Random data

Ask for help here and help each other out !
Posts: 6
Joined: Tue Jul 31, 2007 9:38 pm

Random data

Postby PaulB » Tue Jul 31, 2007 10:11 pm

I'm new the TSM, and was just playing around and generated some random numbers and applied kelly (eg. Evens expected win 50% of time), and it turns anything it generates into a profit. I assume I'm overlooking something here, and I always thought Kelly was based on exploiting perceived value, so what's happening? Can Kelly really make money from noise? I need to know I can trust it before I apply it to my data.

Posts: 2
Joined: Wed Aug 01, 2007 6:45 am

Postby rob » Wed Aug 01, 2007 6:48 am

Could be something wrong.
Kelly does not work on my machine at all and never has.
Posts: 6
Joined: Tue Jul 31, 2007 9:38 pm

Postby PaulB » Wed Aug 01, 2007 8:35 am

Here's what I do:

Generate random data (4000 bets, odds 2, win rate 50%)
I set the advanced settings to Min odds 1.01, max odds 4, average odds 2
Kelly Settings: Bank 100, Divisor 2, win rate 50%

In this particular case is says longest losing run 10, longest winning run 9, average odds 2, win rate 49.85%

I go to Graph and see level stakes which is a random each time (loses about 20 points), but Kelly is a strong and steady climber showing about 9000 points!
User avatar
Site Admin
Posts: 658
Joined: Wed Jun 20, 2007 4:16 pm
Location: London


Postby support » Wed Aug 01, 2007 1:10 pm

Set two of your filters to the following
<= 2
2 >

Now view your results with these filters applied. You should find that there are no bets placed at all for odds uinder 2. For odds over 2 out of 4000 you should have around 1300 bets made. Because the odds range from 2 - 4 the level stakes profit is around 600 and the kelly profit huge !

Return to “Ask For Help ”

Who is online

Users browsing this forum: No registered users and 1 guest