"We don't really know how bikes work"

The somewhat sensationalist language of the video is in my opinion more the fault of the sources than of the authors of the video. The mathematics of control stability are complex for any type of vehicle, and for bikes it is complicated by the fact that the cyclist is not a rigid body and weighs much more than the bicycle. There are, however, more intuitive ways to explain the new findings of recent research. It is not necessary to emphasise the mathematical complexity nor mistify the results.

Gyroscopic forces were once believed to dominate, and they do at high speeds, but an elementary calculation would have shown long ago that it cannot explain low speed stability. Experiments with additional counter-rotating wheels demonstrated graphically that gyro forces are not what keeps a bicycle on the level.

Trail provides a restoring force that straightens the fork, but it is wheel flop that generates the countersteering that rights up the bike, and it is thus the most important stability mechanism. It happens that in all practical bicycle configurations that exist today flop and trail are closely linked, so trail is also a good proxy for flop and stability.

The dutch-american team devised an experiment to show that even trail is not essential for stability: the essential element is spontaneous countersteering, and it is "powered" by the lowering of the centre of gravity of the system. In a conventional bicycle, flop causes the centre of gravity of the whole bike to lower when the fork is turned. The team built an experimental bike with *negative* trail and flop, but they also attached a weight to a long arm projecting forwards from the fork. When the fork is turned the centre of gravity of the frame is raised (rather than lowered as in a conventional bicycle) but the weight attached to the fork is lowered, and that is enough to produce spontaneous countersteering, and thus a self-stable ensemble.

Thanks both. That Nature article was great. I also prefer to read than watch.

Jobst wrote a bit about this too: https://www.sheldonbrown.com/brandt/gyro.html

despite the authoritative tone in that article (plucked from a usenet discussion), he was wise enough to see bicycle stability as a generally unsolved problem. I wonder if he was aware of Jim's work on the subject.

Thanks for this link. I thought it was fascinating. 

Thanks. I haven't watched that fully yet, but found this article on Jim Papadopoulos: https://www.nature.com/news/the-bicycle-problem-that-nearly-broke-mathematics-1.20281

(I usually prefer reading than watching YouTube - you can go at your own speed)