Frank Apisa wrote: Sorry Ican, but you do not know what you are talking about here.
You are corrupting probability theory -- but there is no way to get that through the concrete in your head.
And the analogy of the poker game is exactly on the mark.
I admit that you present me with quite a dilemma. On the one hand you could merely be deceiving yourself that you are competent in the mathematics of probability. On the other hand you could be a fraud thinking you can succeed in deceiving others.
What the heck; I'll assume, for the fun of it, that you fit the first case.
It is the sequence of the bases in the genome that specifies the configuration of proteins comprising a living organism. After all there are only four different kinds of bases in a genome, unlike a card deck in which there are 52 different kinds of cards (jokers removed).
Let's rule out any intelligent influence and play five card stud with all cards up and a single bet of $1 prior to the deal. The meaning (i.e., value) of the five cards is independent of the sequence in which the cards arrived. Whether you are dealt, for example, all spades AKJQ10, or all spades 10JQKA, or all spades J10AKQ, or any of the other of the 120 different possible sequences for those particular five cards, the value of your hand is the same regardless of the sequence of the cards in your hand and you will win the dollar (hands down, so to speak).
So the card analogy is invalid on those accounts. In addition, the genome game consists of "hands" exceeding a million bases for which the sequence of those bases determines the outcome. The question with which we are confronted is how many "deals" (edits/mutations of the genome) containing over a million bases will be required to increase the probability of any specified sequence to at least one in 10^100 (one in a googol=a googolth). In the described stud poker game the odds are far better than a googolth to get any particular hand in a single deal.
A valid analogy would be a game of dice consisting of say a million dice with each die containing its own unique serial number. Of course, a particular die can have any one of six face values, unlike a base position which can have only anyone of four values. But the analogy is close enough to warrant serious consideration. Ok, arrange the dice in the order of their serial numbers. A <ROLL> shall consist of rolling each of the million dice in turn being sure to keep them arranged in the order of their serial numbers.
The question then, is how many <ROLLS> are required to increase the probability of obtaining a specified sequence of face values of one million dice to at least a googolth.
P = N/D;
or N <ROLLS> = P x D = 10^(-100) x 6^1,000,000 = 10^778,151 = 10^778,051.
At 10^99 <ROLLs> per year it will take 10^777,952 years to get 10^778,051 <ROLLs>. That's equivalent to 10 multiplied by itself 777,952 times.
Well, you, Frank, will undoubtedly argue: "So what! The human genome did in fact evolve, did it not? "Yes", I'll answer, "but the probability that it evolved by pure undirected chance in the time alleged to have been available is kinda small, while the probability that it evolved by directed chance is kinda close to certainty." You will then argue, but it's possible it did happen by undirected chance." I will then argue, " Yeah, it's possible, but the odds are better than 50% that it did not occur by undirected chance!" You will continue to insist, "but since it's possible, we have no reason to think directed chance is more likely than undirected chance."
I of course, will then throw up my hands and sigh the words: "non sequitur!"
You will respond, "there is no way to get that through the concrete in your head."
I will respond, "You need to use something more appropriate than a chisel".
You will respond, "Cute, nice try, but you're wrong".