Semiconductor equipment (WFE) & automated test (ATE)

Claim

The most intensively worked field of July 2026 and the one that produced the cleanest negative result: a large drawdown with no value in it.

ATE: 11 of 11 names above their own historical bands. The extreme case was TER at 76% above its own FY2025 high multiple — while screening as a large drawdown. This field is the origin of [[principle-down-a-lot-is-not-cheap]].

Drawdowns (SOX peak close 6/22 → 7/29)

Name	Drawdown
ASML	−19.5%
^SOX	−28.6%
AMAT	−31.8%
KLAC	−36.8%
LRCX	−38.4%

ASML held up best despite being the most expensive equipment name on most measures — another point against reading drawdown as valuation.

AMAT — the field's most important finding

Full analysis: Output/Stocks/Semiconductors/AMAT/analyze-2026-07-29.md (WATCH, conviction 6.5, FV $290–370, entry $300–350, gated on the Aug 13 print).

The durable, transferable part: "record" GAAP EPS was ~26% non-operating. Operating income was flat ($8,289M FY25 → $8,299M TTM); the entire increase came from marks on AMAT's ~9% BESI stake plus a tax rate falling 24.5% → 18.1%. Two agents converged independently — ROIC fell 44% → ~33% on invested capital +72% against operating income +6.6%, while AMAT lost served-market share in 2025 as LRCX/KLAC/ASML gained. Spending more to stand still.

Method notes this produced: [[pitfall-yahoo-misclassifies-strategic-investment-gains]] · [[pitfall-roicai-fcf-field-returns-ocf]] · [[principle-primary-source-beats-vendor]].

Standing order

KLAC before AMAT — KLAC is in its entry zone, AMAT is not.

What would falsify this

An August print cycle that resets the earnings base downward far enough to bring these names below their own bands — the condition that turns the whole field from avoid to shopping list. recheck is set to the day after AMAT reports.

[[semi-materials]] (the adjacent layer, also swept, also no gems) · [[pattern-ai-levered-fields-trade-above-own-band]] · [[regime]] Act II

History