BillionToOne, Inc. [BLLN] — Valuation Analysis (Task 3)

⚠️ SUPERSEDED IN PART — 2026-07-29

The position verdict in this document is retired. Under the current framework (references/criteria.md, 2026-07-29) the memo outputs an analysis, not a position. Whether an analysis justifies a position is a question about a particular book, and two books answer it differently.

The Gate block and the Gate 4 expected-return arithmetic below are also superseded, by the named Criteria (each with a type: BINDING or MEASURED, returning PASS / FAIL / INDETERMINATE), the reverse-DCF implied-path test, and a separate 12-month target.

→ Current analysis: BLLN_Criteria_and_Valuation_2026-07-29.md

Everything else here — the research, the evidence, the mechanism work — stands. Residual references to "Watchlist" in the prose below are the historical record of the 2026-07-28 assessment and are left intact deliberately.

Spot $133.64 (2026-07-28 close, Alpaca SIP). All model outputs below were read back from the live Excel workbook after recalculation in Microsoft Excel, not from Python.

1. Capitalisation

2. Comparable companies

Prices Alpaca 2026-07-28; TTM revenue and cash from EDGAR XBRL, each quarter grouped by the fact's own end date. NTRA's cash tag returned a stale 2022 date and is immaterial at a $36bn cap; flagged rather than silently used.

BLLN is the most expensive name in the set on EV/TTM sales — and also the fastest-growing by a wide margin. On a growth-adjusted basis it sits below the peer regression line (chart 15), and on EV/FY2027E sales (10.8x) it is cheaper than NTRA (~11.9x) and GH (~14.4x) while growing roughly three times as fast with a materially better margin structure. The comps do not support a "too expensive" conclusion. This is stated plainly because it cuts against where the DCF lands, and per references/trade-construction.md a rich multiple in a re-rating peer group is a sector-regime question, not a stock-specific thesis.

3. DCF

Discount rate — and an important correction made during the build. The measured beta from 180 post-IPO daily sessions is 1.75. That figure was rejected. Decomposing it: β = ρ × (σ_BLLN/σ_SPY) = 0.272 × (85.3%/13.3%) = 1.74 — i.e. the entire "beta" is BLLN's own 85% realised volatility scaled by a correlation to the market of only 0.27. That is idiosyncratic risk, which CAPM does not compensate, measured over a sample far too short to be reliable. The house uses a peer-informed β of 1.35 (an explicit assumption), giving:

• Cost of equity = 4.7% + 1.35 × 5.0% = 11.45%
• WACC = 11.45% (a $90m term loan against a $7.1bn cap is immaterial)
• Terminal growth 3.5%; terminal exit multiple 4.5x sales

Base case (Excel DCF tab, recalculated):

Sensitivity (chart 11): spot $133.64 is reached only at roughly a 6.0x terminal sales multiple combined with a sub-10% WACC. Across the whole 5×5 grid of plausible WACC and terminal-multiple pairs, no combination in the central region produces a value at or above spot.

Scenario DCF values: bear $27.75–43.40, base $53.35–79.90, bull $80.41–128.23 (lower figure Gordon, upper exit-multiple). Even the bull case at a generous 5.5x exit multiple lands ~4% below spot.

4. Street and positioning

The stock trades 8.6% above the average Street target and above all but the very top of the range. Consensus FY2026 revenue sits essentially at the low-to-middle of company guidance.

Guidance history — a genuine positive:

FY2025 guidance of $293–299m (initiated Q3-25) produced an actual of $305.1m — a beat above the high end. This management team has beaten or raised at every opportunity it has had as a public company. That is only three data points, and it is stated as such.

Numbers-vs-multiple decomposition (required)

The house Base case FY2026 revenue is $462m vs consensus $454m — a variant of only +1.8%. On FY2027 the house is $615m vs an estimated ~$590m consensus, +4.2%. Therefore essentially none of the gap between the house valuation and the Street's $122 target is about the numbers — it is almost entirely about the multiple and the discount rate. I am above consensus on revenue and still below the Street on value.

That is exactly the pattern logged as calibration item B8, and it is named here rather than buried.

5. Reverse DCF and the required implied-penetration statement

Solving for the revenue path that justifies $133.64 (WACC 11.45%, terminal margin 27%, 4.5x exit):

Required: FY2026 $462m growing at an initial 37.4% and fading 10% per year — reaching ~$3,515m by FY2035, or 11.5x FY2025 revenue.

Translated into units, per references/tam-sizing.md:

At a mature blended ASP of ~$600, $3.5bn implies roughly 5.9m billable tests per year against 610,000 delivered in FY2025 — a 9.6x volume increase. The bottom-up US prenatal SAM tops out near $735–850m even at a generous 25% share of an 80%-penetrated market. The current price therefore requires oncology to become roughly three-quarters of BillionToOne — from $42.9m of annualised revenue, one peer-reviewed publication, no registered MRD trial, and no Medicare coverage for its monitoring assay, against Signatera's 151 publications.

That single sentence is the valuation conclusion. See charts 19 and 25.

6. Factor & Anomaly Scorecard (required)

Synthesis: six of eight computable factors are tailwinds for the long side, and none of them supports a short. Earnings quality is clean on the standard quantitative measures — negative accruals, top-decile gross profitability, a high F-score. A short thesis on this name would have to be argued directly against its own scorecard, which per references/trade-construction.md requires correspondingly stronger narrative evidence. That evidence does not exist here. See chart 16.

The two caveats the scorecard cannot see — unremediated ICFR material weaknesses and the accelerating true-up line — are exactly the blind spot logged as calibration item S5 (the scorecard interrogates historical accounts, not disclosure quality). They are real, and they are not large enough to overturn the quantitative read.