Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Scumbria who wrote (127910)	11/9/2000 11:10:19 PM
From: david_langston	Respond to of 1570745

Scumbria, the odds are astronomical One thing that was drilled into me when I took statistics, just because the odds of an event happening are astronomical that doesn't mean the event never happens. Just ask any of your dinosaur friends about the odds of a 10 Km bolide strike. <g> Dave

To: Scumbria who wrote (127910)	11/9/2000 11:18:44 PM
From: Joe NYC	Read Replies (2) \| Respond to of 1570745

Scumbria, I am not sure, but I think it should approach some limit and not get lower. My quick model of 12 coins shows about 7% chance of getting 3 or less (25%) My guess is that the coin tosses form a normal distribution. Suppose your left extreme is 0, center is 1000, right extreme is 2000. The solution I think is the area under the curve that goes up to 500. I think once you leave small discrete numbers, you start to approach a limit. The area is much smaller than the are under 500 to 2000, but I don't think it is not infitesimally smaller. I think the ratio between the 2 areas should be fixed. But I don't think this is the correct problem to solve. You are looking at net errors per county, and taking a net of that. A county (let's take the first one) that added 65 to Gore 62 to Bush could have had 165 additions to Gore 100 subtractions, 162 additions to Bush, 100 subtractions. Now next step is netting across counties. These 2 steps reduce the "fidelity" so much that you are left with very little of significance. If you are going to base your calculation on errors, you need to include all errors (error = differences between 1st count and 2nd count). Joe

To: Scumbria who wrote (127910)	11/10/2000 1:00:16 PM
From: jcholewa	Read Replies (4) \| Respond to of 1570745

> I'm not certain how to calculate the odds, but I'm quite certain that the odds are astronomical. I wrote a brute force > program to do the calculation, but it can't practically run past 2^24. > 16 coin tosses have about 1 chance in 30 of getting a 75%/25% split. 24 coin tosses have about 1 chance in 100 of > getting a 75%/25% split, and the odds get much, much worse as the number of tosses increases towards 2,000. If you wanted to find out the probability that 2000 coin tosses would get 500 tails or less, you'd be better (I think) to find out the probability of getting at least 1500 heads (in no particular order). I believe this is pretty easy. The chances of flipping 1500 heads where all other flips can be any combination of heads and tails should be something like: 0.5 ^ 1500 x 1.0 ^ 500 = 0.5 ^ 1500 = 2.85 x 10^(-452) Basically, the probability (where certainty is one) of getting 1500 heads or more in a 2000 flip (or actually in any number of flips equal to or greater than 1500) dataset is a decimal point followed by four hundred and fifty-one zeros followed by 2851060965 (approximately). The chances of this happening are about one in thirty-five billion vigintillion vigintillion vigintillion vigintillion vigintillion vigintillion vigintillion, where a vigintillion is equal to one thousand decillion, where a decillion is equal to one thousand quintillion, where a quintillion is equal to one thousand billion. In comparison, there are estimated to be "only" ten trillion vigintillion neutrons (or was it electrons?) in the universe. The ratio of the inverse of the above probability and the number of neutrons in the universe is probably greater than the ratio of the size of the local galactic cluster to the size of your spleen. Apologies if I misdid the math. I think I have it more or less correct. :) And this exercise is only meant to show that Scumbria's hypothesis about flipping 500 tails or less in a 2000 flip contest being astronomically small was a conservative statement. However, this post is not meant to confirm or deny Scumbria's suggestion that this has any relation to the current political episode. Assuming I didn't do something insanely major in my above math, I may followup on that further point, but I dunno if I want to make me into flamebait. <g>     -JC