Phase Space AI

Financial Model Notes

Lam Research [LRCX]

Lam Research [LRCX] — Financial Model Notes & Data-Integrity Log

As of 2026-07-29 · framework v1.5.1

1. Source hierarchy actually used

Quantity Source used Why
Revenue, gross profit, operating income, net income AV INCOME_STATEMENT normalised, FY2026 (Q ending 2025-09-30 .. 2026-06-30) Brief directs AV normalised statements over the XBRL parsing surface
Share count AV OVERVIEW SharesOutstanding = 1,250.571m, cross-checked against AV BALANCE_SHEET commonStockSharesOutstanding = 1,261.102m See §2 — the screen's figure was wrong
Net cash AV BALANCE_SHEET: cash+ST investments $5,579.2m − shortLongTermDebtTotal $3,730.5m = $1,848.7m Screen figure wrong; see §2
Price Alpaca stocks/snapshots latest trade, 2026-07-29
D&A cross-checked; see §3 AV disagrees with itself
Segment / service mix EDGAR 10-K R-files Dimensional XBRL; absent from AV entirely
Geographic / China share EDGAR 10-K R-files Dimensional XBRL; absent from AV entirely
Deferred revenue / RPO EDGAR companyfacts AV's deferredRevenue field is present-but-NULL; see §3
Consensus / revisions AV EARNINGS_ESTIMATES populated, with 7/30/60/90-day revision history

2. Screen-input validation

LRCX was not scored by the screen (reports/scan_v2/ contains no LRCX record), so there are no screen inputs to validate for this name. All figures here are derived from AV normalised statements plus EDGAR, as tabled in §1 and §4.

One cluster-level warning inherited from the KLAC record, which does apply to LRCX. LRCX executed a 10-for-1 split effective 2024-10-03. The KLAC record was corrupted by pairing a pre-split share count with a post-split price; the same trap exists here and was explicitly checked. AV OVERVIEW (1,250.571m shares), AV BALANCE_SHEET (1,261.102m issued) and Alpaca's adjusted price series are all on the post-split basis and are mutually consistent: 1,250.571m × $269.51 = $337,076m, against AV's own MarketCapitalization of $337,166m — a 0.03% difference, which is a one-day price lag, not a basis error. Verified rather than assumed.

Independent scale check. TTM net income $7,265.5m ÷ 1,250.571m shares = $5.81 implied EPS. AV OVERVIEW reports EPS $4.88 — see §3, this is a staleness defect in OVERVIEW, not a share-count error; the statements themselves are internally consistent.

3. Alpha Vantage defects found on this name

Defect 1 — OVERVIEW is stale, and its P/E is therefore wrong. OVERVIEW.EPS = $4.88, PERatio = 55.25. But TTM net income $7,265.5m ÷ 1,250.571m shares = $5.81. OVERVIEW excludes the June-2026 quarter that INCOME_STATEMENT — same vendor, same key, same minute — already carries. Real TTM P/E at $269.51 is 46.4x, not 55.25x: a 19.1% overstatement. Note that AV's PERatio is internally consistent with its own stale EPS ($269.61 ÷ $4.88 = 55.25), which is exactly the brief's point that "a vendor that is wrong consistently defeats every consistency test" — the only check with power came from recomputing off the statements.

Defect 2 — deferredRevenue is present-but-NULL in all 81 quarterly balance-sheet records, as on the other two names. Calibration item D1. The contract-liability and RPO series used in this memo came from EDGAR companyfacts and the 10-K R-files instead. This matters more here than anywhere: LRCX's deferred-revenue trend is load-bearing for the pull-forward question, and AV silently has none of it.

Defect 3 — longTermDebt is 0.0 in the most recent quarter only. At 2026-06-30 longTermDebt = 0.0 while shortLongTermDebtTotal = $3,730.5m; the five prior quarters carry $3,720–3,730m in longTermDebt. Reading longTermDebt would have produced net cash of +$5,579m instead of +$1,849m — a $3,730m error in enterprise value. shortLongTermDebtTotal was used.

Defect 4 — sellingGeneralAndAdministrative is 0.0 for FY2026 (5.33% of revenue in FY2025). Same present-but-wrong pattern as KLAC; the FY2025 bridge was used to validate the terminal margin.

Defect 5 (tooling, not vendor) — av_vs_edgar.py performed ZERO comparisons on LRCX and reported no disagreements. Output: LRCX 0 cmp 0 dis -- 0 inv. A clean-looking result that is actually the absence of any test. Cause: AV normalises LRCX's 52/53-week fiscal quarter-ends to calendar month-ends (2026-03-31) while EDGAR carries the true period end (2026-03-29), and the audit joins on date. The same applies to AMAT (AV 2026-04-30 vs EDGAR 2026-04-26). KLAC got 4 comparisons only because KLA's quarters happen to fall on calendar month-ends. Every 52/53-week filer therefore silently receives no cross-source verification from this script, which is the single most consequential defect found in this run because it disables the brief's only check with power. Fix: join on nearest period-end within ±7 days.

4. Derived figures, shown so they can be checked

TTM revenue $23,232.7m
TTM gross profit / margin $11,725.3m / 50.5%
TTM operating income / margin $8,199.8m / 35.3%
TTM net income $7,265.5m
TTM D&A (computed, absolute value, from cash flow) $441.5m
TTM EBITDA (computed = operating income + |D&A|) $8,641.3m
Market cap = 1,250.571m × $269.51 $337,076m
EV = market cap − net cash $335,227m
EV/EBIT 40.9x
EV/EBITDA 38.8x
EV/Sales 14.43x
TTM diluted EPS ≈ net income / shares $5.81
TTM P/E 43.4x (96th percentile of own 10y)

AV's ebitda field was NOT used, per the brief. It is wrong on this cluster: the field is derived from the same signed D&A that fails on KLAC and AMAT.

AV's ebit field was also NOT used — it is not operating income. On this name the ebit field also diverges from operatingIncome (it includes non-operating items); operatingIncome is used throughout.

5. Reproduction

Raw pulls are in work/wfe/ at the repository root: av_LRCX_INCOME_STATEMENT.json, av_LRCX_BALANCE_SHEET.json, av_LRCX_CASH_FLOW.json, av_LRCX_OVERVIEW.json, av_LRCX_SPLITS.json, est_LRCX.json, ef_LRCX.json (EDGAR companyfacts), LRCX_R*.htm (10-K R-files), snap_LRCX.json.