The charts (stronger teams sit on "higher" lines):
AFC
NFC
That gives a nice visual representation of who should win, a bit of statistics can give us a level of confidence as well. Graphically, this model predicted for the four games of this past weekend:
It got it right in three out of four. No real surprise since only one game was an upset, but interesting. Note that the only prediction to fall short was the Indianapolis-San Diego game, but that was also the smallest confidence level (56% Colts vs. 44% Chargers).
Had I placed bets according to this model I would have won three and lost one (all point-lines were covered), for a payout of 190% after juice. Not bad, but again it was probably more luck than anything else.
Either way, that was too much fun, so I'm going to take a look at this week's games as well.
If we take a hint from last week we ignore the Giants-Eagles game since it is so close. It will be interesting to see if the other three games go as predicted, since they all seem very favorable at over 75% confidence (i.e. 3:1 odds or better).
This is particularly true for the Ravens' game since, along with having the highest confidence level, when the betting lines come out for this week it will probably be the only "upset" of the bunch.
4 comments:
Nicely Done, mate...Interesting little model you've assembled here. Not bad accuracy either!
I'm willing to overlook the Patriots slight since you are obviously a math adept. Turns out we can be friends afterall... :D
Leaving numbers alone, my gut says that it's going to be either the Ravens or the Titans on one side of the field this February. What say you?
Nerd
Guilty as charged... :D
2 out of 3 this past weekend (Ignoring, as you say, the Giants/Eagles game)! Not bad at all...
Ravens v. Cardinals come Superbowl, says I...
I shed a tear for your Titans on Saturday, Meshealle. :(
Post a Comment