company caused it. The market simply decided to pay eighteen times next year's earnings instead of twenty-seven and a half.

That is the most useful kind of fall an owner can be offered, and I want to be careful not to oversell it. The price has come down; the earnings it rests on have not been tested. They are boom earnings, in an industry whose booms have always ended while the customers were still saying demand exceeded supply. On the earnings of the year just ended, Applied still costs more than at any point between 2017 and 2023. At the bottom of the last two cycles it traded at ten or eleven times. So the honest description of today's price is not cheap. It is fair, if the boom runs on.

Two smaller things belong in this letter. I praised management in August for refusing to buy back stock at high prices; the next quarter it spent two billion dollars doing exactly that, at prices well above today's. And I left out a risk I should have reported: the settlement Applied made with the Commerce Department in February carries a suspended order that could, in the worst case, bar some of its exports. Neither changes the quality of the business. Both are things an owner should know, and you should have known them from me.

Now the number. A normal cycle, with 2027's boom and a pause after it, makes Applied worth about two hundred and sixty-five dollars a share today; the company's own plan for 2030, about four hundred and forty-five. I would begin near two hundred and eighty, about fourteen and a half times next year's expected earnings. If the shares get there with the business intact, buy. If they get there because orders have fallen, wait for one more quarter and look again. Either way, write the number down now, while you are calm.

— The Buffett Lens. A lower price for the same earnings is a gift; a lower price for the best earnings is only a discount.

▲The Bull Case
★★ The business beat while the shares fell — Q4 revenue +53% to $10.41bn, adjusted EPS $4.16 vs $4.02 guided; Q1 FY27 guided to a record $11.1bn; FY27 consensus up from $18.41 to $19.29; 2027 guided as 'another record year' with industry spending +30%.
★ A moat that held, and services stronger than we said — gross margin 50.6% through a hiring ramp; AGS +20% like-for-like (our August doubt corrected); machines likely to outgrow the market in calendar 2026; net cash ~$7.6bn.
★ A third cheaper — 18.1× FY27 and 14.2× FY28 estimates, from ~27.5× and ~23× in August; owner-earnings multiple from ~58× to ~34×; dividend raised 15% a year, covered four times.
▼The Bear Case
★★ Still boom earnings — 27.7× fiscal-2026 earnings, above every pre-2024 annual high (17–23×); trough years traded at 10–11×; a 13% revenue fall in 2019 came with no recession.
★ Policy cliffs and a new risk — the Affiliates Rule (~$600m a year) returns on 10 January unless extended; China below 25% in 2027; the BIS settlement carries a suspended denial order we did not report in August.
★ Capital allocation at the top — $2.0bn of buybacks in Q4 at an average market price of ~$479; receivables +73% in a year; two customers ~34% of revenue; no named successor to a CEO in post since 2013.
Closer, Not Yet —
Add Near $280
A wonderful, cyclical business a third cheaper on rising estimates, now priced for the boom continuing rather than beyond it. ★ At $349.23, 18.1× the FY2027 estimate and 27.7× this year's earnings, the price is about 32% above our central value of ~$265, within a range of ~$135 to ~$445. Owners can hold; new money should begin near ~$280. 10-K in mid-December; Affiliates Rule on 10 January; Q1 results in February.
⚡About 14.5 times fiscal-2027 earnings, a little above a normal cycle's value. Add the $280 price trigger to your Watchlist.
◆ The Buffett Lens · Dividend Line Research · As of 18 Nov 2026 · Price $349.23 (17 Nov close) · Follow-up to our 15 Aug 2026 X-Ray
Found this useful? Share the X-Ray.
◆ More on Applied Materials, Inc.

Articles that take this apart

◆ Keep reading

More X-Rays from the same lens

Disclaimer: This is an editorial analysis for information and education, not investment advice, and not a recommendation to buy or sell any security. ⚠️ Price and market data are from our live data pull of 17 November 2026; the price used is the 17 November close. Results, guidance, segment data, the balance sheet, buybacks, the dividend and legal matters are from Applied Materials' releases of 13 August and 12 November 2026, its Form 10-Q for the quarter to 26 July 2026 and its 8-Ks; the Affiliates Rule status from the Federal Register; management quotations from call transcripts (third-party copies); consensus estimates from Yahoo Finance and S&P Global data as reported by stockanalysis.com; peers' figures from their releases or third-party summaries, as marked. Valuation cases are our own estimates and depend on assumptions about the semiconductor cycle that may prove wrong. Past performance does not guarantee future results. Do your own research or consult a licensed adviser.
Dividend Line · X-Ray Analyses — written in the house methodology