joe{at}removethispart.gs.washington.edu wrote 

> [McGinn]
> >> >Genetic drift is nothing but a more complex and, therefore, more 
> >> >conceptually intractable version of the gambler's fallacy.  
> >> >
> >> >You could also do search in google using: Gambler's fallacy 
> >> >
> >> >It seems there are two types of gambler's fallacy, as described 
> >> >on the following webpage:
> >> >
> >> >http://www.nizkor.org/features/fallacies/gamblers-fallacy.html
> >> >
> >> >I think the one associated with genetic drift is the second of 
> >> >the two on this page. 
> >> 
>  [me]
> >> I went there and saw a page on the common fallacious belief that
> >> having got some heads, tails are "due".  This has
nothing whatsoever
> >> to do with genetic drift calculations,
> 
> [McGinn]
> >I agree.  But this wasn't the issue I raised.  I suggest reading my
> >post again.
> 
> [me]
> > which do not commit this
> >> fallacy.  They use binomial distributions in the ordinary,
correct, way.
> 
> [McGinn]
> >Not the issue.  Read it again.
> 
> OK, so I read again.   The relevant passages from McGinn's post are
> cited above.  He clearly says that thinking in terms of genetic drift
> is a fallacy, the second of the two ways that one can commit the Gambler's
> Fallacy.  

Right.

> The description on that web page is:
> 
>  "The second involves cases whose probabilities of occuring are not 
>   independent of one another. For example, suppose that a boxer has won 50% 
>   of his fights over the past two years. Suppose that after several fights 
>   he has won 50% of his matches this year, that he his lost his last six 
>   fights and he has six left. If a person believed that he would win his 
>   next six fights because he has used up his losses and is
"due" for a 
>   victory, then he would have committed the Gambler's Fallacy. After all, 
>   the person would be ignoring the fact that the results of one match can 
>   influence the results of the next one. For example, the boxer might have 
>   been injured in one match which would lower his chances of winning his 
>   last six fights."  
> 
> I can only conclude that McGinn believes that people who do genetic
> drift calculations are thinking fallaciously because the genetic
> segregation, birth, and death events are not independent.

If after performing such calculations they came to the 
conclusion that their results indicated the existence 
of a form of causation distinct from natural selection 
then yes, you would be correct to conclude that I 
believe such thinking is fallacious.

> Without getting into the issue of whether events in the universe are really
> random,

If you are unwilling to confront this issue then it 
is unlikely that you will ever escape the psychological, 
gravitational pull of the gambler's fallacy.

> it is very apparent that these birth, death, and segregation events
> are closely enough modeled by random occurrences that this is a highly
> useful description.

I'm not disputing it's usefulness.  (This is not the 
issue.)  Likewise if you were a bookie taking bets on 
one of the fights of the above mentioned boxer I would 
not dispute your claim that employing statistics of 
past fights was useful.

> It really doesn't matter whether the genetic drift is
> a function of our not knowing the exact state of the universe, or whether
> it somehow is a consequence of quantum mechanical randomness.
> 
> McGinn has presented no argument that using binomial (coint tossing) type
> probability arguments is a poor description of genetic drift.  One can infer
> from the evidence above that he thinks that the outcomes of the
"tosses"
> are not independent.  If they are positively related (as in the boxer case
> above that he cites) then even more gene frequency change will occur than
> imagined by genetic drift calculations.
> 
> It is mysterious why McGinn thinks that this is true enough to lead to a
> noticeable departure from genetic drift calculations.  He has declared that
> a fallacy is being committed, but he has given no evidence.

You provided the evidence.

> 
> Note that we're not discussing whether genetic drift "is causal",

Well, I won't pretend to speak for you, but this was 
the subject I was discussing.  (Uh, so.  What about it, 
Joe?  Do you still maintain that genetic drift is causal?  
[For what it's worth, I'm under no illusion that I will 
actually get an answer to this question.  Not now.  
Not ever.])

> but just
> whether it describes in a useable way what's going on, what to expect to
> happen to the gene frequencies.

Jim
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 6/21/03 8:21:44 PM