Phase Space AI

Trade Construction

Repligen Corporation [RGEN]

Repligen Corporation [RGEN] — Trade Construction & Risk

As of 2026-07-29 · spot $145.00 · framework v1.5.1

This document issues no position verdict. It states what a position would have to look like if the book chose to take one, and what would kill it. The book decides.


1. Liquidity Criteria — the vehicle question, chain pulled first

criteria.md is explicit: "Any proposed options structure requires the actual chain pulled first — open interest and quoted size for the specific strikes and expiry."

Chain pulled 2026-07-29 via Alpaca (/v2/options/contracts for discovery, /v1beta1/options/snapshots for quotes and Greeks). Only one listed expiry exists in the Dec-2026 → Jan-2027 window: 2026-12-18.

Strike Bid Ask Spread as % of mid IV Delta Open interest
130 26.15 29.64 12.5% 0.547 0.701
135 22.45 26.87 17.9% 0.532 0.662
140 20.11 23.89 17.2% 0.528 0.621 15
145 17.45 21.46 20.6% 0.521 0.579 2
150 15.14 19.01 22.6% 0.514 0.537
155 13.63 16.90 21.4% 0.517 0.497
160 11.52 15.18 27.4% 0.512 0.456
165 11.10 12.51 12.0% 0.512 0.418 1
175 7.12 10.99 42.7% 0.509 0.346 12
180 5.50 9.54 53.7% 0.496 0.307
190 4.07 7.66 61.2% 0.501 0.252 1
200 3.14 6.19 65.4% 0.510 0.208 1

Total open interest across the entire Dec-2026 call chain: 37 contracts, maximum 15 at any single strike.

No options structure in RGEN is investable at any size. A 15-contract maximum open interest is a $22k notional at the 140 strike. Crossing a 20–65%-of-mid spread destroys the entire edge before the thesis is tested. This is the HCA failure mode recorded in criteria.md, reproduced exactly: "A vehicle that cannot be filled is not a vehicle."

Consequence: any RGEN expression is common stock only. No calls, no spreads, no collar, no covered write. This is not a preference; it is what the chain permits.

Equity liquidity is adequate. Nasdaq Global Select, $8.18bn market cap, 56.4m shares. IEX-feed 60-day median volume is 93,570 shares = $13.6m/day, and IEX is a small fraction of the consolidated tape, so that is a floor, not an estimate. A position of any size this book would take is exitable in a day or two.


2. If a position were taken — what it would have to be

Recorded for completeness, conditional on the book overriding a Valuation-Criteria FAIL. criteria.md: "Cheap cannot rescue a failure" on Quality, and the Valuation Criteria is BINDING for the long-only absolute-return strategy. Both are FAIL here.

Vehicle Common stock only. Forced by §1.
Direction Long, if at all — the Short Mechanism Criteria does not qualify (growth accelerating, margin runway not exhausted).
Thesis horizon 12 months, not 5 years. The multiple mean-reversion is the trade; the compounding case fails.
Entry The Q2 print already moved the stock +11.0% (2026-07-28). Entry above ~$150 gives up most of the multiple-percentile edge.
Sizing Inverse-volatility. 252-day realised vol 46.5% — roughly 2.5x a typical large-cap, so a half-to-third-size slot at best. This is the active protection while Downside Criteria remains MEASURED.
Target $163 pro forma / $176 standalone (12 months). See RGEN_Valuation.md §2.
Invalidation §4 below.

Momentum Criteria (MEASURED, entry timing only): +9.0% 12-1. Mid-pack, not a timing signal in either direction. The name is 15.8% below its 52-week high and 55.3% below its 2021 high of $324.24; 6-month return is −9.4% while 3-month is +28.2% — a name that broke and has been re-rating off the low for one quarter.


3. Downside Criteria — the permanent-loss case, with a named cause

Type: MEASURED. Logged and scored. It does not reject the name.

3.1 The named cause

The 2025–26 bioprocessing "recovery" is a restocking cycle, not a demand step-up — and it cannot be falsified from the filings, because the order disclosure that would have shown it was retired.

