I would disagree with the reasoning in your statistical model......
Vandal (Regular; 142)
Posted on: 02-05-2013.
Client: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; .NET4.0C; .NET4.0E)
Message views: 60 (Score: 0)
I would argue that the reason for a star system is that certain players are more valuable than others and would therefore be statistically more likely to be all-conference. Otherwise why even bother to rate the players.
Using your model I could state that a 5 star player is 100 times more likely than a 3 star player to be all-conference and therefore worth 100 red balls. Likewise a 4 star player is twice as likely as a 3 star player to be all conference and therefore 50 of them would be worth 100 green balls. You would then have an equal number of all colors and I could argue at least a few red balls should be picked.
Your analysis only works if all players have an equal chance to be all conference. That may really be the case, but if you put any value at all into star ratings, they aren't.
To me it looks like the data says if you have all three star players or above then there is a good chance you'll have a few great players. It also implies 1 and 2 star players will rarely become conference stars.
You do not have the required security level to post to this thread.
Click to report a bug!