Bye Week data

J

JelloYacketDeleted062020

Banhammer'd
Ban Hammer'd
Joined
Sep 15, 2003
Messages
3,418
Saw some posts about bye weeks and how we don't do well coming off of them, and how it's the offense and time to prepare. I took a look at all college teams that played more than 100 games in FBS since 2003 and grabbed their record after a bye week and for all other games. Sheet is below, along with a scatterplot to show the relationship.

The regression line is Record After Bye = .7031*(Record in all other games) + .1131. R2 is about .48.

I included a "Predicted Bye Week" using the equation, and also put a "Residual" column in there to show the difference between actual performance and predicted. Positive numbers mean you're doing much better. We do perform pretty badly compared to our winning % in other games. But, so does Oklahoma and LSU.

I'm sure there are issues with the strength of schedule as well as whether you're in a Big 6 conference or not. No time to delve into that.

 
Last edited:
I got really drunk that weekend, so did we beat Bye or lose to Bye?
 
Outstanding analysis.

I also think it is a safe to assert that the average strength of the team played after the bye week is at least equal to the average strength of the entire schedule which is a roundabout way of saying that most teams don't play cupcakes following a bye week.
 
Is the first game of the season considered "after bye week"? I would think it would have to be if you're looking at relationship to "more time to prepare."
 
So this thing we just got changed in the ACC where nobody should have to play more than 1 team coming off a bye, we should've gone the opposite direction on that?
 
In the old days the schedule said Off instead of Bye. Thus it got a little more interesting when asked who did you beat last week.
 
Easiest bet in college football is after week 5, a team coming off a bye vs one who hasn’t yet had a bye. The rested team covers the spread.

We have faced a disproportionate number of those games. At least the new Johnson rule will bring that down to one per season.

*Duke game this year I believe
 
00Burdell said:
Outstanding analysis.

I also think it is a safe to assert that the average strength of the team played after the bye week is at least equal to the average strength of the entire schedule which is a roundabout way of saying that most teams don't play cupcakes following a bye week.
Click to expand...

That was my thinking too. I think of bye weeks as occurring during the conference schedule, so the overall SOS is probably higher after a bye week, but you're talking about so many games it should matter that much.

Also - quick note, I wrote "since 2013" in the post but the data is from 2003 onwards. I used this website: https://www.teamrankings.com/ncf/trends/ats_trends/?range=yearly_2003
 
You must log in or register to reply here.
Back
Top