GeneDx Holdings Corp. [WGS] — Valuation Analysis (Task 3)

Spot $60.54 (2026-07-27) · Shares o/s 29,666,318 (10-Q, 2026-03-31) · Diluted FY2027E 30.5m Market cap $1,846m · Net cash $71.7m · Enterprise value $1,775m Model WGS_Model.xlsx — every output below was re-derived by independent formula evaluation of the workbook, not read back from the string that wrote it. Two real formula bugs were found and fixed by that check (an H1-2026 aggregation pointing at the FY2024 columns, and an implied-EV formula multiplying the revenue-growth row instead of the revenue row). Recorded here because the check is the reason they were caught.

1. Valuation summary

Headline: on the house base case the stock is worth approximately what it trades at. It is not cheap and it is not obviously expensive. The valuation does not carry a directional decision in either direction — which is the correct and expected outcome, and is why Task 5 turns on the gates rather than on this section.

2. Scenario construction

Built off volume × ASP, not a blended revenue growth rate — because the entire 2026 event was a decoupling of those two, and any model that carries one growth rate cannot represent it.

2.1 Drivers

The FY2026 base of $466.8m sits below the $475m guidance floor. That is deliberate and is the variant: it is what the year produces if the unexplained +5.7% H2 ASP step-up embedded in the reset guide does not happen.

2.2 Base-rate check on every scenario's growth assumption (required)

Anchoring on the reference class first (Chan, Karceski & Lakonishok 2003 — growth persistence beyond chance is close to nonexistent; the modal modelling error is extrapolating the recent rate):

• Base: FY2026–30 revenue CAGR 17.3% off a $427m base. For a $400m-revenue healthcare-services company, sustained mid-to-high-teens growth over five years is roughly a top-third outcome — demanding but not extraordinary. Critically, the Base case does not extrapolate FY2025's +41%; it assumes decay from 30% to 13%, which is the reference-class prior rather than an override of it.
• Bull: FY2026–30 revenue CAGR 27.0% — an explicitly top-decile outcome, named as such. It requires both 25%+ volume compounding and an ASP recovery to $3,870. §6.3 of the research note shows those two requirements are in tension: the marginal patient added at higher penetration is by construction in a worse-reimbursed channel. The prior being overridden is named; the override is not defended as likely, which is why the weight is 25% and not higher.
• Bear: FY2026–30 revenue CAGR 12.7% — approximately the median five-year outcome for a company at this revenue scale. This is not a disaster case. It is the base rate, with the price problem simply not fixed.

3. Comparable companies

Market cap ÷ TTM revenue is used consistently across all names including WGS, because enterprise value is not uniformly obtainable for the peer set from a free source. WGS's own EV/Sales is stated separately below.

WGS EV/Sales: 4.01x TTM · 3.71x on FY2026 consensus ($478.4m) · 2.98x on FY2027 consensus ($594.9m) · 3.16x on the house FY2027 base.

What the comps do and do not say. WGS trades at a ~22% discount to the peer median on P/Sales while growing volume faster than any of them. On that reading it is cheap. But the peer set is not like-for-like — NTRA and GH are oncology/MRD franchises with rising realised price and a different payer dynamic; MYGN, NEO and FLGT are the low-multiple end precisely because their realised economics broke. WGS's multiple has moved from the NTRA/GH end of the distribution toward the CDNA/NEO end over nine months, and the reason it moved is the same reason the low-multiple names are cheap. The comps therefore support a fair-value band of roughly $58–78 (3.0x–4.0x FY2027E) rather than a target, and the honest statement is that the peer set brackets the answer without deciding it.

Sector-regime check (required for a rich/cheap claim): the diagnostics peer set is up a median +101% over twelve months while WGS is −24%. This is emphatically not a sector de-rating. It is stock-specific, which raises rather than lowers the evidentiary bar on any claim that the market is simply wrong.

4. DCF

4.1 Cost of capital

4.2 Base-case free cash flow (10-year explicit, SBC charged as a real cost, no cash tax to FY2029 on NOLs)

4.3 Output — and why it is reported but not used

The spread between $13 and $46 on the same cash-flow stream is the finding. Essentially 100% of this equity's value is terminal — the explicit-period PV is negative for the first three years — so the DCF cannot discriminate and is not used as the price target. This is the same archetype already logged in CALIBRATION_WATCH.md for TXG/TWST/NTRA/GH (pre-profit life-science names whose DCFs return terminal-dominated or negative values). Handling it by disclosure rather than by quietly dropping it is the required treatment.

The one genuinely useful DCF output is the inversion. Solving for the discount rate at which the base-case stream equals today's $1,775m EV:

At $60.54 the market is either discounting the base case at a 7.3% cost of capital — implausible for a 96%-realised-vol, loss-making, single-product small cap — or it is discounting the bull case at a normal risk-adjusted 15.0%. In other words, the stock is priced for the bull case, correctly discounted. That is a valuation-negative observation. Per references/trade-construction.md it is also, on its own, not a thesis: a stock can be priced for the bull case and still go up.

5. Street consensus, positioning and the required bridge

5.1 Consensus estimates (Alpha Vantage EARNINGS_ESTIMATES, one call, cached 2026-07-27)

FY2026 consensus revenue ($478.4m) is essentially the guidance midpoint ($482.5m). Consensus is guidance-anchored, not independent. FY2026 EPS has been cut −107% in 90 days and FY2027 EPS −46%.

5.2 Consensus rating and price targets

5.3 The bridge — numbers vs. multiple, decomposed rather than asserted

House base target $65.06 vs Street $81.78. Gap = $16.72, or 20.4%.

