Repost: Exclusive: things tend to have a greater effect on the world once they exist

So the Institute of Advanced Motorists have press released the fact that casualties are up on 20mph streets (deaths are down, but they were already in single figures, so that’s random). I thought it might be worth reposting this sarcastic rubbish that I bashed out last time some idiot tried to claim that an increase in casualties on 20mph roads is evidence of their failure.

I heard on the lunchtime news on Radio 4 today the shocking news of an increase in the number of people injured on 20mph streets. Back when there were fewer 20mph streets, fewer people were injured on 20mph streets, they revealed. Now that there are more 20mph streets, more people are being injured on 20mph streets. This road safety intervention, they concluded, isn’t working.

This watertight logic perhaps also explains why BBC News have been so quiet on the destruction of the NHS. Before the NHS existed, literally nobody at all died in any of the then non-existent NHS hospitals. Almost as soon as the NHS was created, people started dying in the newly created NHS hospitals. Clearly the NHS doesn’t work.

Members of the Association of British Nutters will no doubt be getting very excited about these numbers, but before they make rash recommendations they should remember that back before the British motorway network was built, there were literally no people injured on the British motorway network, whereas now that the British motorway network exists, there are lots.

I hope that the main elements of the astonishing innumeracy that went into the BBC story — the failure to put the raw numbers into any kind of useful context, either of the rapid growth in the number of streets with 20mph limits as it has become easier to set the limit (or their changing nature as 20mph starts to roll out beyond quiet residential streets onto busier high streets), or of the far higher number (and, more importantly, rate) of injuries and death on either equivalent 30mph streets or on the same 20mph streets before the speed was lowered — should be obvious. Needless to say, reducing speeds on a street from 30mph to 20mph cuts injuries, regardless of the entirely banal fact that those few injuries which remain will thenceforth be added to the tally for 20mph streets instead of that for 30mph.

So, mockery over,  there’s a more important point: should an increase in injuries, if there really had been one, automatically kill off further roll out of 20mph zones?

Those who dwell at the bottom of Bristol’s Evening Post presumably think so

It beggars belief that the council intend reducing the 30mph speed limit. A limit introduced when there was no such thing as MoT’s, ABS brakes, crash zones on the front of cars and good street lighting.

I can see no justification in spending this money and would dearly love to know who Bristol City Council think it will benefit? It certainly won’t be the youth, disabled or elderly.

James R Sawyer clearly thinks that the 20 zones must be all about safety, as he argues that his ABS brakes and crash zones are already plenty enough to keep him safe as he drives through Bristol at 30. But Bristol have always been clearabout why they’re moving towards a 20mph city:

Councillor Jon Rogers, Cabinet Member for Care and Health, said: “…20 mph zones create cleaner, safer, friendlier neighbourhoods for cyclists and pedestrians. They are popular with residents, as slower traffic speeds mean children can play more safely and all residents can enjoy calmer environment.”

Slower speeds are not a simple issue of cutting crude injury statistics. They’re more about reviving communities which have been spoiled and severed by traffic speeding through them, reclaiming a little bit of the public realm that has been monopolised by the motorcar, and enabling liveable walkable neighbourhoods to thrive. Far from “certainly no benefit for the youth, disabled or elderly”, we know much — some of the research having in fact been carried out in Bristol itself — about the many adverse effects of higher speeds and volumes of traffic, and the loss of shops and services due to car-centric planning and living and the blight of high streets by arterial traffic, on the mobility of those most excluded from the car addicted society, particularly the young, the elderly, and the disabled. If they’re lucky, these people will be forced into dependency on those willing to help them get around; if they’re unlucky, they will simply be left isolated and severely disadvantaged. But of course, we don’t like to acknowledge the existence of the large numbers of people who are excluded from much of our society, culture and economy by our rebuilding the world with nobody in mind except car owners.

The injury statistics cited in the BBC News piece include minor injuries, which is most injuries at slow speeds — little things which don’t require a hospital stay. What are a few more cuts and bruises if it means that thousands of kids are free to walk to school with their friends instead of stuck inside mum’s car? Would we rather keep the infirm all shut up and sedentary with no access to the shops and the services they need, too intimidated by the anti-social behaviour of motorists to cross the road, than risk one person having a fall?

These strands can be tied together by the other piece of context that would have been worth including in the BBC piece: in the same year that injuries in 20mph zones increased, injuries to pedestrians and cyclists in general increased — in part because there are more to be injured. It has always been the case that the great road safety gains that successive governments have boasted of have been won mainly by making streets so dreadful that people find them too frightening, stressful, unpleasant, humiliating or ineffective to walk, cycle, or do anything other than sit in a secure metal box on. Start making the streets a little bit less awful and people return to them.

“The overall results show that ‘signs only’ 20mph has been accompanied by a small but important reduction in daytime vehicle speeds, an increase in walking and cycling counts, especially at weekends, a strengthening of public support for 20mph, maintenance of bus journey times and reliability, and no measurable impact on air quality or noise.”

Like cycle tracks, which people still like to claim increase car-cycle collisions (they don’t) despite before-and-after studies largely ignoring the fact that the point of cycle tracks is to widen bicycle use from the confident and quick witted to the people who were are otherwise too scared, stressed or infirm to do so, so invalidating the before-and-after study design, an increase in minor injuries after speed limit reduction, even if it were really to happen, would be far from proof of a failure.

