This is only really meaningful for discrete events. How many times has the Sun not exploded? Well, it didn’t explode a second ago. And it didn’t explode two seconds ago. And it didn’t explode 2.00000001 seconds ago, and so on. One could discretize the problem by considering some span of time: In this case, the question would be something like “What is the probability that the Sun will explode in the next year?”. Now, you can count the number of years in which the Sun didn’t explode.
But that’s tricky, too, since “zero” is an awfully small number to base statistics on. If we could say something like “In the past billion years, the Sun has exploded 30 times”, then we’d have a good basis to estimate the probability, but with a sample size of zero, the best we can do is lay down some very fuzzy bounds.
Of course, all of this is assuming that the probability of the Sun exploding in any given year is independant of all other years, which (as you alluded to in your second paragraph) is not true. There are some circumstances for which the probabilities are truly independant (for instance, decay of a radioactive atom), in which case “the closer you get to the time it will occur” is useless. But the Sun, unlike an unstable atom, does have an internal state which progresses towards a state where it will die. This means that if we can understand that internal state sufficiently, then we can dispense with all of the probability, and just look at our watches. Well, we think we do understand it sufficiently to say that we’ve still got a few billion more years.
Cite? I can’t find anything on lasvegian statistics on Google.
That would work for something like earthquakes, where strain builds up and eventually has to be released.
It sounds more like the gambler’s fallacy- the belief that if you’ve (say) tossed a coin ten times and it’s come up heads every time, it must be due to come up tails on the next toss. The probability of tails on the next toss is still actually 50%.
I’ll take that bet! If the sun doesn’t blow up by tomorrow, you pay me $10,000. If it does blow up by tomorrow, I’ll pay you $10,000.
Larry Niven postulated such an event in his Hugo-winning story “Inconstant Moon.”
diggleblop, it takes 8.5 minutes for light to reach Earth from the Sun, IIRC. So if an astronomer saw a killer flare, he’s already dead, because the radiation-front from same would arrive at the same time the light did.
That assumes that the deadly radiation levels start at the beginning of the flare. Solar flares can last for a few hours, and it might be possible to see it starting up, before it got too deadly.
There are also protons released in solar flares, which travel slower than light- they can take 15 minutes or several hours to get from the Sun to the Earth. I know those are a concern for astronauts outside Earth’s atmosphere- there have been flares that could have killed an astronaut on the Moon protected by only a space suit. I don’t know if those events can be bad enough to affect someone shielded by Earth’s atmosphere.
Didn’t mean that to sound like I was disparaging all your excellent commentary, just trying to sum up. (great review of the physics involved though, thanks!)
Note that when a physicist or astrophysicist says something is “very unlikely”, they’re usually saying something several orders of magnitude different than you are when you say “it’s very unlikely I’ll win the lottery this week.” Lottery and Vegas odds look pretty good next to the 1 in 10 billion chance of the Sun colliding with another star when the Milky Way and the Andromeda galaxies collide. I wanted to emphasize that difference. It’s kind of like when an astronomer says a star will go supernova “soon”, and when you press for details it turns out that it will probably go supernova sometime in the next few thousand years.
I meant to say, but forgot earlier- this wasn’t a smartass question at all. It’s a very good question, actually- you should ask questions like “how do we know that”. I would have been thrilled to get a question like that from a student when I was TAing an astronomy course for non-majors.
The How to Destroy the Earth page says it might work with Mercury (number 3 under this heading), if you could somehow teleport it directly into the Sun’s core without it being vaporized by the outer layers of the Sun first. I should note, however, that this is listed under “Other, less scientifically probable ways that Earth could be destroyed”.
