Fewer Helmets, More Deaths | Hacker News Reader

Fewer Helmets, More Deaths

nytimes.com

3 pointsjere4 comments

Hmm, that's some pretty misleading presentation.

Basically all of the "rise" in deaths, which they are trying to pin on helmet law changes, is due to a large drop in car deaths, not actually anything to do with motorcycles or helmets (unless car drivers have suddenly started wearing helmets and that's what caused the reduction in deaths?).

They don't even show a graph for any/all states that retain the helmet laws, presumably because its mere presence would show up this piece as intentionally misleading.

They do show actual deaths for Florida (which, even as a percentage seem very high compared with other states) but since it jumps up and down by 100 deaths per year both before and after the law change, I'm not sure it means anything, particularly if the number of people on bikes rose and fell, or it was linked to peak tourist activity.

I can't think of any reason why someone would scheme to mislead in this matter, some shadowy cartel of helmet manufacturers maybe?, so I'm guessing they think they're doing the right thing by blatantly misrepresenting the stats.

"Basically all of the "rise" in deaths, which they are trying to pin on helmet law changes, is due to a large drop in car deaths, not actually anything to do with motorcycles or helmets..." This is simply not true. The last three graphs all show a large rise in absolute numbers of motorcycle fatalities, independent of any automobile numbers. I agree that the authors have a ways to go before they've proved causation, especially given a significant bump in helmeted motorcycle fatalities seen in the Florida data.

I admit I misread a couple of graphs. I was caught out by the fact that the graphs were overall so similar to relative ones that came before them and that they "melt" into each other as you scroll, even though they're showing different things. But they do show that in at least two states, the death toll rose notably in absolute terms, though in one in was rising steadily anyway before the change and in the other it goes up and down dramatically after.

I also note that they're actually selling their own argument short, since they've smoothed the points between years rather than showing them as a bar chart, it looks like the death toll started rising before the law change, but (I think) that's actually an artifact of the presentation, probably depends on when in the year the law change took effect.

All but the last two graphs are possibly misleading. It is unclear if the increases are due to more riders and/or more miles per rider as the Great Recession made people pick more frugal transportation.

The last two graphs where they show riders without helmets becoming an increasing share of fatalities. Even there, however, they should have plotted those fatalities against the number of riders riding with and without helmets, and they should have plotted fatalities per miles.

I'd guess they could show an increase, but not as large an increase as the graphs they are using show.

So, wear a helmet. AND be aware of the limitations of helmets.

One more thing: What's the harm of some statistical white lies about helmets? The harm is that the real need might be for autonomous baking systems in cars and trucks. A vehicle that brakes faster and better than you can if you are about to hit a pedestrian, cyclist, or motorcyclist should probably be mandatory.

Reply on news.ycombinator.com