Re: Need Some Help From Mathematicians or Statisticians

On Oct 23, 10:00 am, "alwaysaskingquestions"
<alwaysaskingquesti...@xxxxxxxxx> wrote:
"r norman" <r_s_norman@xxxxxxxxxxxx> wrote in message

news:9j7sh3h51u25tl076jba0e2a2ipir0mfrb@xxxxxxxxxx

On Tue, 23 Oct 2007 16:37:25 +0100, "alwaysaskingquestions"
<alwaysaskingquesti...@xxxxxxxxx> wrote:

I'm trying to explain to somebody that the improbability of life evolving
is
nowhere near as remote as some people make it out to be.

I'm taking the Perfect Bridge Hand - 13 spades - as an example

[...]

I think you are on quite the wrong track. The probability arguments
against evolution involve numbers that are infinitesimally small, so
small that there is essentially no way that an event could occur even
given billions of years to try.

I've made that argument in a different way. The UK National lottery is based
on matching 6 random numbers drawn from 49 numbers in total. The odds of
doing that in any one draw are approximately 1 in fourteen million. There
have been 1,234 draws so far so the odds against the particular sequence of
winning numbers generated since it started are approxinately 2 E8818
against. Yet it happened!



For example, if you randomly grab
amino acids and put them into a chain, the naively calculated
probability of obtaining a particular protein of length 100 is 1 in 20
to the 100 power. If that is the way proteins are made and the way
they evolve, then the argument might even be correct, but the problem
lies in the fact that that particular probability has no relationship
to the way evolution works. If, on the other hand, you ask for the
probability of finding some protein of whatever length that has at
least a small capability of performing a specific function and so is
useful to its owner, you get a very different answer. And then you
can ask the probability of working over this protein, combining it
with sequences of other useful proteins, to gradually arrive at a
protein of length 100 with a lot of important features. Again, each
step is reasonable and, when you combine it with the ratchet effect of
selection, you don't multiply probabilities together to get the
probability of the overall result.

What you call the 'ratchet effect of selection' is exactly what I'm trying
to get across in my example of trying to deal 13 spades. According to my
calculations, it reduces the time from 1 M years to about 1K years, i.e. a
reduction of 90%.

I'd take a different approach. You'll need about 4 decks to complete
this.

Assume a 'card eating beastie' that likes red cards and a second 'card
eating beastie' that prefers rounded objects over pointy ones.
Start with one deck, and deal the hands normally.

After the first deal, remove 10 of the red cards in the deck and
discard them. Replace them with 5 red and 5 black from the other
decks. This simulates the effect of the red-card beastie +
reproduction.

Also remove 10 of the rounded objects (hearts and clubs) and replace
them with 5 rounded and 5 pointed. This simulates the effect of
rounded-card beastie + reproduction.

Within a few deals you will have almost all spades, and the odds of
your achieving your perfect hand will be much greater.

Point out that the effect would be magnified in nature because the
population of all black cards should tend to produce only black
offspring, etc.

The point here is that NS changes the population, so the odds of
reaching your goal get better with each iteration. The 'goal' is a
'best fit' to the environment (including predators).

.

Relevant Pages

  • Re: Need Some Help From Mathematicians or Statisticians
    ... I'm taking the Perfect Bridge Hand - 13 spades - as an example as the probability calculations are worked out quite well here: ... I've taken the example that if I set out to deal thirteen cards and get thirteen spades, I could expect it to take 80 billion deals on average. ... probability of obtaining a particular protein of length 100 is 1 in 20 ...
    (talk.origins)
  • Re: Need Some Help From Mathematicians or Statisticians
    ... probability calculations are worked out quite well here: ... I've taken the example that if I set out to deal thirteen cards and get ... calculations reduces the total time from 1 million years to just under 300K ... probability of obtaining a particular protein of length 100 is 1 in 20 ...
    (talk.origins)
  • Re: My views on evolution
    ... The simple fact is that the probability calculations proposed by IDers ... is the probability of obtained "that exact" sequence? ... What is a bacterial flagellum? ... of protein at all of whatever length and size and shape that at least ...
    (talk.origins)
  • Re: what are the odds of....
    ... The probability Pof getting a straight flush with suited connectors ... is the probability that you hit one of four combinations of cards by the ...
    (rec.gambling.poker)
  • Re: what are the odds of....
    ... The probability Pof getting a straight flush with suited connectors ... is the probability that you hit one of four combinations of cards by the ...
    (rec.gambling.poker)