Fatality risk on Boris-bikes?

As of the 23rd May 2022 this website is archived and will receive no further updates.

understandinguncertainty.org was produced by the Winton programme for the public understanding of risk based in the Statistical Laboratory in the University of Cambridge. The aim was to help improve the way that uncertainty and risk are discussed in society, and show how probability and statistics can be both useful and entertaining.

Many of the animations were produced using Flash and will no longer work.

I was very saddened by the death on Friday of a Boris-bike rider in Whitechapel High Street, particularly as I am a frequent and enthusiastic user of the scheme. But as a statistician, I also immediately wondered how surprised I should be about the fact that this was the first fatality of the bikes. My conclusion, using a very rapid and crude analysis, is that it does not suggest that Boris-bikes are of higher risk than average cycling, and if anything we have been fortunate that it has taken this long for the first fatality. Of course this does not lessen the tragedy of the event.

Transport for London report that between December 2010 and 31st May 2013 there were around 22,000,000 Barclays Cycle Hire (the official name) trips in London. There were 750,000 trips in May, so let’s assume that by July 7th there were around 23,000,000 trips. These journeys were an average of 20 minutes during the week and 28 minutes at the weekend, so conservatively we could assume 1.5 miles for each trip, giving a total of at least 34,000,000 miles cycled on Boris bikes since the opening of the scheme to non-members.

The Department of Transport reports that in 2011 there were 22 cyclist deaths per billion km (620,000,000 miles), which works out as one cycling fatality expected every 620,000,000/22 = 28,000,000 miles [see page 234 of this report, eventually found through the shambolic chaos of the government statistics web-links]. Of course Boris-bike users are not average: they are probably somewhat higher risk since in London and include inexperienced tourists, compensated by being lower risk by not being very old or young, and cycling extremely heavy and slow bikes. They also rarely wear cycle helmets, but I am not getting into that tricky area .

If we very crudely assume these factors cancel out and Boris bike trips are of average risk, then to have a fatal accident after 34,000,000 miles is, unfortunately, not surprising. In fact, very roughly, there is perhaps less than 30% chance that it would have taken this long.

So I am not very surprised to hear of this tragic accident, but do feel shocked that it happened on a so-called cycle ‘superhighway’. My personal opinion, as someone who has negotiated that particular stretch of road with some trepidation, is that far more needs to be done to make cycle-friendly and protective routes in London.


Good Afternoon, First of all I would like to question the government, Boris Johnson or anyone in the decision making process of where bicycles should travel on Britain's roads… I find it incredible that anyone could even think of Bicycles sharing the road with Motorised Traffic! In Europe there are cycle lanes the run as an extension of the pavement but totally separate from pedestrians and traffic. Bicycles should not be near anything with a motor as we don't travel as fast, therefore a hazard to the driver and ourselves. I get to a junction to go straight on and the car next to me is turning left!! Do you get my drift? I'm cycling along in the designated lane and suddenly !! it disappears!! what am I to do now… Disappear too!! Unbelievable. It isn't rocket science, just a little careful planning and a not too great expense which will pay for itself in the long run in lives alone.

I agree with you David that the estimated risk derived from this one fatality on a Boris Bike is well within expectations of risks derived for all cyclists using the governments published figures in the documents you link to. What has always concerned me about those government figures however is what is the likely error on measurement of deaths per unit distance? I can believe that the counts of deaths are reasonably accurate, but what is the error in measuring total distances travelled? More importantly are the errors in determining these figures likely to be the same for all modes of transport e.g. HGVs, cars, buses, bicycles, pedestrians? For some of these modes of transport there are multiple ways of estimating distances travelled and thus cross-checking the estimates independent of simple road side surveys e.g. motor vehicles have their mileage recorded for MOT purposes, motor vehicles rely on fuel, the sales of which are also recorded and average fuel consumption estimates are also available. However for cycling and walking I wonder just how accurate the surveys are in determining the actual distances travelled each year? So how sure can we be that when the government reports an annual change in the death rates for cycling or walking per unit distance (as has happened in the last few years), that these changes aren't within the limits of error for determining the estimated distances?

