How Much (Statistical) Protection Do U-Locks Offer?

Bike theft is a topic that gets lots of air play these days. We probably know the basic facts, but, curiously no one has so far produced a statistic that would help us understand the probability of having a properly U-locked bike stolen (here in town, or anywhere). Gerald Fittipaldi recently ventured that upwards of 95% of bikes stolen on the PSU campus were *not* locked with a U-lock, but beyond that all we hear are anecdotes. One would think that the Bike Theft Task Force would have some figures.

My understanding of the relative security of different locking methods is that although most any bike lock can be defeated with available technology (bolt cutters, battery powered angle grinder, etc.) the time, spectacle, noise associated with cracking a U-lock makes it far (10x? 100x?) less likely to be attempted than just taking an unlocked bike, a bike where only the front wheel secured with a QR was locked, or snipping a cable lock. But those are just hunches. I would like to see a number so in our conversations here we can cut to the chase.

Notify of

newest most voted
Inline Feedbacks
View all comments
Michael Andersen (Contributor)

Awesome question! I’d love to see research on this. Sadly I have no idea where one might find it.

Andy K
Andy K
8 years ago

The only solution is to buy a $200 bike and lock it up like its worth $2,000

Scott H
Scott H
8 years ago

Have you tried contacting Kryptonite and OnGuard for a printout of their annual claims? If they break it out by state you could glean some sort of percentage based on the number of bikes stolen here in Oregon compared to the number of U-locks that the manufacturers know have been defeated. It’s likely that they also have their own estimations regarding the likelihood of one of their u-locks being defeated, because they would need to use those numbers to estimate their own insurance policies. Much the same way an auto-insurance company tracks the number of accidents and thefts in each state in order to calculate the premiums they need to charge to turn a profit.

Granted, no matter what documents you get your hands on, it seems like the statistics you’re after will still be enigmatic: There’s no way to count the number of thieves that abandoned their plans after discovering that a bike was secured with a u-lock, or the number of thieves that simply forgot their angle grinder that day and had to settle for the bike that was only protected with a cable lock. Even the bike theft task force would have trouble tracking these variables, assuming they’re collecting numbers at all! Simply comparing the number of bikes stolen with the number of u-locks used won’t give you the whole picture.

Maybe someone at is already tracking some of these data points or maybe this is something we could collaborate with them on. It might be possible to start building a table based on their theft descriptions that starts to paint a picture of how many u-locks are being defeated.

I’ve already found a few defeated u-locks while spot checking the index for Portland:
My bike wheel and frame were locked with a u-lock to a rack. The theft happen
Lock is missing, along with the bike. Locked to rack with U-lock, gone!

8 years ago

How do you possibly expect to get such a number?

An accurate count of the bikes that weren’t stolen because a thief passed them over? Well, we did write that new bike-thief tracking app, but for some reason they’re refusing to use it.