Dislocation

Penumbra fell 25.8% from a $303.76 peak on 19 Feb 2025 to a $225.54 trough on 5 Nov 2025, but that decline was a slow multiple de-rating with no dated adverse trigger — earnings beat consensus every quarter through it — and peak-leg volume rose only 2.07x its pre-peak median, orderly rather than panicked. The drawdown fully round-tripped by December 2025. On 15 Jan 2026 Boston Scientific agreed to acquire the company at $374 a share. At $321.19 today the stock is a merger-arbitrage position, not a fear-repriced compounder.

The drawdown, quantified

Source: peak, trough and depth from fit_features.capitulation_gauge.drawdown (daily price history); deal price from the 15 Jan 2026 merger announcement [1].

The capitulation_gauge records a 25.8% fall over 259 days, from a $303.76 close on 19 February 2025 to a $225.54 close on 5 November 2025. Two features of that window matter more than its depth. First, the peak was itself an earnings-driven spike: the stock closed up 12.0% on 19 February 2025, the session after the Q4 2024 release, so the reference point is a high-water mark, not a settled level. Second, the trough on 5 November 2025 coincided with the Q3 2025 release and reversed immediately — the stock closed up 17.3% the next day and finished November at $293.17. The low was a single-session touch, not a base.

Source: daily price history, data/prices/daily.json; deal price ($374) from the 15 Jan 2026 merger announcement [2].

The fall came in one long leg, not a cliff. From the February peak the stock stepped down through 2025 — $292.84 at end-April, $256.63 at end-June, $227.37 at end-October — punctuated by earnings pops that faded within weeks. It then recovered the entire decline: by 31 December 2025 the close was $310.91, above the February peak. The current $321.19 sits above that peak and 71% up the trailing 52-week range (low $224.84, high $360.00). Measured against where the stock trades now, the drawdown the feature file records is a closed historical episode.

The trigger

There is no dated adverse event behind the 2025 decline. The stock did not gap down on a guidance cut, an earnings miss, a recall, or a regulatory action. Every quarter reported during the fall beat consensus — Q4 2024 EPS of $0.97 against $0.89 (+8.5%), Q1 2025 of $0.83 against $0.67 (+24.5%), and Q2 2025 of $0.86 against $0.82 (+4.7%) — and the shares rose 12.0%, 7.0% and 11.9% respectively on the sessions after those releases. A price that falls while earnings beat and estimates hold is multiple compression, not a fear repricing; the de-rating carried the premium down without a shock to point to. The one contextual drag was China, where the preliminary FY2025 update put reported growth at 17.3–17.5% against 24.7–24.9% excluding the region — a headwind to the reported line, not a dated selling event [3].

The only identifiable, dated event that repriced the stock recently is the opposite of an adverse one. On 15 January 2026 Boston Scientific and Penumbra announced a definitive agreement for Boston Scientific to acquire the company at $374 a share, an enterprise value of roughly $14.5 billion [4]. Each share carries the right to elect $374 in cash or 3.8721 Boston Scientific shares, subject to proration to roughly 73% cash and 27% stock [5]. That is the dated event that moved the tape: the stock closed up 11.8% at $350.49 on 15 January on 9.67 million shares — 13.7x its trailing 50-day average, the largest volume day in its listed history — and made an all-time high of $360.00 later that month.

The fear gauge

Volume through the decline points to orderly repricing, not capitulation. The capitulation_gauge measures peak-leg intensity as the maximum 20-day average volume between the February peak and the November trough, divided by the median daily volume over the 180 days before the peak: that ratio is 2.07x. A genuine washout typically prints a far larger multiple; 2.07x is a modest elevation consistent with a steady bleed rather than forced, emotional selling. The single largest volume prints in this stock — 13.7x on 15 January 2026 and 9.9x back in December 2020 — sit outside the drawdown window entirely; the January spike was the acquisition, not the fall. No session inside the peak-to-trough leg registered as an unusual-volume event of that order.

Who was selling

No reported short interest is available. FINRA returned no reported short-interest position rows for PEN, and no borrow-pressure or public net-short disclosures are on file, so the level and change of shorting through the drawdown cannot be quantified from the record — daily short-sale volume, where it exists, is trading flow and not a substitute. The tab records this as a data gap rather than inferring a short thesis.

What the record does show is a change in the holder base after January 2026. A definitive merger agreement converts the register from fundamental owners toward risk-arbitrage holders, whose position is the spread to the $374 consideration rather than a view on the business. On 6 May 2026 stockholders adopted the merger agreement overwhelmingly — 28,564,786 shares for against 85,334 against, roughly 99.7% of votes cast [6]. Beyond the arbitrage community, there is no disclosed forced or structural seller — no index deletion, fund liquidation, or insider distribution — in the corpus.

Estimates versus price timing

The framework's signature is a price fall that outruns an estimate cut. Penumbra shows a stronger version of the mismatch and a weaker one at once: the price fell 25.8% through 2025 while estimates were not cut at all. Consensus EPS was beaten every quarter of the decline (surprises of +8.5%, +24.5% and +4.7%), and forward numbers rose rather than fell. So the price move did outrun the fundamentals — but downward, against improving estimates, which is de-rating, not a fear discount to a cut number. Since January 2026 the question is moot: with a definitive deal in place the company withdrew forward guidance, and consensus free-cash-flow estimates ($179.9m for FY2025, $173.1m FY2026, $216.1m FY2027) describe a business whose equity value is now set by the merger terms, not by a discounted cash-flow path. The estimates-versus-value arithmetic sits in the Damage Math and Yield tabs.

What the price reflects now

The gap between price and value today is a merger-arbitrage spread, not a dislocation. At $321.19 the stock trades 14.1% below the $374 cash reference and would return 16.4% if the deal closes at that value. That discount is what remains to be earned for bearing deal risk: stockholder approval is already secured, so the open conditions are regulatory — expiration of the U.S. Hart-Scott-Rodino waiting period and antitrust or foreign-investment clearances in certain non-U.S. jurisdictions [7]. The 27% stock component adds a smaller sensitivity to Boston Scientific's own share price. The company itself frames the residual risk plainly: in its Q2 2026 release it flagged "the risk that the pending acquisition by Boston Scientific Corporation will not be completed in the expected timeframe or at all, including the risk that required regulatory approvals will not be obtained," and declined to provide FY2026 guidance or hold a call given the pending deal [8]. The underlying business kept growing through the pendency — Q2 2026 revenue of $390.0 million, up 14.9% year over year [9] — but that growth now accrues to Boston Scientific on close, not to the standalone equity.

Bottom line

There is no qualifying dislocation here in the framework's sense. The one drawdown the record contains was a 2025 multiple de-rating with no dated trigger and muted 2.07x volume, and it has fully round-tripped — the stock trades above its February 2025 peak. The only recent event that repriced the shares is a definitive acquisition at $374, which turns the stock into a risk-arbitrage instrument priced 14.1% below the deal. The temporary-versus-permanent question the framework asks of a fallen compounder does not apply to a company under a binding merger agreement; that reading belongs to the Fit tab.