74% of the disagreement with the Street is about the multiple; only 26% is about the numbers. This is stated plainly because it is the weakest part of the negative view, not the strongest, and the framework requires it be proven rather than assumed.

The Street's own behaviour is the sharpest observation available. Between 23 June and 23 July 2026, four analysts raised price targets — TD Cowen by 55% ($55 → $85) — while over the identical window the Alpha Vantage revision fields show FY2027 EPS falling from $1.2313 to $1.1038 and five of nine analysts cutting. Numbers down, targets up. That is not underwriting; that is marking the multiple to a recovering share price — the same pattern already logged in CALIBRATION_WATCH.md for the TXG and TWST target raises.

And the real risk to the house view, stated: the Street's target-raising may be correctly anticipating that Q2 confirms an ASP inflection under the new commercial leadership, in which case the multiple should expand and the house base case is simply too pessimistic on a single quarter's data. I have one quarter of ASP break and three of DSO deterioration. That is a pattern, not a proof.

5.4 Positioning, short interest and the options market

Options market read (Alpaca snapshots, 2026-07-27):

Front-week IV is ~2.1x 30-day realised. Bid/ask on the Aug-7 $60 put is $4.38 / $6.81 — a 41% spread. Historical print-day moves: −49.2% (5 May 2026), −7.3% (24 Feb 2026), −4.2% (29 Oct 2025). A 19.9% implied move is not obviously mispriced against that distribution, and the variance risk premium plus a 41% spread makes long premium negative-expectancy here. Load-bearing for Task 5 Gate 5.

6. Transcript / prepared-remarks Mention-Frequency table (REQUIRED)

Source: SEC 8-K Item 2.02 earnings press releases, one source across the entire series, never mixed. Window: 17 quarters, 2022Q1 – 2026Q1. Normalised per 10,000 words (word counts 2,755–4,412). LIMITATION: press releases carry no prepared-remarks vs. Q&A split — that column cannot be populated and is not faked. Full 17-quarter grid is in WGS_Research.md §1.2.

7. Factor & Anomaly Scorecard (REQUIRED)

Read the two right-hand columns together — they disagree, and the disagreement is the point. Because of the SPAC/wind-down history the brief flagged, each row states explicitly which entity it is measuring.

7.1 Synthesis — do the factors support, contradict, or split on the fundamental view?

They split, and they split along an interpretable line.

• Every factor measuring the CURRENT ongoing business corroborates the negative fundamental view: momentum, 52-week-high, trend, SUE, estimate revisions, asset growth, TTM F-score (2/9), and DSO.
• Every factor that contradicts it is measuring something other than the ongoing business: gross profitability (measuring a transaction-inflated asset base), accruals (measuring an impairment), and the FY2025 F-score (measuring a year that ended before the break).
• The one factor that contradicts and is NOT distorted is short interest, and it contradicts the trade, not the analysis — 30.79% of float says the negative view is already widely held and expensively expressed.

The honest statement demanded by Gate 1's quantitative-corroboration standard: the two named corroboration inputs — accruals and F-score — do not cleanly support a short on a naïve read. Accruals look clean and the last full-year F-score is 6/9. But in this specific name both are restructuring/impairment artifacts, and I state the disagreement rather than ignoring it. The F-score recomputed on the trailing four quarters collapses from 6/9 to 2/9, and the receivables trend — the one metric untouched by the discontinuity, the write-off, or purchase accounting — has deteriorated for three consecutive quarters. On the evidence that is actually measuring the business as it operates today, the scorecard corroborates the mechanism.

7.2 Would this name have survived the screen? (base-rate information)

No — and it would have failed on both sides. WGS is a user-named ticker, not a screener output. Run against the idea-screener funnel: the momentum/trend stage would have rejected it as a long (−24% 12m, 0.66× the 200-day, 0.36× the 52-week high), while the estimate-revision overlay would have flagged it as a short candidate — and the EDGAR quality overlay would then have passed it through on clean FY2025 accruals and a 6/9 F-score. The screen would have produced a contradictory read. That is base-rate information and it is consistent with the split scorecard above: this is a name in transition where the standardised signals disagree with each other, which is exactly the regime in which single-name factor readings are least reliable.

8. The four conclusions, kept separate (input to Task 5)

1. Fundamental. Volume demand is intact and accelerating (+34%). Realised price is deteriorating (−14.3% over two quarters, company-disclosed at −5% YoY), receivables are aging (DSO 36→68), and operating leverage has inverted (opex +48–57% vs revenue +17%). TTM operating cash flow is negative. A $31.3m acquisition write-off 11 months post-close names "go-to-market execution" as the cause. Deteriorating, on a business whose underlying demand is fine.
2. Expectations. Consensus is guidance-anchored at FY2026 $478.4m / FY2027 $594.9m and has already been cut hard (FY2026 EPS −107%, FY2027 −46% in 90 days). House base is −2.4% on FY2026 revenue, −5.4% on FY2027 revenue, and −54% on FY2027 adjusted EPS. The house FY2026 number sits below the guidance floor. A real but modest gap on revenue; a large gap on earnings; and 74% of the target-price disagreement is multiple, not numbers.
3. Valuation. Probability-weighted target $66.71 (+10.2%). Base $65.06 (+7.5%). The market is pricing the bull case discounted at a normal 15% WACC, or the base case at an implausible 7.3%. Full to fair; not a valuation short, and not a valuation long.
4. Portfolio. Deferred to Task 5. The binding facts are 30.79% of float short on an 18.94m float, ~$106.6m of clustered insider buying under the price, front-week IV at 2.1× realised with 41% option spreads, and a catalyst seven days away.