TCU vs. Oregon, plus an explanation

First let me post the plays, and then I’ll explain how I derived them:

Site Team Risked To Win Result Net Results
The Power Rank TCU 712 1745 0 83818
ESPN FPI TCU 8083 19803 0 -19985
BCF Toys TCU 4774 11697 0 -12412
S&P+ TCU 894 2190 0 -68911
Compu-Picks TCU 2611 6396 0 -42768

As you might note, these aren’t particularly huge numbers, despite the fact that all five systems list TCU as the favorite. Interestingly, all five systems would have bet on TCU at the last line before Trevone Boykin was arrested and declared out for the Alama Bowl. In theory, I could simply give everyone the play as of when Boykin was still eligible, but I have been giving everyone gameday lines, and there was no reason to expect the Alamo Bowl to be different than any other bowl game. As such, I derived the market value of Boykin in terms of TCU’s Pythagorean rating. I held Oregon as a constant, and derived what each of the five systems rated TCU. For the record, I don’t assume each had the same rating for Oregon, but that wouldn’t matter for our purposes as any change in Oregon’s rating would be reflected in the relative difference to TCU’s. After solving for each system’s rating for TCU with Boykin, I applied the adjustment that the market applied. I then took that new Pythagorean rating and applied put it against the constant I had used for Oregon. That gave me a new expected win% for TCU, which is what was used to generate these wagers. I kind of wish at least one of these systems had favored Oregon before Boykin’s loss, but alas, it wasn’t meant to be.

 

 

 

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: