Amazon Fell 95%. Webvan Fell to Zero. Same Crash, Same Chart Shape, Opposite Endings.
Everything in this piece is general market history — publicly documented, widely reported facts about two companies from the same era — not a claim tested against our own data the way our backtests are. Same disclosure as always: what we've verified against our own archive and what's just publicly known are two different categories, and a good story doesn't get to blur that line.
With that said: two companies, one crash, roughly the same chart shape on the way down, and about as different an ending as two stocks can have.
Amazon, December 1999 to October 2001
Amazon went public in May 1997 selling books online — a business model most of Wall Street treated as a novelty act, not a category. By December 1999 the stock had ridden the same dot-com euphoria lifting every internet name with a ".com" in its ticker to a split-adjusted high in the low hundreds of dollars.
Then the bust arrived. Amazon's stock fell roughly 95% peak-to-trough, bottoming around $6 in the fall of 2001. The company came close enough to running out of cash that its own founder later described the period as genuinely existential for the business — this isn't hindsight talking; contemporaneous coverage at the time was openly questioning whether Amazon would survive as an independent company at all.
It did more than survive. Amazon posted its first-ever quarterly profit in the fourth quarter of 2001 — a few cents a share, small enough that some outlets barely covered it — and went on to become one of the most valuable companies in the world.
Webvan, November 1999 to July 2001
Webvan's pitch was genuinely similar in spirit: order groceries online, get them delivered same-day, skip the trip to the store. It IPO'd in November 1999, months after Amazon had already proven online retail could pull in enormous capital, and it raised similarly enormous sums to build automated warehouses across the country.
The chart shape into the crash looked a lot like Amazon's: a euphoric run into late 1999, then a grinding collapse through 2000 and into 2001 that erased the overwhelming majority of the stock's value. Same era, same investor enthusiasm, same brutal drawdown.
Webvan filed for Chapter 11 bankruptcy in July 2001, roughly eighteen months after its IPO. The warehouses shut down. The stock went to zero.
The chart couldn't tell you which one you were looking at
Pull up Amazon's chart and Webvan's chart side by side in early 2001 — both down more than 80% from their highs, both still falling, both about as "extended below any moving average you'd care to draw" as a chart gets — and price action alone doesn't tell you which company is which. A purely technical read of both charts would have generated the same signal on both: deeply oversold, way below trend, exactly the kind of setup a mean-reversion model exists to flag.
One business had a real, if unprofitable, path to scale, real revenue growth, and a founder willing to run the company close to insolvency to protect the long game. The other had a business model that never worked economically — same-day grocery delivery at 1999-era costs on 1999-era logistics, burning cash faster than any conceivable scale could fix.
The price didn't know the difference. Only the businesses did.
This is exactly the limit we disclose on the methodology page: a score built from price, volume and trend can tell you an asset is statistically stretched — cheap relative to its own recent history — and it will read Amazon-in-2001 and Webvan-in-2001 as the same setup, because from a pure price-action lens, they were the same setup. What it can't do, and was never built to do, is tell you which one is which. That part was always on the person doing the buying.
Scores in this piece come from the ZonedInOut model. How the score works →
Frequently asked
How much did Amazon stock fall during the dot-com crash?
What happened to Webvan?
Did Amazon almost go bankrupt during the dot-com crash?
Can a stock chart tell you whether a company will survive a crash?
Get the Sunday Brief
One email a week, before the open: what the board did, which calls worked, which didn't, and what changed. Free. No spam, unsubscribe anytime.