>> |
Anonymous
>>441016 >But there is also the fact that teams will/can fluctuate in their performance over the season in a matter that is not normal.
And the same thing can happen in the post season. With much more dramatic consequences. Typically you have three off games and you're done. Take the Rays. A one-run game, something that tends to balance out over a large sample, went to the Phillies and it became 2-1. They had a bad game three and then faced a matchup problem in Cole Hamels. A larger series you don't know they don't bounce back and win. Fortunately, we have a larger sample size to look at.
>And really, records are scewed based upon the teams they play.
This is why third-order record is a great way to evaluate teams.
>If you are going to bring in sample sizes and scientific method into this then each team would need to play each other team an equal amount of time, or a randomized schedule in which match-ups are as likely between the Rays and Red Sox and Rays and Phillies.
This would undoubtedly be better than what we have at determining which team would is better, but it's also boring.
That would be better than the current 162 game schedule for evaluation purposes, but the current 162 game schedule is also better for INTELLIGENT evaluation than looking at a collection of short series arranged as a tournament.
|