Postscript, July 2014

The IAM make a thing of the DfT stats showing a 26% increase in serious injuries in 20mph limits and a 9% decrease in 30mph limits. Given that the base figures for the two sets are so different, that amounts to 87 more injuries in 20 zones and 1102 fewer injuries in 30 zones. Of course, the only figures that would really matter (in the absence of a double blind randomised controlled trial) are before/after comparisons of the streets that have switched and/or case-control studies of those streets (at least, for measuring injuries; as I said before, there are other important outcomes to 20 zones besides injury rates). And given that these numbers are not (and could not really be) normalised to the changes in total length of the two types of street, and are influenced by far too many confounding variables, I wouldn’t dream of suggesting that they’re worth drawing any conclusion from. But if you’re intent on drawing a conclusion, given the trend in switching 30mph streets to 20mph streets, a net reduction in serious injuries of 1015 seems like a far more pertinent one than a 26% increase in injuries on 20mph streets.

12 thoughts on “Repost: Exclusive: things tend to have a greater effect on the world once they exist”

  1. so did they divide the number of casualties by the total distance of 20 mph zones for both sets of figures?

    Very devious to claim casualties are up without having ‘normalised’ the two data sets correctly…

    1. No they did not, any more than the “Association of British Drivers” who fed the same canard to the Sun a couple of years ago did. That is because no-one actually knows how many miles of 20mph road there are in this country. We do know that it has increased substantially, that is all.

      Well not quite all, Portsmouth adopted 20mph across the entire city except for distributor roads which generally remain 30 (an represent the busiest roads but a tiny percentage of total mileage). Their findings were that while average speeds didn’t actually fall that much (a couple of mph from mid 20s to low 20s) the incidence of collisions (of any kind) fell, and so did vehicle/pedestrian incidents and casualties. At the level of all incidents and all casualties the numbers are meaningful, but at the level of KSI they are too few to be statistically significant.

  2. The IAM post is clearly a pretty easy target, but actually the bit at the end of Best`s comments is far from silly
    “In locations with a proven accident problem, authorities need to spend more on changing the character of our roads so that 20mph is obvious, self-enforcing and above all contributes to fewer injuries. In Europe, it is long term investment in high quality segregated or shared surfaces that have led to a much safer environment for cyclists and pedestrians.”

  3. Was in Middlesbrough the other day, which has extensive 20mph limits in residential roads, and I was gobsmacked at the amount of children outside playing in the road it took me back to my childhood.

    I assume being outside playing they are in more danger of being hit by a car but in less danger of becoming overweight and suffering associated illnesses.

    The swings & roundabouts of public health on a national scale.

  4. they have 20 mph zones in Gloucester… and the vast majority of motorists don’t obey them… I’m often overtaken by cars and vans that must be in front despite the fact that I’m already cycling at or above 20 mph myself…

    they really need to enforce them with a network of average speed cameras… then they don’t have to put speed tables/bumps in or wiggly bits and then emergency vehicles can put their foot down bringing injured people back to A&E withough having to slow right down for every bump…

    1. The lack of enforcement – police or engineering – is a common criticism of 20mph limits (“zones”, by the way are the other ones, the ones which do have physical calming measures are intervals determined by DfT guidance).

      However, while zones typically achieve average reductions of 7-8mph, which is essential where average speeds are significantly too high already, limits tend to be about 1-2mph. But don’t underestimate the value of that. It is estimated that each 1mph reduction in average speeds results in a 6% reduction in collisions, and thus presumably in casualties. You can introduce 20 limits in about 60 miles of road for every one mile of zone, for the same cost. Is 60 times 1mph worth more than 1 times 7mph? Sounds like it might be

    2. Ah, then you have exactly the same situation as in the Netherlands (where speed cameras were invented and are not painted bright yellow).

      No drivers take much notice of speed limits here either. Where lower speed limits are imposed in places with through traffic, drivers almost always break the speed limit. The Dutch realised this more than five years ago, hence the title of that post “Lower speed limits are not enough”

      The answer is not more speed cameras and it’s not speed tables or bumps either. It’s removal of through traffic from roads which don’t need to be through roads. When the only cars which you encounter are at the ends of their journeys their drivers behave a lot differently. In the middle of journeys, drivers simply want to be somewhere else so do not behave in a rational ma

  5. ITV Central are running with this tonight about injuries being up in 20 mph zones… they’re looking for comments:

  6. The crucial issue is how we measure safety. I have written a brief guide to this issue with regard to cycling here for Local Transport Today – and the same principles apply to walking.

    More pedestrians might well be reported killed or injured if:
    A. There are more walking journeys, but the chances of being killed or hurt goes down – which is a GOOD THING
    B There is a greater chance of the injury being reported, which is a GOOD THING

    What we are concerned about is the level of DANGER presented to pedestrians. If it goes down, people may well be encouraged to walk about more – which is a good thing. Also, elderly and disabled people and children may be more likely to walk. They are particularly likely to be hurt or killed with even quite small impacts – but avoiding this by scaring them out of the way is simply not civilised.

    Also, the raw figures say nothing about who was at fault: as a pedestrian an injury which is due to my carelessness is irresponsible – but not as downright unjust as one caused by somebody else breaking the rules/law.

    Discussing the benefits of 20 mph – or any other intervention based on reducing road danger – in terms of totting up aggregate reported casualties of pedestrians is missing the point.

  7. 20 zones work they include traffic calming, the speed limit is not meant to be a traffic calming measure, sign only 20 limits are creating a socially acceptable disregard for speed limits and just driving up non compliance giving pedestrians a false sense of security and a false indication of traffic speeds. 20 limits might be more effective if they didn’t just put them everywhere, you water down their effectiveness where they are more needed boy who cried wolf style,.

    They only reason they’ve been so blunt on many streets is because 25mph limits aren’t allowed, and some of Bristol’s 20 limits are set against DfT guidelines, they have even include a stretch of road that was 40 and some duel carriageway with no pedestrian access in the centre.

    paulc properly set speed limit are meant to be self enforcing without the need for excessive police enforcement.

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

%d bloggers like this: