2107: Launch Risk

Explain xkcd: It's 'cause you're dumb.
Jump to: navigation, search
Launch Risk
Don't worry--you're less likely to die from a space launch than from a shark attack. The survival rate is pretty high for both!
Title text: Don't worry--you're less likely to die from a space launch than from a shark attack. The survival rate is pretty high for both!


This comic deals with the faulty application of general statistics based on a large population, such as all Americans, to specific situations with vastly different statistics, such as astronauts.

A manned rocket ship is about to be launched into space. Mission control counts down from "T-minus 20," where "T" stands for the time at which the rocket is scheduled to launch. In the capsule, one astronaut asks another how they are feeling. The second admits that they are nervous. The first one offers a supposedly reassuring observation that they are more likely to be struck by lightning than to be selected to become an astronaut. Such comparisons are commonly used to illustrate that a particular probability is very small, and therefore not worth worrying about.

The second astronaut is about to agree that they have a good point, but then realizes the problem with their argument: the likelihood of being selected as an astronaut is a moot point, because they both already are astronauts. The comparison ignores the relevant concern, which is the danger involved in being an astronaut and launching into space. It may also provoke a false understanding that, given two events each with given likelihoods of happening, where chance has proven sufficiently provident enough for the less likely one to have occurred, the 'luck threshold' has been exceeded enough to make the more likely one now a practical certainty (a variation upon the hot hand fallacy, though with nominally independent events and opposing desirabilities).

The second astronaut's nervousness is understandable as space missions are historically quite dangerous, and have numerous avenues for potentially fatal failure, certainly far beyond the minuscule risk of being struck by lightning, approximately 1 in 14,600 throughout your entire life.

The title text creates additional confusion by referencing another common statistical reference point, the probability of dying in a shark attack. In addition to shark attacks being uncommon, they are also less likely to kill their victim than is commonly assumed. Still, while shark attacks are more frequently fatal than rocket launches, this comparison is once again useless, as the astronaut is not in any danger of sharks, but is literal seconds from launching into space. The astronaut is presumably not especially reassured by the "pretty high" survival rate.

Of the 557 people who who had been in Earth orbit (as of the time of this comic) 18 (3%) have died in related accidents, not specifically at launch (List of spaceflight-related accidents and incidents, Astronaut/Cosmonaut Statistics). Of the 93 incidents logged for 2018 in the Global Shark Attack File, 4 (4.3%) were fatal, but the statistic has been higher in the past when there was likely less education against provoking sharks.

A large metal rocket, such as depicted would be more likely to be struck by lightning than nearby structures. However launch controllers generally monitor weather carefully to reduce the chances of attempting to launch when lightning is likely.

A spacecraft launch can also trigger lightning, by creating a conductive path through electrically charged clouds. Apollo 12 was struck by lightning twice during the launch phase. Thankfully backup systems allowed the flight to proceed. For more information, see NASA: Lightning and Launches

The perceived value of risk is a recurring topic and is also featured in 795: Conditional Risk and 1252: Increased Risk.


[A large spaceship rocket is standing on its launchpad, just decoupled from the tower that supports it ready for launch. At the very top of the rocket where the astronauts are sitting there are two lines coming out, indicating the speak from the two astronauts inside the rocket. Next to the top of the rocket there is a speech bubble drawn with jagged line indicating sound from a loudspeaker. It is the countdown to the launch.]
Astronaut 1: How you feeling?
Astronaut 2: Honestly, pretty nervous.
Astronaut 1: I know it seems dangerous, but just remember: you're more likely to be struck by lightning than to be selected to become an astronaut.
Astronaut 2: Oh that's a good-
Astronaut 2: ...Wait.
Countdown: T-Minus 20...19...


  • During the time that this comic was on the front page of xkcd, the Header text changed to promote Randall's new book How To.
    • As this could also be seen as a launch (a book launch) with certain risks, there could be some relation to putting it up together with this comic.

comment.png add a comment! ⋅ comment.png add a topic (use sparingly)! ⋅ Icons-mini-action refresh blue.gif refresh comments!


