Infographic: Cycling through red lights

Red light jumping is very, very naughty – and dangerous. But is it true that most red light violations are done by cyclists? Do we have a balanced media coverage on this matter? Well, get ready to change your perception with our infographics. GLUE team decided to visualise the findings based on reports by Cycling Info and Transport for London‘s Road Performance and Network Research Team.

For bigger image, please click here. Feel free to share love.

Remember, jumping the reds is so not worth it.

Additional note: Please read an insightful post on Red Light Jumping vs Speeding Car by Rad Wagon.

GLUE Infographics


  1. If 4% of road users are cyclists then they are equally culpable…. if less than 4% of road users statistically more culpable…. its not a useful statistic.

  2. Please use the word crashes – especially in this context ignoring a stop aspect is an intentional action so in no way are these accidents.

    It might have been more valuable research if the following data had been collected at the locations measured

    a) the number of motor vehicles also ignoring a stop aspect (ideally broken down by type bus-LGV-taxi-car)

    b) the nature of the red light infringement
    accelerating or maintaining speed through a recently changed signal
    pulling forward through a stop signal after stopping
    filtering left or across a T junction (in absence of filter lane)

    c) compartive proportion and direct count of cycles/cars failing to observe stop signal

  3. Thank you Julia and Dave for constructive feedbacks. I agree with Dave we need better breakdown of vehicles involved. Perhaps more regular report by TfL and research by transport charities would be useful to monitor this.

  4. Thanks for this – very useful. Three quick comments:

    1. Even as a cycle campaigner, I would have to say that the TfL stat about 16% of cyclists jumping red lights (in London) is perhaps unduly lenient towards cyclists – the 84% who don’t jump a red light includes those who arrive at the light when it is green anyway. The figure would doubtless be somewhat higher if it was the proportion who cross the stop line while the light is red, and who then continue through the junction (i.e. they don’t just stopping a few yards ahead of the line – technically that’s also an offence but it’s not in the same league).

    The study this comes from is downloadable from:

    …or there is a summary here:

    Click to access behavour-at-advanced-stop-lines-summary.pdf

    2. In relation to the stat showing that cyclists are only involved in 4% of pedestrian injuries at London traffic lights, Julia is also right that you need to know what proportion of vehicles passing through the junctions are pedal cycles, before you can interpret this. However measuring this would itself be less-than-straightforward, as pedestrian injuries probably aren’t evenly distributed at traffic lights throughout London. I suspect they are more concentrated in central and inner London, where both pedestrian and cycle flows are above the average for London’s road network.

    In short, statistics are rarely as simple as they appear!

    Finally there are CTC campaigns briefing on these issues of cyclists’ involvement in pedestrian injuries, and the (separate but related) topic of cycling and the law (and the safety impacts of law-breaking by drivers and cyclists alike). See and

    A range of other CTC briefings are accessible from .

    Roger Geffen
    Campaigns & Policy Director
    CTC the national cycling charity

    • Thank you for your comment and useful notes, Roger! Most insightful indeed. It is true that statistics are not as simple as they appear, let alone simplifying the message for an effective visual info. GLUE team will definitely have a look at CTC stats to inform our future cycling infographic project.

  5. Great response on GLUE infographic by Prof Michael Egan from McMaster University, Canada: “And I came home to this infographic from smellslikeglue, which is a very cool London-based style, cycling, and music site. My Arts & Science students are tasked with designing their own visual assignments, and it occurred to me that this infographic offered an apt model for thinking about their projects.”

  6. As already mentioned, please don’t use the word ‘accidents’ because many of those collisions were due to negligence or worse – they weren’t accidental. TfL and the police actually call them ‘collisions’, as you can see in their reports.

    Also, as somebody who cycles almost everywhere all year round, I do find that red light jumping varies a lot by the type of location and so take these statistics with a big pinch of salt. For example, naughty people are much less likely to jump the lights at busy and dangerous four-way junctions than at pedestrian crossings on a relatively quiet road. There are a few junctions I pass on my daily commute to work where jumping is absolutely endemic and a regular nuisance to pedestrians.

  7. Don’t get me wrong I am not supporting jumping red lights however, as a cyclist I have sat at a red light waiting and waiting and waiting fo rit to change because the traffic light has not worked out or sensed that I am sat there waiting because I am not surrounded by a few tons of tin and a polluting engine.

    Possibly if the traffic lights were not strictky geared to these things called cars there would be less red light jumping- just a thought.

    • The sensor should detect normal bikes (I’ll forgive them missing the all-carbon stealth bikes).
      Highway Code says “If the traffic lights are not working, treat the situation as you would an unmarked junction and proceed with great care.”

      Please contact the local authority (eg via
      In my experience they say “checked and found to be working to requirements”.
      Reply again – “Then the specifications are wrong. If you are taking no action because it’s not causing a problem, then it could very easily become a problem.”
      Then either they turn up the metal-detector sensitivity, or you park on the sensor and hold up traffic.
      If your friends can stop at other lights that don’t work at the same time, it can help.

      By the way, some lights have little radar sensors, but I don’t think any react to flashing headlights, although that seems to have become a popular urban myth!

  8. Great little infographic however as a few others have pointed out RLJing does vary by junction and type of red light. When riding down CS7 I’ll see some riders go through a “safe” red light (ped crossing with no one there) yet wait when we get to a traffic light controlled junction. I recently made this clip up to show the difference in technique for RLJing between cyclists and motorists:

  9. 16% of cyclists make less that the number of drivers that violate red lights, because those latter are so much more numerous, no matter their percentage. But the rate of violating cyclist is likely to be higher, thus implying cyclists respect red lights less than drivers.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s