“The computers got it right and the humans got it wrong,” was the gist of last week’s Wall Street Journal article about attempts to capitalize on the Brexit vote. This was partly due, it said, to the fact that too many investors were prone to “projection bias”: that is, betting on what they wanted the outcome to be.

A fund category sometimes referred to as commodity trading advisors (CTAs) uses trading algorithms to identify market trends and place bets on futures and other derivatives. Most of these models didn’t factor in the British election polls. Instead, in the weeks leading up to the vote, many went on the defensive, favoring high-quality government bonds, gold and safer currencies (like the yen). For Société Générale, that stance paid off, as its CTA Index gained 1.5% the day after the vote. Lara Magnusen, portfolio strategist for California-based Altegris, says “our models aren’t going to be affected by the same sentiments a human would be.” Her firm stuck with gold and yen bets and fared well post-Brexit.

CTA fund managers say a key to their models’ success is that they can “tune out noise around market moving events…. that are important to investors but can be difficult to accurately forecast.”