快猫短视频

IT IS 2011, a couple of years after the Great Recession. Quantitative analyst Vincent (Jesse Eisenberg) and his programmer cousin Anton (Alexander Skarsg氓rd) have found a way to steal a march on Wall Street: trading a millisecond ahead of the competition.

Where will they find this tiny, telling pinch of extra time? They plan to make it themselves, by stretching an optical fibre from Kansas City Internet Exchange to New York in as straight a line as possible. While everyone else waits 17 milliseconds for their information (the beat of a hummingbird鈥檚 wing is the film鈥檚 poetic, and accurate enough, conceit), Vincent, Anton and their backers will only have to wait 16 milliseconds. That鈥檚 time enough to squeeze in a few thousand algorithmically generated trades.

The trick will be to lay the cable as straight as the law allows. Never mind Amish farms, Appalachian mountain ranges, loneliness, obsession or physical frailty. They will build this 1600-kilometre-long, 10-centimetre-wide fibre tunnel if it kills them.

Scripted and filmed like a true-life story (after all, who in their right mind would make up a thriller about high-frequency trading infrastructures?) The Hummingbird Project, incredibly, springs entirely from the head of writer-director Kim Nguyen. It can鈥檛 quite decide whether to be a think piece or a buddy movie, but it can be staggeringly funny. Salma Hayek has indecent amounts of fun as Eva, the cousins鈥� abandoned boss. In a frantic attempt to keep them on her payroll, at one point she shouts: 鈥淚 think we can break the walls of perception together!鈥�

鈥淒o you recall when it took a microsecond to win or lose a fortune? What slowcoaches we were, eight years ago鈥�

It is one of those stories that, in being made up, encapsulates a lot of historical and technical insight. Hayek鈥檚 Eva can talk 鈥渘anosecond financial engineering鈥� all she wants. As a sceptical investor notes, her style of trading is really just scalping: profiting off small, short-lived price anomalies between financial exchanges.

Scalping is hard because one hefty loss wipes out millions of tiny profitable trades. And it is also impossible to do without computers because markets adjust quicker than the eye can follow.

When world markets crashed in 2008, this took a lot of the heat. It was easier for politicians to point the finger at runaway tech and artificially accelerated trading than to challenge and dismantle key institutions. But while trading algorithms have caused the odd 鈥渇lash crash鈥�, they do far more to sustain a market economy than to threaten it. This is why so-called mechanical arbitrage runs over half the trades in many markets.

Vincent and Anton鈥檚 project is entirely reasonable in a world that puts commercial operations as close to market exchanges as possible to steal millisecond advantages over competitors. Hanging over the cousins鈥� project is a rival bid to leave fibre behind and send financial information by microwave (and the discussion of 鈥減ulse-shaping algorithms鈥� will warm the heart of any telecoms engineer). Today, the industry is even more complex, with atomic clocks to arbitrate the timing of financial information. Financial instruments that scalp multiple markets are driving the creation of strategic data centres in unlikely places, as banks head for space via Elon Musk鈥檚 Starlink servers.

All of which gives the film a curiously nostalgic feel. Do you recall when it took a thousandth of a second to win or lose a fortune? What slowcoaches we were, eight years ago.