πWeek 7 Recap & SP+ Model Results π
What a weekend of college football! My buckeyes unfortunately lost to Oregon at the last second (should have listened to my model and not my heart). LSU sneaks a win out vs. Ole Miss in what was a sneaky important game in the SEC and playoff picture. Penn State continued to limp through their schedule struggling with a 2-loss USC team. Texas showed why they are the #1 team in the country, and don't sleep on those nerds from Vanderbilt!
Let's dive into the data. The model picked every single game this week and hit at a 60.8% clipπ°. Again, if you like beating Vegas, it may be time to start taking notice.
Week 8 Predictions will be out soon.
Match-up Data: Using row one as a guide (Liberty vs. FIU)
Home/ Away: Home & Away team for the match-up. The green highlighted team is who the model is picking to cover the spread based on Differential and Home Line (below).
Home/ Away SPS+: Encompasses an array of data all filtered into one metric. This number tells you how much better a given team is relative to the average college football team. For example, Through 7 weeks, Liberty is 3.3 points better than the average team and FIU is -15.6 (15.6 points worse than the average team).
Differential: Calculates a match up based on SP+ (Home team SP+ - Away Team SP+). I additionally layer in 3 points for home field advantage as this is what long-term data would tell you about college football games. Using the first column below, the model says that Liberty should beat FIU by 21.9 points. ex: 3.3 (-) -15.6 + 3
Home Line: Looks at the home team spread. At the time was Liberty -19.5 favorites.
Spread Differential: Takes the SP+ Differential and subtracts the Vegas line to make the pick. ex: 21.9 (-) -19.5 = 2.4. This tells you that the model believes that Liberty has 2.4 points of cushion vs. the spread, meaning they will cover it.
Result: Did the model pick correctly based on the outcome of the scores.
No comments:
Post a Comment