TfL Official Collision Map gives open access to data


Transport for London has just released "London's first interactive collision map".

There are now many collision maps on the internet. They are of variable quality and are not always up to date.  TfL's version, like most others, is based on statistics collected by the police and recorded in their STATS19 system.

Distorted View

All these maps present a distorted view of the level of danger on the roads. The data is victim focussed and so fails to identify the causes of casualties. Almost all the cyclist and pedestrian casualties on London streets involve motor vehicles. A classification of casualties by the type of vehilce involved would be much more useful than by the type of victim.

The maps only show absolute numbers of casualties, not related to actual flows of cyclists or other victims. That can lead to policy which ignores the most frightening roads which, if they were less dangerous, would be used by more cyclists and pedestrians.

The TfL map allows quick filtering by different classes of casualties. We expect that it will be regularly updated, unlike some of the other casualty maps. Our image shows serious and fatal injuries to cyclists in East London during 2013. Like all these maps there are serious limitations. It is important to understand what is not shown as well as what we can see.

  • There is no traffic data, is a group of casualties due to the number of users or is there a serious problem with the road?
  • The map only shows one year at time
  • The classification by type of victims hides easy identification of the cause of danger, was it an HGV, bus or another cyclist?

What is really useful

  • This data is publically available, from the same official system used by all the boroughs as well as TfL.
  • The collision map is linked to downloadable data tables that allow smart users to interrogate the system for details of any location. We are hoping easy apps will become available to let anyone do this.
  • The 10 year data should mean the end of transport engineers, planners and 'safety' teams only using 36 months data when looking at junctions and prioritising works. That has led to consistent mis-allocation of investment and action to fix dangerous roads.

At Coroner's inquests into recent fatalities at Bow and Holborn the initial 36 months data led to claims that these were not dangerous locations. Looking at the five or six year casualty history showed a very different story.

Traffic engineers use of only 36 months data is outdated. It made sense 30 years ago when records were kept on paper and searching was difficult and expensive. Today there is no excuse for not having useful relevant data availble instantaneously.

Campaigners should now demand 10 years information for every location they want highway authorities to act on.

Rate based casualty analysis

Measuring casualties agains the amount of travel for that type of road user is the only effective way of understanding the risk of travel. For example the North Circular Road is so hostile very few people cycle along it. The number of casualties is low but the risk is extremly high. It is an effective barrier to cycling for many people in North London yet the casualty map would suggest that there is not a problem.

Road user risk and fear of being hurt are a major influence on how people choose to travel. Understanding risk and how to reduce danger on the roads depends on having the right information. For decades London Cycling Campaign, and every other sustainable transport pressure group, have been calling for authorities to measure risk, to publicise levels of risk and set risk reduction targets.

By not putting risk evaluation and risk based targets at the centre of danger reduction policies it is likely that planning and resources will be wasted by concentrating on the wrong places.

Transport for London's Cycling Safety Working Group seemed to accept these ideas several years ago but the latest version of the Cycling Safety Action Plan fails to set rate based casualty reduction targets.

The TfL Collisions and casualties on London's roads: Annual Report 2014 was also published this week. It does discuss road user risk and gives some rate based data for vulnerable road users. This is in the final chapter of the report, starting on page 31 out of 35.  The headline tables at the begining of the report show cyclist casualties up 64% compared to the baseline years 2005-09. Only five chapters later do we see that the risk to cyclists, measured against distance cycled, has actually fallen by 9%.

While media and public attention has been guided to the headline figures suggesting cycling has become more dangerous the truth is that is has become less dangerous to cycle in London.