Ahh.. thought so. Statistical fail, I think.

If I’m reading this right it says that not using lights was a contributing factor in 2% of accidents. But it does not say how many of the total accidents were using lights, nor does it say how many accidents happened in the dark.

The statistic we need is how many riders with lights were involved in accidents vs total riders with lights; and how many riders without lights were involved in accidents vs total riders without lights. In both daytime and night time.

Am I right?

😆

But you don’t actually know if I’m right, you’re just guessing?

No.

But if you can’t tell if I’m right or not then you don’t understand stats enough to quote articles. So I might cordially invite you to get back inside your own box 🙂

There must be some stats on this somewhere – I genuinely don’t know the answer..but how far ahead does a car on dipped headlights throw its beam so the driver can see an obstruction clearly. Is the distance greater or lesser than the stopping distances at 40/50/60/70 in the highway code?

In order to be able to make any even slightly meaningful inferences from the data.

Actually the Stats19 data does tell you those things. But it doesn’t tell you how many people are riding around with or without lights, nor the total distances covered by each. And without that you still can’t make any meaningful inferences (if there are 100 collisions in each set, what if there were 100 people in one set and 1,000 in the other? but then what if the 100 in the first set all rode 1,000 miles each and those in the second only rode 10?). And there’s no data set that’ll tell you those things.

And even then you’d have to account for population behaviours: let’s take the (quite reasonable) hypothesis that people who ride without lights are less safety-conscious and more inclined to take risks. This makes them inherently more likely to be involved in a collision. Thus even if you know the per-mile collision rates for lit and unlit cyclists, you still have a population bias that you pretty much can’t filter out.

And then you’d have to consider the population bias in terms of locations, and…

What Bez said.

The question we are debating is ‘are lights worth using?’ not ‘what’s the greatest risk to cyclists?’

And given we don’t know how much of a risk not using lights is, we don’t know how much the non light users are to blame, do we?

It would be pretty ridiculous to argue that whilst there are careless drivers on the roads we don’t need to use lights – which is the only conclusion from your post.

Disagree. It is a legal requirement to be lit, for obvious reasons. Drivers should be looking out for everything, that is true, but being unlit makes this very much more difficult.

Are you saying that if a driver is paying attention then lights should not be necessary? Cos that’s bobbins IMO.

It really does sound like you’re arguing against the use of lights..?

Look – it’s dark at night, even under streetlights, so why should lights not be mandatory? They are for cars, so why not for bikes? Really not following you here.

You have a duty to follow certain safety rules when driving, riding and walking. Otherwise you could dress completely in black and leap out from behind a lorry into the road and blame the driver that hits you..? The safety rules involve the highway code, watching where you are going and using lights at night.

Can a driver reasonably be expected to see a rider without lights in all instances?

I absolutely NEVER advocate taking anything at face value. In every accident all the facts need to be considered, both by the real courts and the internet ones. Consequently I would never assign blame to any cyclist in an accident for not using lights unless I knew all the circumstances. However I WOULD condemn a cyclist for not using lights *in general* because it’s risky behaviour. Note the subtle difference between those two statements.

Sometimes the cyclist IS to blame, you know. As is the pedestrian.

No it’s not.

Do keep up – I think us idiots have already agreed that it’s not reasonable to expect a driver to see a cyclist without lights when the driver is waiting at a T junction and the cyclist is riding along the road the driver is looking to join (assuming no street lights). I’m sure there are other examples, but that’s a good one.

I don’t think anybody has come up with an example of where it would be unreasonable to expect a driver to see a cyclist without a rear light though.

So you reckon it’s reasonable for a driver driving into a low sun or in bright sunlight not to see a cyclist without a rear light? Would it be reasonable for the driver not to see them in those circumstances even if the cyclist did have a rear light?

Let’s just drive into a space whether or not we can see what is in that space.

Probably when riding into a low sun…

I’m not convinced, unless the light is more powerful than the sun, then the rider will be a shadow, light or no light.

Jesus. If you’re riding at night just put some bloody lights on. FFS.

I’ve said it before – the true offence here is ‘failing to spot the police’

Of course I’m bloody well not, read my posts, especially the bit where I said this:

The bit about the low sun was in direct response (and ONLY in response) to the bit where you said you couldn’t think of a situation where a driver might not reasonably be expected to SEE a rider without a rear light. I made no allocation of blame in any hypothetical accident in such a situation.

Not necessarily. Where a collision occurs and the cyclist is unlit, this is likely to be recorded as a factor. But what you don’t know is whether the same collision would have occurred with lights*. To get a better idea of that, you’d need to compare normalised collision rates across the with/without lights sets, adjusting for confounding factors.

–

* Example: driver is distracted by phone or similar, collides with cyclist, obviously says “I didn’t see him”; cyclist doesn’t have lights, so it’s recorded as a contributory factor (and I suspect may well be in all cases where it appears), but it’s not necessarily the cause of the failure to see him.

PS I don’t think anyone’s claiming that there are no incidents in which a lack of lights was a genuine contributory factor.

To think that it’s not unreasonable for the driver not to see the cyclist but that he would be at fault if he ran the cyclist over is a failure of logic, hence by giving your opinion on one you’ve also given your opinion on the other, whatever your disclaimer.

Or to put it another way, if you think a driver who runs down a cyclist is at fault even if they are diving into a low sun, then the logical progression from that is that it is unreasonable for a driver not to see a cyclist in those circumstances.

At the point I came into this thread I thought we’d already done “lights are a good idea”, and moved onto discussing whether motorists should be expected to look where they’re going whether or not cyclists are lit up.

So slow down so that you can process the information and not run into something in front of you (which might be all sorts of other unlit things apart from cyclists). The thing is, the whole argument you’re making here is very much based on the hegemony of the car.

No, because I also include cyclists running into other cyclists.

Don’t invent prejudices for me. Keep an open mind on the internet, rather than believing what you want to believe. I’m NOT pro car hegemony, and this whole debate is about practicalities of the current situation, not future transport policy.

So bloody stop it, it’s a pain in the arse.