But I am a nerd who is going to win you MONEY!
I was sitting here in my jammies last night until almost 4 AM analyzing the results of all 373 games played this season by a Division 1 college football team. I was creating a power ranking of sorts that can predict the pointspread for any pairing of the 120 teams.
Here are the re$ult$:
The way to interpret this is that if Oklahoma played Alabama on a neutral field, then [given what has happened in the season so far] Oklahoma would be expected to win by 4 points. For home-field advantage, you can add about 2.5 points to the spread (so the pointspread would become 6.5 points if the game was played in Oklahoma and 1.5 if the game was played in Alabama).
This does a nice job of retroactively predicting results. Using this ranking, if a team is favored by at least a touchdown, then, historically-speaking, they have about a 94% chance of winning the game.
I could explain how I came up with this, but then you’d be almost morally obligated to give me a wedgie.
If these results can be trusted – and of course they can be, I mean consider the source – then it suggests that the current polls are, for some teams, wildly off the mark.
In particular, these teams are hugely overrated:
Nebraska
Clemson
Arkansas
Georgia Tech
Louisiana State (aka LSU)
Oklahoma State
And these teams are hugely underrated:
Notre Dame
Tennessee
Temple
Cincinnati
Penn State
Missouri
Toledo
Also, it tells us that Memphis is really, really bad.
So now you know where to put your life savings. You don’t need to thank me. You can just slip the check for 50% of the winnings under my door.
In all seriousness, please don’t be stupid. I trust that these rankings are better than the ones in the super-political AP and Coaches polls, but they stand no chance of reliably beating the Vegas pointspreads.