The chain of evidence:

  1. Repligen's customers — biopharma manufacturers and CDMOs — over-ordered through 2021–22 (revenue +83% then +20%), then cut −21.1% in FY2023 with a Q3-2023 trough of $141.2m.
  2. Book-to-bill was disclosed for exactly five quarters, Q3-2023 to Q3-2024, at 1.07, 1.03, 0.99, 1.02, 1.04 — barely above parity throughout — and has not been disclosed since. Backlog has never been disclosed, in 15 consecutive earnings releases or the FY2025 10-K.
  3. Since Q4-2024 the only order commentary is unquantified: "continued momentum in orders", "recent order trends", "order momentum ... continued". There is no way to distinguish restocking from demand from outside the company.
  4. The FY2025 10-K notes commissions are "driven by customer purchase orders which are short-term in nature" — there is no contracted revenue floor.
  5. Tariff mentions appear in every release from Q1-2025 onward (3, 2, 1, 1, 1, 1) — pull-forward ahead of tariffs is a known mechanism for exactly this pattern in supplied consumables.
  6. The control environment cannot corroborate it. Repligen restated FY2023 and five quarterly periods after the PCAOB questioned a $17.3m COVID cancellation payment booked as Q1-2023 product revenue, and management concluded that internal control over financial reporting was NOT effective at 2025-12-31, with inventory-valuation and financial-close controls unremediated. The 290bp of gross-margin expansion that underwrites the margin leg of the bull case runs through exactly those controls.

If it is restocking, growth reverts towards end-market volume growth — call it 5–7% — once safety stocks are rebuilt, and it does so without warning because there is no order metric to telegraph it.

3.2 The impairment arithmetic

Compounding causes, all named and all already contracted or announced:

Pro-forma revenue ~$900m × 7.33x = $6,597m EV; less ~$300m net debt = $6,297m equity ÷ 63.0m shares = $100/share.

Bear case: $100, −31.0% from $145.00. Probability 27%.

Logged to trade_recommendations.jsonl for Brier scoring per criteria.md.

This is not a going-concern case and is explicitly not argued as one. $810m of cash and securities, a 2028 maturity, positive operating cash flow ($61.1m in H1 2026), no covenant structure on a 1.00% unsecured convertible. The impairment is a valuation and capital-allocation impairment, not a solvency one.

3.3 What would make the bear case wrong


4. Invalidation triggers — dated and checkable

# Trigger Where it would appear Action
1 Organic growth prints below 9% in any quarter quarterly 8-K exhibit 99.1, organic reconciliation table Thesis broken — the mean-reversion trade rests on 12–13% organic
2 FY2026 revenue guidance cut below $813m Q3 2026 release Exit
3 BioLife deal breaks (regulatory block or stockholder vote fails) 8-K Re-underwrite: standalone target rises to $176 but the growth mix worsens
4 Purchase price raised or consideration shifted materially to cash 8-K / S-4 amendment Re-underwrite the dilution and the balance sheet
5 Adjusted-to-GAAP gap widens further — new add-back category beyond "transformation costs" reconciliation tables Accounting-quality escalation
5b ICFR still not effective at 2026-12-31, or a new material weakness, or any further restatement FY2026 10-K Item 9A (late Feb 2027) Accounting-quality escalation. A third consecutive year of ineffective ICFR while integrating a $1.5bn acquisition would be disqualifying on its own
6 Multiple exceeds 13.10x EV/Sales (post-destocking p75) daily Target reached; the mean-reversion trade is complete
7 Converts called on/after 2026-12-18 without a refinancing plan 8-K Liquidity re-examination against the $540m BioLife cash outflow

5. Summary for the book

Quality Criteria (BINDING) FAIL — ICFR not effective at 2025-12-31 (inventory-valuation and IT-general-control weaknesses unremediated, after a 2024 revenue restatement); op-margin change +4.1pp organic-adjusted vs ~5pp separation; ROIC 2.7% vs WACC ~10%
Valuation Criteria (BINDING) FAIL — required 27.4% vs demonstrated 3.3% / guided 12%; margin −24.1pp / −15.4pp
Liquidity Criteria (BINDING) equity PASS · options FAIL (max chain OI 15 contracts)
Downside Criteria (MEASURED) $100, −31%, p=27%, named cause: restocking mistaken for demand, unverifiable because the order disclosure was retired
Momentum Criteria (MEASURED) +9.0% 12-1
Consensus Criteria (MEASURED) INDETERMINATE — quota exhausted, blocks nothing
12-month target $163 pro forma (+12.3%) / $176 standalone (+21.1%)
Vehicle Common stock only
Sizing input 46.5% realised vol → inverse-vol slot, ≤⅓ size

The honest summary of this name: a genuinely improving, genuinely organic 12–13% grower whose multiple sits at the 14.5th percentile of its own history — and which is priced for 27% compounding it has never demonstrated, does not guide to, and has no named mechanism to reach. The screen said PASS by +17.6pp. On corrected inputs it is FAIL by 15–24pp. The 12-month setup is nevertheless the most attractive part of this name and is the reason the memo is not a dismissal.

The one thing that would change the analysis fastest is not a valuation input. It is Repligen reinstating a quantified order metric and reporting effective internal control at 2026-12-31. Those two disclosures would convert the memo's two largest unverifiables — is the recovery demand or restocking, and can the reported margin expansion be relied upon — into evidence. Neither is a forecast; both are scheduled disclosures.