The physics of running red lights

It’s the Christmas silly season and newspapers are again rounding on cyclists (aka ‘lycra louts’) for running red lights and putting other road users at risk. Never mind the lack of any hard evidence of injuries or deaths caused by cyclists running red lights, it’s a story that appears to please news and online editors, such as at the Sunday Times yesterday and The Times today.

To borrow a memorable phrase from Peter Mandelson, I am intensely relaxed about cyclists running red lights, where it is safe to do so, i.e. when there is no obstruction caused to anyone using the junction who has priority at the time. Josh Hart has a great blog post on the subject.

Josh says it’s important not to confuse safe behaviour with law-abiding behaviour. “You can follow every law and still put yourself in a terribly dangerous position (i.e. in the door zone). By the same token, you can slow and look around carefully at red lights and stop signs and proceed when no one is coming and you’ll likely never get into trouble. Blindly following the law is a recipe for getting hurt on your bike. Better to trust your own hearing, sight, and instincts than the government’s rigid idea of ‘health and safety’.”

There’s some evidence to suggest that cyclists stopping at traffic lights on the left hand side make themselves more vulnerable to getting squashed by lorries and HGVs/LGVs that find it hard to see a cyclists in that position. This may explain why London Mayor Boris Johnson is considering allowing cyclists to turn left on red as well as riding the ‘wrong’ way down one-way streets as is already the practice in Brussels, among many other cities.

This would go some way towards addressing the second plank of Josh’s argument, that the Highway Code is inappropriate for cyclists. He say’s “unfortunately we live in a society where the needs of one class of road user are prioritised at the expense of more vulnerable road users.”

Buffalo Bill made the same point long ago on The Bike Show when he said that traffic rules were only needed to address the problem of automobiles on the roads, not cyclists, and therefore the rules should not apply to cyclists in the same way. As Josh argues, “The bicycle is a kind of a hybrid animal– somewhere between a pedestrian and a vehicle, and we need to treat it as such… Let’s stop trying to fit the round peg of cycling into the square hole of overly regimented traffic regulations.” Amen to that.

But before anyone accuses me of encouraging a free-for-all at red lights I strongly believe that to run a red light safely you need to have all your senses about you. You need to be looking hard enough to see that there’s no danger. You need to give way to pedestrians who have priority and anticipate those that have not yet stepped off the pavement. Setting traps to catch cyclists running red lights is so obviously an instance of wasting police time, though there is one small respect in which I do approve of Fixed Penalty Notices for red light jumping: If you don’t see the rozzer who nicks you, you weren’t looking hard enough to be safe running the red light, were you? So pay up and consider it a £30 lesson in cycle training. Sometimes when I’m cycling I can’t really be bothered with the effort involved in running red lights so I am quite happy to stop and wait, assuming there’s no potentially left-turning lorry looming over me.

What never seems to be discussed in the whole debate on red lights is why cyclists feel such a strong urge to run red lights, why it makes so much sense for us to do it. Those who write about it invariably ascribe it to the perceived arrogance and smugness of the cyclist, an expression of their innate sentiments of self-satisfaction and superiority. I feel a lot of things when riding the streets of London but superiority is rarely one of them. Certainly not when it’s pouring with rain and another bus has whistled past within a few inches of me, accompanied by a billowing cloud of filth and spray.

Cyclists run red lights not because we consider ourselves to be supreme beings but because of the forces of physics. The nature of the bicycle is that we have to use our own energy to move. Getting going is not simply a matter of disengaging the clutch with the left foot and easing the right foot onto the accelerator. When cyclists stop we have to give up all the kinetic energy we’ve built up and when we start again we have to overcome the inertia of our own body masss and that of the bicycle, the road resistance, wind, and so on. Cyclists running red lights are like people walking in a park who walk in the direction of where they want to go rather than following the paved paths. If enough of them are doing it, the result is a muddy rut in the grass. We don’t blame the walkers, we blame the planner who didn’t think about desire lines when making the paved paths.

