Posted by David D'Aquin on July 13, 1997 at 16:46:39: In Reply to: Re: Reasons for Kelly posted by Bisser on July 13, 1997 at 12:30:09:Grimy is talking about exp logarithm of growth, not your expected win. To put it simply, betting too much causes your bank to jump up and down in a strange fashion such that the expected number of wins brings about a losing result if you are betting more than twice the optimal amount. I'm sure Grimy thought you understood that his use of the word expect applied to the expected logarithm of growth. The same for his number one. The optimal proportion maximizes the rate at which you expect your bank to grow. This also does not mean expected win, which would be maximized by betting 100% of the bank on each play, but it means the expected log of growth is maximized... the outcome obtained via the expected number of wins and losses... what I call the median outcome.
- Re: Reasons for Kelly Bisser 11:17:26 7/14/97 (17)
- Re: Reasons for Kelly Pete Moss 17:10:05 7/14/97 (16)
- Isn't log($) arbitrary? Thicko 04:02:09 7/15/97 (13)
- Re: Isn't log($) arbitrary? Pete Moss 13:24:50 7/15/97 (1)
- P.s. (Re: Isn't log($) arbitrary?) Pete Moss 13:27:33 7/15/97 (0)
- Re: This is the million dollar question! Bisser 12:22:53 7/15/97 (10)
- Re: This is the million dollar question! David D'Aquin 17:48:55 7/15/97 (2)
- Re: This sounds like a great reason to use Kelly... Bisser 18:08:48 7/15/97 (1)
- Re: This sounds like a great reason to use Kelly... David D'Aquin 21:06:55 7/15/97 (0)
- Re: This is the million dollar question! Pete Moss 13:32:04 7/15/97 (6)
- Re: This is the million dollar question! Bisser 17:00:45 7/15/97 (5)
- Re: This is the million dollar question! Pete Moss 19:23:43 7/15/97 (4)
- Re: E(log G) or E(G) Bisser 20:20:09 7/15/97 (3)
- Re: Reasons for Kelly Bisser 20:06:48 7/14/97 (1)
- Re: "bet" or "bad" Bisser 20:58:47 7/14/97 (0)