Hi mrc7, regarding your comment on the accuracy of 'miles cycled' you raise a very valid point. I used to be the person who produced these stats at DfT and can agree with you that they are just an estimate. They hold up to some scrutiny at an aggregate level, but they really are not robust when broken down (as they can be by road type, region and even local authority). Furthermore it is pedal cycle data which is really hard to calculate. Traffic on major roads is fairly accurate (although PC traffic here is small - none on motorways, for example), however it is less accurate in some other place: - traffic estimates for London are a little shady (due to slow moving traffic - vehicles are classified by the automatic counters in the ground as 'long' or 'short' and then expanded into vehicle categories based on 24 hour manual counts around the year) - traffic estimates on minor roads are of lower quality. Due to the sheer volume of minor roads a sample of these roads are counted (every year) and these estimates are multiplied up to give an estimate of overall traffic on minor roads; given that (I think!) the majority of PC traffic is on minor roads - we have another problem - DfT data doesn't include off-road/private road traffic although perhaps not a problem here I remember the pedal cycle traffic was particularly our least robust set of estimates. I think this is an interesting piece of analysis – but you are correct to be cautious of the DfT vehicle kilometre data.

In answer to Jabbo5's comments above, I would simply point out that it is a myth that all or even most cyclists "in Europe" ride on segregated cycle facilities. I've toured on my bicycle in France and in Switzerland, most of the time on unsegregated roads shared with motor vehicles. Even if you visit countries like Denmark and The Netherlands where dedicated cycle lanes are commonplace in urban areas you still observe that there is plenty of situations where traffic conflict is possible, e.g. at the many junctions and crossing points. I would maintain that it's not so much the segregation of traffic that makes cycling safer in some countries compared to others, but it is more down to the attitudes of the drivers. In my experience in many of the countries I list here, the drivers seems to be far more considerate towards and tolerant of cyclists compared to drivers in the UK, most probably this is because many of those drivers also have a lot of personal experience as regular cyclists. I think it's much more of a social issue than it is an infrastructure issue.

I concur with mrc that it's attitude as much as infrastructure that affects behaviour around cyclists. I've been pondering whether there's scope for a change in the law to create a presumption of guilt on the part of a motor vehicle user who hits an unmotorised user. The countries where it's not socially acceptable to kill cyclists and pedestrians seem to be the same countries that have this presumption in law. Did the law cause the behaviour or the attitude lead to the law? I've also been pondering the British tendency - or is it just that of our media? - to cast all things as Good Things (motherhood, apple pie, low taxes, freedom from control, free healthcare) and Bad Things (foreigners, speed cameras, cyclists in lycra, children killed by speeding cars, having to pay taxes). Cyclists are classed as Bad Things because a few of them are lawless, often as a result of fear and/or a desire to make progress. There's an argument that runs: 'when bad cyclists stop jumping red lights and riding on pavements we'll support the right of riders not to be killed' which if applied to any other situation would cause outrage: 'when the naughty children stop breaking windows and vandalising things then the nice children will have a right not to be abused'. It's like the irrationality about response to speed cameras which implies speeding is OK in some places (some dead children for example) but not others: how about 'these children may be abused but not those others' or 'some houses may be burgled but not others'. The relationship between speed and risk is well-proven, but still we fight enforcement. I think leadership is the key; when national government commits to safety (I believe we are the only country in UN without casualty reduction targets; ministers still of talk about 'ending the war on the motorist' despite car users now being a minority of deaths) then we can start changing things. Boris has supported boldness of change in London, but where space is tight (Aldgate) decisions are tough, and people waver. 'Screw your courage to the sticking point and we'll not fail' - OK, maybe things didn't work out so well for Mr and Mrs Macbeth, but - regicide aside - the approach would serve politicians well (providing it is underpinned by evidence-based policy....)