Dislocation

Dislocation

There is a real dislocation here — a 48.5% peak-to-trough fall in five months — but its shape is unusual on two counts the framework cares about. The dated trigger is the February 10, 2026 fourth-quarter print, whose headline 9% revenue miss was roughly $168 million of one-time reserve changes on a quarter that also set records for profit and cash. And the price fell far ahead of the numbers: through the drawdown, consensus forward earnings estimates rose about 14% while sustained selling volume ran only 1.35x its trailing median.

The drawdown, quantified

Source: derived from the daily price series; figures match fit_features.capitulation_gauge.drawdown (peak $24.57 on 2025-11-12; trough $12.65 on 2026-03-30; 138 days).

Source: daily closing prices, as reported; peak, trough and current dates per fit_features.capitulation_gauge.

The fall came in three distinct segments rather than one clean break. From the November 12 peak of $24.57 the stock drifted to about $16.85 by February 10 — a 31% decline over three months on ordinary volume and with no single dated catalyst, the multiple compressing as autonomous-vehicle disruption fear repriced the whole rideshare complex. The framework treats this first leg as drift, not the moment: a stock down 10-30% with no event and normal volume is the slide, not the capitulation. The event leg was February 10-12, when the fourth-quarter print took the stock from $16.85 to $13.05 — a 22.6% two-session fall. The stock then bled to its trough of $12.65 on March 30 on light volume, and has since recovered to $14.20, still 42.2% below the peak and 12.3% above the trough.

The trigger

The dated adverse event is Lyft's fourth-quarter and full-year 2025 report, released after the close on February 10, 2026; the stock fell 17.0% the next session, from $16.85 to $13.99. The report was, in its own words, a record on most axes — gross bookings grew "19% year on year" and accelerated, the company "printed over a billion dollars in cash" [1], and the year closed "with 51.3 million riders taking 946 million rides" [2]. What the tape reacted to was the top line: reported revenue of about $1.59 billion came in roughly 9% below the ~$1.75 billion consensus.

Most of that gap was a single non-recurring item. The CFO told the call the quarter carried "a one-time impact of totaling about $210 million under the category of certain legal tax and regulatory reserve changes," of which "$168 million of that $210 million impacted revenue," so that without it "our revenue would be closer to $1.8 billion" — in line with the prior quarter's revenue margin [3]. Guidance was not cut: management guided first-quarter gross-bookings growth "accelerating at the high end" with margin in line with the prior year [4], and reaffirmed the 2027 targets of $25 billion gross bookings, a 4% adjusted-EBITDA margin, and over $1 billion of free cash flow [5]. Whether a one-time reserve-driven revenue miss justifies a permanent repricing is the temporary-versus-permanent question the Damage Math tab decides; this tab only fixes what fell and when.

The fear gauge

The capitulation signal is muted. The deterministic gauge measures a volume spike of 1.35x — the peak 20-day average volume during the fall divided by the median daily volume over the 180 days before the peak (about 16.1 million shares), per fit_features.capitulation_gauge.volume_spike. A 1.35x sustained multiple is closer to orderly repricing than to an emotion-driven flush.

Source: daily traded volume, as reported; median pre-peak volume of ~16.1M shares per fit_features.capitulation_gauge.volume_spike.

Where the volume did concentrate is telling. The one true panic day was February 11, when 74.2 million shares traded — 4.6x the pre-peak median — as the earnings gap forced anchored holders out. But that spike was a single session inside an otherwise ordinary tape: the earlier drift from the peak ran at 10-20 million shares a day, and the March 30 trough itself printed on just 11.4 million shares, below the median. Peak fear, in the framework's sense, showed up for one day on the trigger and did not sustain into the low. The selling exhausted quietly rather than in a flood.

Who was selling

Lyft carries a large, structural short base. Publicly reported short interest stood at 89.6 million shares — 23.8% of float — at the July 15, 2026 settlement, down about 2% from 91.5 million on June 30; the level has run near a quarter of the float through the drawdown, roughly double the ~12.7% average of its peer group. Days-to-cover sit around 3.5-9 depending on the volume window used. This is a stock the disruption bears have been positioned against for some time, not one where a fresh, forced seller appeared.

Insider behavior points the other way, on small size. Across the drawdown window there were no forced or liquidation sales; the only sizable insider disposition was a director's roughly $3.9 million sale on December 4, 2025 at about $22.30 — before the event leg, near the top of the drift. CEO David Risher, by contrast, has bought roughly $100,000 of stock in the open market each quarter, and added again on February 13, 2026 — two days after the crash — purchasing 7,490 shares at $13.38. The composition of the selling is therefore a heavy anchored/short base and index-priced multiple compression, not disclosed informed insiders heading for the exit.

Insider figures per SEC Form 4 filings; short-interest figures per publicly reported bi-monthly short interest (settlement date July 15, 2026). Neither has a source PDF in the corpus.

Estimates versus price

The framework's signature is a price fall that outruns the estimate cut. Here the price did not merely outrun the estimates — it moved opposite to them. Over the roughly six months spanning the drawdown, consensus forward EPS rose about 14% while forward revenue was essentially flat, even as the shares fell 42-49%.

Source: CapIQ estimate snapshots, data/sp/estimates.json momentum series (as-of 2026-01-25 vs 2026-07-24); price change is peak (12 Nov 2025) to current (24 Jul 2026), per fit_features.

The consensus forward cash-flow numbers say the same thing from the other direction: the sell side models roughly $1.13 billion of free cash flow for FY2026 rising to about $1.29 billion by FY2027 — an 18-22% forward yield on the current $5.93 billion market capitalization (fit_features.consensus_forward_yield). Estimates did not fall before, with, or after the price; they held or rose while the multiple compressed. Whether that forward yield clears the framework's balance-sheet-scaled bar is worked in the Yield tab, and the gap between the earnings change and the price change is quantified in Damage Math.

What this tab establishes

The dislocation is genuine in magnitude — down 48.5% peak-to-trough, 42% from the peak today — and it has a clean, dated trigger in the February 10 revenue miss, most of which was a one-time reserve item on a record-profit, record-cash quarter. Two features complicate the classic setup, and both are stated here without diagnosis. The fear gauge is weak: sustained volume ran 1.35x its median, with a single 4.6x panic day and a low-volume trough. And the price moved independently of the numbers, falling more than 40% while forward earnings estimates rose about 14% — an estimate-independent multiple de-rating carried by a large, elevated short base rather than by forced informed sellers. Where the November-to-February drift sits relative to the February event leg, and whether the de-rating is temporary or permanent, are carried into the Damage Math and Clock tabs.