[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

MLB Forum

Effect is a little bigger than I expected

Posted By: Colin Caster
Date: 28 Aug 01, 6:01 pm

In Response To: Schedule correlation effect (James)

When you first raised the point, I correlated some random binary vectors with one another (45 cases each, representing the approximate number of remaining games in the season). I replaced a random pair of values with inverse outcomes (i.e., [W,L] or [L,W]) for each shared game between pairs of teams. The median pairwise correlation between AL "team results" (91 correlations in all) was -.07.

At the time, I figured that this wouldn't be big enough a negative correlation to affect the second decimal place of the pennant probability. Of course, as you point out, there are fewer games now, and as the number of games remaining decreases, the inverse correlation should increase in strength (between certain pairs of teams).

One question: is your sim size still 10,000?

CC

Password:

Messages In This Thread

Schedule correlation effect -- James -- 28 Aug 01, 3:08 pm
Effect is a little bigger than I expected -- Colin Caster -- 28 Aug 01, 6:01 pm
Sim size -- James -- 28 Aug 01, 11:19 pm
Value of a game -- James -- 31 Aug 01, 11:11 am
Nice analysis -- thanks. (nt) -- Colin Caster -- 31 Aug 01, 12:14 pm

 


Post Response
Your Name:
Your E-Mail Address:
Subject:
Picks Post?
Message:

If you'd like to include a link to another page with your message,
please provide both the URL address and the title of the page:

Optional Link URL:
Optional Link Title:

If you'd like to have the option of deleting your post later,
please provide a password (CASE SENSITIVE!):

Password:

If you'd like e-mail notification of responses, please check this box:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

MLB Forum is maintained by Pi Yee Press