For Wiki Wednesday, here's something I ran into the other day in reference to the idea of miracles.
Littlewood's Law says that if you define a miracle as a one-in-a-million occurrence, you should encounter something you'd describe as a miracle about once every month. The law is an application of the Law of Truly Large Numbers.
Whether or not we accept Littlewood's Law, or the assumptions that it requires (that an "occurence" happens to a person once per second, for example) it is meant to drive home the idea that, in a very large universe, uncommon things actually happens more often than you might expect. We humans have a tendency to attach undue meaning to exceptional events. This is, in part, because it's so difficult for us to grasp numbers that are beyond the scope of our daily lives.
A month of seconds is well over a million seconds, (Littlewood is only counting 8 alert hours per day) but when we look back over the month we do not remember it as a month of seconds. We don't even experience it as a month of seconds. It's no surprise to me, then, that when something amazing happens, people don't remember all the times that nothing amazing happened.
Posted by James at June 11, 2008 4:47 PMSo that means I should play the lottery more, because the one-in-10-million odds really aren't that bad?
Posted by: mjfrombuffalo at June 11, 2008 10:37 PMIf you can play it for free once every second for 8 hours a day for about 10 months straight, it's a pretty good deal.
Posted by: James at June 11, 2008 10:39 PMI know that the big lottery in MA is a one in over half a billion chance of winning.
So play a hundred times. Play a thousand. Then sit down and count "one" for every time you played, once a second. You'll be counting for about fifteen minutes.
Then count "one" for every chance you won't win. You'll be counting for about sixteen years.
Like your chances?
With one in ten million, you'd be counting for less than half a year. With one in a million, which is the chance James writes about in the post, you'd be counting for eleven days. BIG difference. Still a pretty darn crappy chance of winning. And this is not assuming eight alert hours, this is constant counting. (But the point of the post is quite interesting, that you're not going to notice the hours and days of nothing interesting happening.)
You're not improving your chances much by playing every day, though, because the lottery starts over every week. So you'd better find a way to play every number combination for free immediately. Of course, a machine wouldn't be able to print out all your tickets before the lottery happened, and you would need a good system for storing them. I think your stack of tickets would be about 33 miles high (although with thin lottery paper, maybe it would be as short as 15 miles high). That's based on a stack of 500 sheets of paper being 2" high.
I recommend time traveling to the future, look at the paper, then time travel back to play the lottery. I think you'd have a better chance of figuring out how to time travel.
Posted by: Maggie at June 12, 2008 6:12 AMRemind me next time to include the tongue-in-cheek smilie.
Posted by: mjfrombuffalo at June 12, 2008 2:03 PMWe knew you were joking, but it's irresistible to revisit the reasons why lotteries are not good bets.
Posted by: James at June 12, 2008 2:07 PMOne Million. Pah! I studied Quantum Mechanics. I spit at that paltry number!
A one in a million occurrence is a pretty low threshold for something I'd consider a "miracle".
Call it one in a billion and you're talking approximately one "miraculous" event per human lifetime.
If you assume the corollary (that in any one second you have a 1/1,000,000,000 chance of observing a miracle) then
Assume 300 million in the country and across an 8 hour day of seconds you get just over 8600 miracles a day or nearly 250,000 per (lunar) month in the United States alone. Times 20 for 6 billion world population and that's 4.8 million miracles globally a month.
I wondered if you'd need to account for multiple people observing the same miracle, but since special relativity dictates that multiple observers will observe a miralce at distinctly different times, each observance is in fact a different event.
So there. 4.8 million miracles a month.
Which suits my academic background better.
One in a million totally floods and de-values the miracle market, and leads us to assume that miracles are in perpetual supply. How long have we been going on this one in one million thing? Will this proposed scarcity in miracles drive miracle futures through the roof? Have we reached "Peak Miracle" yet? If people with influence continue to ignore the evidence that miracles are simply a finite supply of fossilized supernatural beliefs and "assume" that they are rapidly replenished from "we don't know where" then we are headed for a miracle crash. How do we prepare for it?
What were we talking about again?
Posted by: Bull at June 12, 2008 3:05 PM
Sorry ... didn't carry my swags ... make that 5 million / month.
You get the idea.
Posted by: Bull at June 12, 2008 3:07 PMShit.
MPER (Miracle Producing and Exporting Religions) has officially announced they are not stepping up miracle production, and a former MPER exec has claimed miracle reserves are lower than published.
I'm buying guns and freeze-dried food.
Posted by: Bull at June 12, 2008 3:13 PMHave fun stormin' the castle.
Sorry all this miracle talk reminded me of Miracle Max (He's only MOSTLY DEAD).
And speaking of Miracle Max and The Princess Bride. The WPI summer theater company will be putting on an adaption of the book at 7pm in the Higgin's House gardens tonight, tommorrow and Saturday. I can think of worse ways to spend a nice summer evening. I'll be there Saturday (weather permitting) in case anyone out there is interested.
Posted by: B.O.B. (bob) at June 12, 2008 3:17 PM