The methodology in the actual paper where they calculated the risk seems roughly right: https://www.nature.com/articles/s41550-022-01718-8

Not sure how to trade on this market, though, as I think it's probably less than 70% that we would actually learn about it if it actually happened, as the person could just "go missing" in case the impact was strong enough. There's probably some calculation to be done about how likely the death is to be reported, I think the probability would land around 50%.

Interesting question. Note that most space objects are Starlink satellites, and it seems these satellites are small enough that they disintegrate before hitting the ground.

https://en.wikipedia.org/wiki/Operation_Burnt_Frost

Related market:

Note that I don't think this market being at 4% disproves the Reddit post title. The chance of a piece of falling debris killing someone is surely much higher than the chance that a Manifold user finds out about it. A significant fraction of people in the world live in places where such a thing may not be widely reported on.

@IsaacKing For the record (since I have the largest NO position by far), my thought process in betting the market down so far didn't involve such considerations. I'm betting based on the true probability, not on the probability of an observation.

Of course there's another issue, which is that the underlying paper's claim is made assuming no interventions, whereas this market will of course be resolved based on events that follow whatever interventions are deemed wise.

Does the death have to be unintentional?

Uh, yes. I forgot about that concept

Suppose a piece of space debris falls to earth, I pick it up, bring it to the top of a skyscraper, drop it into a crowd of people, and it kills someone. Was that person killed by falling space debris?

In some sense, yes. In a sense relevant to this market? No

Minor tangential rant, because I made the mistake of trying to read about this...

The press releases (sorry, "news articles") associated to this study contain the claim "The Southern Hemisphere is more at risk of falling debris, even though countries in the Northern Hemisphere are more responsible for the debris." Think about that for a bit. You should have geometry questions!

In fact the claim in the paper is more plausible---that there's a greater risk of space junk falling in the "global south". "Global south" != "southern hemisphere", by a long shot! (Equatorial countries are typically members of the "global south", and equatorial orbits are somewhat cheaper to reach, so this form of the conclusion is reasonable.)

Everything else about... all of this... screams "untrustworthy". If this market fluctuates above 5% I'll do something a bit more quantitative. (Limit order should notify me.)

(At first glance it seems that the paper assumes that a rocket re-entry kills everybody within a 10km radius. Um. This may be a classical example of some authors determined to get a certain result.)

@ScottLawrence I want to push back against this part in particular:

"The Southern Hemisphere is more at risk of falling debris, even though countries in the Northern Hemisphere are more responsible for the debris." Think about that for a bit. You should have geometry questions!

Essentially all orbits cross the equator. If you want to service North America, your satellite is going to go just as far south as it does north. Take a look at https://stuffin.space if you want to see this more clearly

@LivInTheLookingGlass Yeah, that's precisely the "geometry questions" I was hinting at. (Sorry, I really need to remember to be more explicit on the internet.)

With the exception of a precisely equatorial orbit, all orbits cross the equator, because orbits (projected onto the earth) form great circles. Moreover every orbit spends exactly the same amount of time north of the equator as it does south, blah blah blah you know all this just wanted to remove ambiguities.

Anyway. Paper in vanity journal sketchy, press releases are terrible, nothing's new.

Damn, the site is dead :(

https://web.archive.org/web/20220331095820/https://stuffin.space/

@ScottLawrence

Moreover every orbit spends exactly the same amount of time north of the equator as it does south

That is only true for circular orbits. A classic example are the Soviet Molniya satellites which used a very elliptic orbit and spent most of the time above Siberia and Canada.

Isn't the discount rate on MM, over 10 years, already of order 10%?

@ScottLawrence No, it's only a couple percent for anywhere from years to decades - because of loans it doesn't change that much for longer timeframes

https://manifold.markets/jack/this-question-will-resolve-no-on-ja

@jack Ah, I forgot about that, thanks.