Sunday, November 11, 2012

One quarter of a micromort

Belated thanks to Roger Browne who responded to my last blog post and pointed out that there already is a standard unit of risk-- the micromort, which measures a one-in-a-million probability of death.

In the time it takes me to write this blog post, I've got something like a quarter of a micromort risk of dying from all causes. The average person racks up about 40 micromorts per day.

That's a pretty good benchmark! My little brain can understand numbers like one-quarter or 40.

There a great list of relative risks for various things on the wikipedia page for micromort. I've always wondered just how dangerous skydiving really is, and it isn't as dangerous as I thought. My risk of dying if I ever decided to jump out of an airplane (7 micromorts) would be about the same risk as sitting on the couch and watching TV for 4 hours.

Since I wrote my last blog post, we've had a hard frost, so my risk of dying from an Eastern Equine Encephalitis-infected mosquito is now zero micromorts.

I'm not sure what to think of the fact that while we were being warned about mosquitos here in Massachusetts, dying from fungus-tainted steroid shots manufactured here and administered by our doctors turned out to be a much higher risk. How much higher? I dunno. But I would if health officials and reporters started telling us how many micromorts of risk we're getting when we go outside for an hour during mosquito season or get a tainted injection from our doctor...


Wim Coenen said...

Perhaps interesting to note: you can only sum the risk of separate events as long as you're talking about small risks. For large risks, summing them is an overestimate.

For example, the combined chance of dying when you do two things which individually give a 50% chance of dying is not 100% but 75%.

For chances as small as a micromort you can safely add them because 1-(1-p)(1-q) = q+p-qp ~= p+q for very small p and q.

Anonymous said...

Health officials would need whole new department of micromort-o-metrology then. And who is going to pay for that?

Gavin Andresen said...

RE: paying for a micromort-o-metrology office:

Maybe if we had a better understanding of the comparative risks of things we'd spend a lot less money on stuff like Wars on Terror or Drugs.

So a micromort-o-metrology office might be a great investment.