Trading algorithms are getting smarter, human traders are not

If you have ambitions to become a hedge fund portfolio manager in the near future, the CEO of a large quantitative UK hedge fund has some bad news for you – no matter how talented, skilled or smart you think you are, computers will always be one step ahead of you.

“I realise to you that I am a dinosaur and that some day you will replace me,” said the CEO of a large quantitative hedge fund speaking anonymously to university students at the London School of Economics Alternative Investment Conference this week. “That is if the job of a portfolio manager isn’t entirely replaced by a computer.”

His argument, in a nutshell, is that computer processing power is accelerating at an ever-greater rate, the volume of available data for quantitative strategies has never been greater and even that traditional ‘human’ strengths – difficult to replicate with an algorithm – are likely to be within reach of computers in the near future.

And, while computers are getting stronger, he says, there are still plenty of mediocre traders out there: “The barriers to entry have traditionally been quite low for hedge fund managers. A lot of people who were managing money had never done so before.”

This argument comes from a position of power when quantitative hedge funds outperformed a lot of other strategies in 2014. The AHL diversified fund returned 21% in 2014, and six other Man Group funds made the top 25 European hedge funds tracked by Bloomberg. Generally, quant funds made a comeback last year, while other firms had a less than stellar year.

But at the end of 2013, after five years of mediocre returns, quant hedge funds were being written off as among the worst performing hedge fund strategies. Unpredictable moves by central bankers and politicians, anticipated or reacted to by human traders, left the super-computers struggling to react.

The hedge fund CEO admits that – along with ‘guts’ and intuition – one area that computers still struggle to compete with human traders is when there’s a “regime shift”. Human traders also have a better strategic knowledge of certain industries and an understanding of the impact of litigation and special situations.

“But the computers are much better than they were, and they’re only improving,” he said. “Slowly but surely they are catching up, and in 30 years’ time there will only be more quant funds and fewer human traders.”