One of the great frustrations of bike theft is that we know almost nothing about it. Auto insurers have an obvious incentive to research auto theft, which has led to numerous innovations in theft prevention. Things like locking wheel nuts, radio coding and immobilisers have hugely reduced automotive theft. With no such interested body in cycling, any efforts to reduce cycle theft are just guesswork.

We have absolutely no idea whether chains are better than D-locks. We don't know which kinds of bicycle are most likely to be stolen. We don't know what the most common methods of lock removal are. We don't know if any of the tracking products actually provide a deterrent or facilitate recovery. It's practically impossible to give good advice beyond the most basic aspects of using a good quality lock and parking in a busy area.

If anyone can think of a good approach to gathering this data then please let me know - I believe it's the biggest single solvable problem in cycling development.