In article <[EMAIL PROTECTED]>, Robert J. MacG. Dawson <[EMAIL PROTECTED]> wrote:
>Paul Bernhardt wrote: >> Gary Carson said on 3/28/03 5:45 AM: >> >I play a lot of poker. The result of any individual hand is whether I >> >win or lose a pot. And as long as I have the process right -- make >> >the right decisons in the play of the hand -- the particular outcome >> >in terms of turn of the cards simply doesn't matter. If I get the >> >process right (and am well bankrolled) eventually I'll get the money. >> I'm a serious Bridge addict, traveling to play in tournaments and I teach >> it. One of the things we say is "don't pay attention to results, pay >> attention to good playing decisions." We know the people who never really >> improve, they are too focused on results instead of making good >> decisions. Over time, you'll score better in the game if you are >> consistantly applying good methods, improving your methods, even though >> on any give hand you can get 'fixed' by a bad lay of the cards. If you >> adjust your methods each time you get a bad result you'll never develop >> good methods. > Isn't there a QC teaching expreriment (by Deming?) designed to make >this point - that the strategy of chasing your last error is not as good >as a consistent one? This one is well known. If the probability of winning by a particular one of two actions is 2/3, most people will play it 2/3 of the time, resulting in only a 5/9 result, which loses 2/3 of their edge over choosing the actions half the time each. Mathematics shows that "common sense" can be very much wrong, even in less clear cases. -- This address is for information only. I do not claim that these views are those of the Statistics Department or of Purdue University. Herman Rubin, Deptartment of Statistics, Purdue University [EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558 . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
