For a city so suffused with bikes, Cambridge is not known for them being ridden safely. Given how often people ride them with both hands on their phone, swerving erratically around tourists, it’s really a wonder that there aren’t more accidents.

Looking at the science behind it all only serves to make this even more mysterious. For something so commonplace, there’s still remarkably little consensus on how bicycles manage to stay up almost effortlessly. One of the common arguments is that it’s essentially a gyroscopic effect, as shown by a spinning top – essentially, if objects rotate quickly they have a tendency to remain close to their spinning axis – but it can be shown that the resulting forces seem far too small to have much of an effect. However, it can’t be entirely due to the rider, either – most bikes will stay upright for a good twenty seconds if you push them at a decent speed and let go, much longer than you’d expect for a typical object. You can see how it’s a bit of a problem.

In a truly wonderful paper published in 1970, the Imperial chemist David Jones decided to try and work out the mechanism once and for all, which he elected to do by making a series experimental bikes, each with a new feature designed to somehow reduce the bicycle’s stability. His end goal, to produce a bike that was completely unrideable, proved remarkably hard to reach. He first attached a flywheel to the bicycle’s front wheel, intended to counteract the alleged gyroscopic effects at play by spinning in the opposite sense. Although initially a somewhat strange riding experience, it soon proved to be fairly easy to ride.

He played around a bit more – making the tyres extremely thin, reversing the tilt of the front fork – but neither had the desired effect (indeed, the latter change in fact made it even more stable). What finally provided the killer blow was a factor barely anyone had considered: the offset between the front wheel’s centre and the extension of the fork. He showed that the sum of this distance (scaled by the wheel radius) and the angle of the front fork to the vertical gives a very good measure of a bike’s natural stability, which turned out to fit the data for almost every bicycle ever made.

There’s more to it than that, of course – the actions of the rider are naturally key to the bike’s motion – but it seems that they don’t need to be that big to keep it steady. It’s interesting, though, that something which appears so simple is still not fully understood – sometimes we don’t yet have the answers.