It’s my sense that running red lights is much less about saving time than saving energy. A cyclist who runs red lights generally doesn’t save much time because she’ll usually end up stopping at a major junction at which point a law-abiding cyclist will have caught up, or nearly caught up. The vital point is that the red light running cyclist will have expended much less energy. And this is important. Cycling can be hard work, and the perception of effort puts a lot of people off riding bikes. The easier it is, the more people are likely to take it up. (By the same token, all those roadies who use their daily commutes to work out as a form of training, should perhaps reconsider their red light running, since stopping means they use more energy and therefore get a better workout. The expenditure of energy to get up to cruising speed might be considered as a form of interval training.)

All this is just my intuition based on my own experiences of cycling. I’d really like to find a physicist or an engineer who would be able to properly quantify the time loss and the energy loss involved in stopping at red lights. I think the results would be interesting and would help illiuminate the debate. The implications might go beyond the debate on red light running and could be useful for transport planners who, if you ask me, should be working hard to design routes for cyclists that minimise the amount of stopping that’s required. If you’re a physicist or an engineer who could apply your knowledge to these questions, please get in touch.

30 thoughts on “The physics of running red lights

  1. Ingenious argument. And I agree. But not with running reds until such time that its legal.

    Learning to track stand will help move off more rapidly, but no idea on the energy aspect.

    Good podcasts, thanks.

  2. From a practicality standpoint, rather than a physics one: to stop, many cyclists must — in addition to slowing their vehicles in a much more measured fashion than motorists (slowing from 15 to 0 is going to take more than 15 feet, *thank* you), a cyclist may need to unclip from a clipless cleat, or loosen a strap to remove his/her shoe from a toe clip before the bike comes to a complete halt, less s/he fall over. Again, a few seconds more time. On the restart, seconds are lost reclipping.

    Stopping (and unclipping) is not unreasonable where the road to the intersection is clear and the cyclist has enough warning time to complete these actions (for example, if the light turns red while the 10mph cyclist is 30-40 feet from the stop line). In urban traffic, and some suburban traffic, the status of the light is not visible through motor traffic, and the only “red light” warning a cyclist gets is the brake lights of the cars in front of him/her. Usually this is insufficient warning to stop in the queue, requiring the cyclist to coast-and-brake in the door zone until s/he gets to the stop line (and the front of the queue). Unfortunately for cyclists, stopped traffic is also a cue to pedestrians to come out from the cars parked at the curb and jaywalk between the temporarily-stopped cars, adding the hazard of a bicycle-pedestrian collision in this time interval.

    The issues of lead time, door zones, and pedestrian safety aside, the potential (or lack thereof) for time savings depends on the cyclist’s time-and-distance required for stopping and starting, his/her cruising speed, and how the sequence of lights is timed. For example, two lights in my town, a block apart, are timed such that while traveling north, if I start off from the stop line the instant the first one turns green, I will be stopped by the second light unless I can accelerate to 15mph or greater within 5 seconds. (In practical terms, this means I will always be stopped by both lights.) This suggests that in absence of cross traffic, it should be possible to make time by running lights. (In reality, there is always cross traffic at both these intersections, making it unsafe for a cyclist to attempt to run a light. Not that I don’t see sidewalk-cycling, helmetless immigrants with packages hanging from both wrists doing it regularly.)

  3. How can we campaign against bad driving when we’re condoning breaking the law?
    Manoever and position yourself away from trucks and door zones and just take the stop/start as a fact of life.

    Ride on the road, follow the rules. Ignore the rules with similar reasoning to speeding drivers and perpetuate the lycra lout myth for the media to beat us with.

  4. As I mentioned on twitter, for a commuter on a bike, a constant power output on a flat course ia a reasonable assumption, and there’s an online tool that’ll do the required calculations:

    My own commute is 5k and there are 11 sets of lights, some of which will be green (NB this assumption never seems realistic…); so lets say I hit a red light every 1k on average. I’ll compare the time taken for 1k at constant speed versus 1k from a standing start at the same, constant, power; since effectively the stop/start time is just 5x this 1k time. I’ll assume I stop suddenly then the light immediately turns green, so the time doesn’t include waiting at lights.

