Nothing to see here: NZ road toll edition

By Eric Crampton 21/01/2015 3


I had a short op-ed in today’s Herald. It’s not online It’s now online;* here’s a clip. 

The most recent holiday road toll statistics provided fodder for a lot of armchair reckonings over the past week. There were fifteen fatal crashes this holiday season; last year, there were six.

Nobody has been keen to take credit for the increase, though we can be pretty sure that if the figures had halved instead of doubling, the Police would have credited tough speed enforcement measures and the drop in the blood alcohol limit. 

And so we all look for stories to explain the difference. Perhaps everyone had their eyes on the speedometer rather than on the road for fear of breaching the tighter speed enforcement rules. Maybe strict enforcement of the 100kph limit led to more dangerous overtaking situations. 

But the best explanation is likely that events like this can fluctuate a lot from year to year pretty easily.

When events are really infrequent, like fatal traffic accidents, small things can push around the total number of accidents. Blogger James K., at Ordinary Times, points out that this year’s holiday season was 2 days longer than last year’s, so we should have expected a twenty percent hike in accident rates solely on that basis. He also shows that, when we look at longer-term trends in the daily number of crashes, this year’s figures remain within the expected range.

They ran it with a picture of a car wreck; I prefer this picture, from JamesK.

* Thanks Sam and Thomas! Note this was first published at Offsetting Behaviour 16 January.


3 Responses to “Nothing to see here: NZ road toll edition”

  • It is very easy and common, to overlook the role of ‘statistical variation’ in generating deviations from the mean. There seemed to be a lot of blaming of tourists and foreign drivers with the armchair experts I listened to.

    Good point also about the longer holiday period. I’d missed that.

  • If we could log near misses as well as the hits that might be useful. I suspect the near misses vastly outnumber the hits therefore you really are playing Russian roulette. Probably of the order of a 100 to 500 chamber revolver and 1 bullet.

  • “near misses” are logged if you take that to mean injury and non-injury crashes (as against fatality only).