Wednesday, 20 March 2013

Cyclists' 10 worst junctions revealed

Cambridge News have published a story with collision counts at the junctions either end of Lensfield Road in Cambridge. The story contains some factual errors which have been analysed and corrected.

Cyclists' 10 worst junctions revealed

08:12 Monday 18 March 2013
Written byDebra Fox

It begins:

The 10 most dangerous junctions for cyclists in Cambridge have been revealed, with Lensfield Road the top hotspot for accidents.
The Lensfield Road, Trumpington Street and Fen Causeway junction was recorded as the most dangerous with 41 accidents recorded in six years, while 20 accidents put Perne Road, Radegund Road and Birdwood Road roundabout in at number 10.
The latest figures from Levenes Cycle Injury, the personal injury lawyers, show the number of accidents have increased at all of the 10 locations in the city. However, Cambridgeshire County Council disputes the figures.

The story is similar to one published a year ago in a number of sources. Levenes have previously published their own story on Levenes' web site dated 29th March 2012. Cambridge News themselves published a very similar story Most Dangerous Junctions for Cyclists Revealed 15th March 2012 with numbers. And the figures from Levenes were also used in a similarly named story by The Cambridge Student 22nd March 2012.

The story references the total number of collisions at two junctions, each at either end of Lensfield Road. Its a confusing set of numbers which become clearer when you see the source.

Levenes provided these numbers to Cambridge News and gave me the same data to help clear up factual errors:

The Junction2005-2010
6 years
7 years
1. Lensfield Rd/Trumpington St/The Fen Causeway3641
2. Queen Edith's Way/Fendon Rd/Hills Road Triangle3439
3. East Rd/Mill Rd3133
4. Maids Causeway/Victoria Ave/Jesus Lane Roundabout2733
5. Hills Road/Gonville Pl/Lensfield Rd/Regent St Cross Roads3031
6. Castle St/Northampton St/Chesterton Ln2627
7. Milton Rd/Elizabeth Way Roundabout2326
8. Hills Rd/Cherry Hinton Rd2325
9. Emmanuel St/St Andrews St/Downing St2022
10. Perne Rd/Radegund Rd/Birdwood Roundabout1520

Quotes from the story:

The Lensfield Road, Trumpington Street and Fen Causeway junction was recorded as the most dangerous with 41 accidents recorded in six years.

There is a mistake, 41 is the Levenes figure for 7 years, not 6 as stated.

The Catholic Church junction, where Hills Road, Gonville Place, Lensfield Road and Regent Street meet ... came fifth on the list after 31 cyclists were said to have been injured there between 2005 and 2011 [7 years].

This number is correctly quoted.

But a spokesman for Cambridgeshire County Council said the Lensfield Road junction with Fen Causeway had only had 31 accidents between 2005 and 2010  [6 years] – 10 fewer than the Levenes statistics.

This number is not correct. Given the earlier mistake by Cambridge News quoting the wrong year, we cannot be sure what question CN asked the council to answer, or if there were any mistakes whilst writing the story. This could be a case of "garbage in, garbage out".

A small part of the story said:

The latest figures from Levenes Cycle Injury, the personal injury lawyers, show the number of accidents have increased at all of the 10 locations in the city

The story did not give the numbers as they did in previous years so it was impossible for the reader to check this claim.  The total number of collisions has increased - obvious given one extra year of data - but the collision rate has not accelerated.

It is worth adding that Levenes and the Council take different approaches when working out which are the worst junctions for collisions.

Levenes have simply counted the total number of collisions involving cyclists.

Cambridgeshire County Council use an algorithm score the sites which skews results towards more severe collisions. The scoring process considers all collisions and road users, not just those involving cyclists.

Last year the story was more topical as cyclists were busy marking their most dangerous junctions on The Times Cycle Safe Map with the Hills Road end of Lensfield Road having twice as many submissions as the TrumpingtonSt/Fen Causeway end. Shortly after, the council announced they had plans to update the Catholic Church junction at Hills Road / Lensfield Road.

STATS19 data for 2012 will be released to the general public around July 2013 and that is when we'll start to see the next round of updates and headlines from all the news sources. Levenes have said they will be updating their map when the new data is released.

If you spot any errors or can provide clearer information, please do get in contact. 


  1. The Cambridge "News" article did attract a lot of comments though! ;-)

  2. Your report says that the Hills Road end of Lensfield Road had twice as many submissions as the Hills Road end. Should that have read instead: "The Hills Road end of Lensfield Road had twice as many submissions as the Trumpington Street end"?

    1. Thanks bikemapper, good spot. I have corrected the article above with the correct road names.

    2. My pleasure. Would you mind ...? I have added a few more routes to my proposed design for a strategic cycling network, which you can see here (best viewed with the terrain box ticked). Could I get some feedback, please? Do you think the map is easy to read? Is there anything obvious that is missing?

      Do you agree with Al who thought the reason the authorities are not planning to do anything more adventurous at the Catholic Church junction is because it would upset traffic flow at other locations, and they don't have a plan for how to deal with that. To which he responds, of course: WHY THE HELL NOT? "It highlights that there's a need for a detailed plan of what a pro-pedestrian and cycle road network could be for the whole of Cambridge," he says, "so that it can be implemented. The lack of such a plan has just been used here to do basically f**k all."

    3. Thanks to everyone for their suggestions. I have tried to incorporate as many of them as I could, as can be seen here. Please keep them coming!

    4. Can't see a terrain box. Nice concept - like a tube map - but doesn't overlay well over a Google Map, just too many different colours so some will get lost when overlaid . A custom OSM background would be the way to go. I'd prefer names for each line, way easier to remember especially if they go to/from/past a useful destination, need to study existing world transit maps for the best idea.

      When you say proposed, I assume you mean asking the highlighted roads to be upgraded - there are some pretty hellish roads highlighted like The Backs, I always take the alternative through the centre or Grange Road.

      Re, Catholic Church. Vehicles flows is clearly one of their main aims. The council(lors) won't just give roads over to cycles due to loss of votes but maybe can be persuaded/forced on safety grounds.

      There are very few places that are designed with a giant strategic plan. The way they get funding only allows growth piecemeal. But with a clear vision it might get bit one small part at a time. Major infrastructure upgrades are rare (eg guided bus, A14 upgrade, Northstowe).

    5. Thanks for your feedback. The terrain box reveals itself when you hover your cursor over the box marked 'Map' (top right).

      When I say 'proposed', what I mean is that it is not cast in stone. Even so, you're right to assume that the routes on the network would need to be upgraded.

      It is "a basic precondition of mass cycling" that there be a comprehensive, city-wide cycle network. There are two ways to approach the development of such a network. The first way is known as an Adjustment policy (this is the piecemeal approach you describe). However, according to Cycling: the way ahead for towns and cities, "it is possible to go much further than this strictly pragmatic and ad hoc approach", and that is to pursue a Global or Holistic policy.

      Of course, it is simply not possible to develop a dense network of high quality bicycle infrastructure overnight, and this is why, on page 2 of the chapter entitled How to begin?, in big bold letters, Cycling: the way ahead suggests introducing the network to a minimum level of functioning. In practice, this amounts to the installation of interim measures.

      Unfortunately, many cycle advocates do not accept this approach. In fact, to the best of my knowledge, I am the only cycle advocate who does. This perhaps explains why I am routinely given the cold shoulder by most cycle advocacy groups (Cambridge Cycling Campaign have still not responded to my email, for example).

      It is of course essential that once the network has been "introduced", it is then developed further. The key here is sustained investment. But clearly, if it's going to get better, it's going to take time.

      "It's a long-term plan," Andrew Davis from the Environmental Transport association once explained, "and it's not about being anti-cars, because our members are car-drivers, you know, we're not anti-car. It's having the right form of transport in the right place and at the right time. And we need to do it, and we can do it gently and purposefully, that's the point. You know where you're going to go, and you tell everybody this is why we're doing it, and you bring them on board. If you make swingeing changes, no one likes that, it's quite understandable. A change of ramping up petrol prices or blocking roads is not on. It's got to be saying, 'This is where we're going in our cities, and we're going to do it purposefully, and we're telling you why we're doing it.' That's the main thing.

      If you can spare five minutes, please read this comment which I posted on Rachel Aldred's blog.

  3. As suggested by Richard Jennings, I replaced my use of the word accident with collision. The quotes used the word accident consistently, but they do mean collision. The STATS19 dataset still uses the term Accident in its filename. A hard habit to shake.