    At 150W, starting at 0m/s, it says I’ll take 119.6s per km. The steady-state speed at 150W is about 9.3m/s, which will take 107.5s. So, roughly 10% s more time, and since this is constant power, 10% more energy.

    That speed (around 20mph) is a bit unrealistic, most commuters will be going more like 15mph: 6.7m/s. Sticking in the figures to make that work turns out to need just 66W (this model takes no account of efficiency, so the power figures aren’t realistic). That changes the times to 149.1s versus 163.8s. A slightly smaller difference, as you’d expect from the lower air resistance.

    However, 10% difference in energy isn’t very much on a short commute like this. One biscuit would cover it.

  5. More publicity should be given to the US States that have laws enabling cyclists to treat red lights as a ‘stop’ or ‘give way’ sign. I am sure I came across some research this year that indicated cyclist/motorists collisions were down in these States since the introduction of these laws.



  6. “Blindly following the law is a recipe for getting hurt on your bike.” Could you give some more examples of this, and does that mean cyclists should ignore the entire Highway Code, or just the bits they find inconvenient?

    I was a cyclist, and didn’t run red lights, not out of any sense of smugness, but I just think that cyclists and motorists have to share the road and all follow the same rules. The other day when crossing the road, I had to stop halfway to avoid being hit by a cyclist who was running the red light – a pedestrian stopping while the green man was illuminated to let traffic go by!

    Regardless of how safe or unsafe it is, it comes down to the fact that running red lights is illegal. I also believe that when drivers see it happening, their opinion of cyclists goes down a little, and they might be just a little less aware or careful next time they’re driving near one. This of course could have fatal consequences.

  7. Cycling home in the dark tonight I passed 17 cyclists. 7.5 of them had lights – the .5 guy was the best. He had only a dying rear light mounted facing forward.
    As cyclists we regularly see the majority do little to preserve their safety or their responsibility as road users.

  8. @ Matt: Note that I stress cyclists should not obstruct other road users who have priority. I see this is principally a question of courtesy, politeness and safety. I think the law needs changing and if the law is an ass I’m quite happy that it be broken.

    @mike1727: I don’t campaign against law-breaking driving, I campaign against dangerous driving. I think there’s a difference.

  9. Whether it can be done safely or not, running red lights brings cycling into disrepute (as evidenced by all the bad press). So, for that reason alone, I don’t do it.

  10. @matt: Let’s make a list of other things that cause drivers’ opinion of cyclists to go down:
    Not using the cycle lane
    Riding two abreast
    Not wearing a helmet
    Taking up the whole lane when the road is narrow
    Overtaking when they’re stuck in a traffic jam
    Complaining when they overtake with less than 1 foot of space
    Disagreeing with their imaginary version of the highway code
    Not paying “road tax”
    Eating tofu

  11. Thanks for this article. I completely agree. And I have absolutely no time for people who’s idea of the law is that if it’s the law you have to abide by it. Too much to even start… But simple courtesy would be good. Especially from cyclists.

  12. From my Dad, a retired physics lecturer:

    Jack’s original blog is a very good one but I doubt whether an analysis of the science involved would be very rewarding.

    The time “saved” is simply the time for which the lights stay red. This would be easy to estimate for one cyclist on a fixed journey but very difficult to quantify over all cyclists in any one day. I suppose some kind of estimate could be made but, as the Chinaman said on seeing his first 100-metre sprint, “Please, what do you do with time saved?” There might be an economic argument for people doing deliveries or taking messages on a bike.

    The energy involved is fairly easy to calculate using E = 1/2 x M x V squared:

    Slowing down from a speed, V, of 10 metres/second, a cyclist and machine of mass 80kg would lose 4000 joules. This is converted almost entirely into heat energy by the friction of the brakes resulting in a very slight warming of the atmosphere. It would be interesting to devise a means of using the energy to drive a dynamo that would charge up a battery for the cycle lights. To do this without using the brakes however would mean that slowing down would take much longer.

    Starting up from rest to get to the same speed would involve the same energy.

    Pulling up behind or alongside a lorry involves a simple bit of optics. If you, the cyclist, can see the lorry driver in his mirror, then he can see you. If you can’t see him, then he can’t see you. Know this and act accordingly!

