Reply to thread

Message

[QUOTE="Cusefan95, post: 4825807, member: 173"]
From the linked article:

"...when you tweak the method to strengthen its weaknesses, you also weaken its strengths. In particular, if you use a modified margin of victory, the numbers don't have as nice an interpretation."

That really gets to one of the issues with statistics generally, you can often manipulate them to produce the results you want.  This is a chronic issue in scientific research right now, where scientists are showing an amazing ability to find the results they want (or the people funding the research want them to find...) - but its also leading (IMHO) to the replication crisis in the sciences.

Syracuse has successfully shaken my belief that aggregated analytics can get you close to a projected point spread this year.  Starting with Clemson, we've deviated so far from what an expected point spread should be based on any publicly available modelling I could find...it seems to hold better for NFL (maybe because there's more betting so wisdom of the masses applies better in the NFL than in college?).  My best guess is the four wins led modelling to massively overvalue Syracuse, people figured out we were garbage and the modelling was wrong - and it took until THIS WEEK for the computer models to catch up to Syracuse's "true value". (True value" meaning "CDRW doesn't see the field").
[/QUOTE]

Verification

What is a Syracuse fan's favorite color?