Gave a short explanation, but I think it would be good to mention probability based logical fallacies and https://what-if.xkcd.com/55/. Don’t know how to link without it looking bad. This is my first page! Netherin5 (talk) 17:28, 4 February 2019 (UTC)

Revised to a more extensive explanation including the fallacy that the second astronaut apparently realizes in mid-reply. SteveMB (talk)

What are the odds that one or both astronauts are female? I see "he" being used to refer to the second astronaut, but we don't actually know the sex of either one. 17:56, 4 February 2019 (UTC)

Fixed 18:07, 4 February 2019 (UTC)

This seems wrong, at least with the lightning explanation. I believe the joke is that since he already is an astronaut, being hit by lightning doesn’t seem unlikely. Netherin5 (talk) 18:03, 4 February 2019 (UTC)

Would be nice to add something about risk perception of common vs. uncommon and dramatic vs. more mundane seeming events. e.g. in US, lifetime chance of death from flu, 1 in 63; from automobile accident 1 in 84; from lightning 1 in 79,746; from shark attack, 1 in 3,748,067 https://www.floridamuseum.ufl.edu/shark-attacks/odds/compare-risk/death/ 18:52, 4 February 2019 (UTC)

I find it strange that 1 in 63 citizens die from flu, while 1 in 84 die in auto accidents. Those sound like old numbers to me.
ProphetZarquon (talk) 22:44, 4 February 2019 (UTC)

The risk to be killed as an astronaut should be add somewhere (it is easy to find number of death/total number of astronaut) if someone want to make the morbid calculation. Xavier Combelle (talk) 18:55, 4 February 2019 (UTC)

From some impatient Googling and Wikipedia scanning there have been just over 360 people in space and 18 deaths (excepting training including Apollo 1). That puts the death rate at just over 3%.

These were mostly Shuttle as the crews were larger. However,the title is Launch Risk, so the figure would be less than half that, but still about 1.5%. Furthermore, if you ignore the Space Planes the Launch Risk is probably very low. RIIW - Ponder it (talk) 19:07, 4 February 2019 (UTC)

Many of those 360 have been in space multiple times reducing the risk further. Sebastian -- 07:30, 6 February 2019 (UTC)

We should get a better source for the lightning info: The current citation is confirmed[citation needed] as a biased source[citation needed] owned and controlled by socialist Jews[citation needed]. 19:10, 4 February 2019 (UTC)

I would like to hear some statistics on lightning-related death rates, as compiled by anarchist Buddhists.
ProphetZarquon (talk) 22:44, 4 February 2019 (UTC)
Sorry, I've never compiled those statistics. Otherwise I fit your requirements, though. 08:33, 6 February 2019 (UTC)
The statistics compiled by followers of Zeus would be even more interesting. 09:55, 6 February 2019 (UTC)

I'd say that part of the joke was the phrasing. The astronaut's friend said "You're more likely to be struck by lightning than selected as an astronaut," which isn't very reassuring; if the friend had said "You're more likely to be killed by a lightning strike than to die in spaceflight," it might have been a consolation (albeit a fallacious one).

Removed the shark death rate statistic, since it was 1) not typical, 2) not comparable to the other statistics in the paragraph. The statistic given was the percent of shark attacks that are fatal. It used reporting from one beach in Brazil, noted for having particularly high death rate statistics [1]. The other rates listed are lifetime chance of death from particular cause - a totally different statistic.

The rocket closly resembels Soyuz. Might be this comic releted to recent Soyuz launch accident? If it is so, the one who is trolling is russian cosmonaut. And it also meeans some meta-trolling.

It could be a Soyuz, thought it looks like the conical part just below the escape tower has windows. Soyuz has just a closed fairing. 11:38, 5 February 2019 (UTC)
Nah, the booster shape is completely wrong. I think it might be Gaganyaan / GSLV-III. -- 08:01, 6 February 2019 (UTC)

Launch pads usually have lightning protection systems, as a lightning strike on an assembled rocket would be bad news. See https://ams.confex.com/ams/pdfpapers/38831.pdf for example