  13. @Nico – “And I have absolutely no time for people who’s idea of the law is that if it’s the law you have to abide by it.” so where do you draw the line? Is it OK if I nick your bike?

    Sorry but I don’t think you can pick and choose.

  14. Pingback: Why Cyclists Run Red Lights –

  15. @Neil, I think Nico is envoking the idea that natural law can take precedence over human (civil) law. Imagine you live in a really oppresive country with terrible laws. They are still the laws, would you feel the same urge to obey those laws?

    Isn’t it possible that even though our country is democratic and among the more reasonable countries in the world, we may have some laws which don’t conform to natural/moral law?

    Recreational use of drugs, consenting sex between 15 year olds, assisting the suicide of the terminally ill, camping out on open land… These are all examples of things which are strictly speaking against the law but where, in many people’s opinion, the current law cuts across the grain of natural morality.

    Stealing a bike is clearly against both the human (civil) law and against natural law as it involves harm being done to another. I just don’t see the harm that’s done by running red lights, subject to taking due care and attention. Eventually I’d like to see the law changed so that red lights applied in a different way to cyclists, so we could treat them as yield signs. In the meantime, I’ll do what I think is right and considerate.

  16. Thoughtlessly jumping lights can be anti-social, so too is going thru’ a green light when a pedestrian has mistimed their amber gamble and is still in the carriage way.

    At the risk of paraphrasing Jack, we are people, not automatons, as such we should behave with courtesy and use our discretion when ever we interact with our fellow man.

    For all those who follow the law to the letter I refer them to a cautionary ditty from the 19th century, it relates to the advent of steam launches on the Thames:

    Here lies the grave of Mike O’Day
    Who died maintaining his right of way.
    He right was clear his will was strong.
    But he’s just as dead as if he’d been wrong.

  17. Pingback: Red light means go (or does it?) | The Bike Show - a cycling radio show and podcast from Resonance FM

  18. physics schmysics.

    The reality is that running red lights is often safer than waiting for a green. Taking off surrounded by motor vehicles is not a safe practice. If you get a chance to avoid having to deal with motor vehicles you should take it for safety’s sake.

    Japan seems to lead the way in this respect by having bike lanes across pedestrian crossings. It means you can turn left at red lights and more importantly you can join in with pedestrian crossings instead of waiting in the middle of busy intersections to turn right.

    The motor vehicle always hurts the cyclist, its absurd to expect cyclists to obey the same rules as motorists.

  19. Pingback: Running Reds. « * SPINWELL CYCLING CLUB *

  20. Agree with you 100 % Its about not losing that energy that you have gained, all for a silly red light although I stop most times. The other point is rules of the road are there for cars and everyone else can get out of the way.
    Maybe they could have a bye law where a cyclist can run a red at his own risk. Ie if you run the red and something bad happens, tough youre at fault.

  21. Pingback: Bus driver jailed for 17 months. - Page 4 - HEXUS Community Discussion Forums

  22. Seriously? You are whining about having to move a part of your body. The thing that cyclist forget about is their actions affecting others. Just because its inconvenient for you to stop doesn’t give you the right to not stop. That law is there to protect everybody using that intersection. If you want the road laws to change in favor of cycling then don’t be an asshole! If cyclists are courteous of other road users (from the vast majority of my experiences indicate that they are not) then why should we cater to you. Acting like a bitch when you are inconvenienced is just not a way to get what you want.

    I’m not saying drivers are faultless when it comes to breaking traffic laws and I feel the same about those people as I do bad cyclists. Bad drivers are bad drivers whether its a car a bike or a bus. That doesn’t mean everybody should be bad drivers. Where would the get us?!?! Just learn to share the road with everybody. Drivers have to do it why can’t you?

  23. The difficulty of getting started is all about overcoming inertia, replacing all that kinetic energy that went to heating up your braking system when you stopped. I don’t think air and rolling resistance really come into it, since you’re working even harder against those while cruising at a steady speed.

Leave a Reply

Your email address will not be published. Required fields are marked *