Asteroid Spotting

worksinprogress.co

25 points·bekind··3 comments

I know this has nothing to do with the article, but this is one of the most beautifully designed blogs/sites I've seen for a while. Wow! Kudos to the designer. It kept me on the site for a while.

As a former student of probability (ok, I took Math 361 at UIUC and worked in Google Ads Quality!): I've always felt that probabilistic statements like this:

the number of near-Earth objects above a given length, L, in kilometres can be calculated with the equation N(>L) ≈ 1,148L-2.354, and that for any given near-Earth object, there’s a ~1.6 × 10-9 chance of hitting the Earth in any given year

are meaningless. "Whew! Only 1.6 × 10-9! I guess we can feel safe, then!"

The question isn't "what's the probability?" but "could it plausibly happen, and is it worth defending against?" (The answer is Yes to both.)

What was the probability of the 2007 housing crash? All the financial models said it was negligible, and yet it happened.

I suppose you can tell I'm a Bayesian, not a frequentist.

The other thing about that statistic is that it is the change of a single near-Earth object hitting earth. And although they propose that equation we are still working on the 140m and above asteroids. There is an estimate based on what we have discovered but as the technology improves to find the harder to detect asteroids we are still finding them at a generally increasing rate each year [0].

[0] https://cneos.jpl.nasa.gov/stats/site_140.html

Reply on news.ycombinator.com