This is cool, but shouldn't be all that surprising. The article doesn't go into any metrics used to test beginner racing candidates. It's essentially like pair

This is cool, but shouldn't be all that surprising. The article doesn't go into any metrics used to test beginner racing candidates. It's essentially like pairing astronauts who have done extensive virtual testing with people who have limited field experience. You expect the former group to outdo the latter, but obviously they will underperform against those who have done extensive field work in space. And so on. Many programs use virtual testing as a precursor to field work. So I'm not surprised that expert gamers who have probably clocked in near 10,000 hours of virtual racing can beat comparatively greenhorn racers who know how to drive a car but don't have much actual racing experience.