Predicting the Tennessee-Indiana game using SPM comps

The TaxSlayer Bowl pitting the Tennessee Volunteers against the Indiana Hoosiers opened with Indiana as a slight 1.5-point favorite. It has since flipped so that now the Vols are a 1.5-point favorite. Regardless, Vegas is expecting a close game. Here’s what the GRT Statsy Preview Machine has to say about this year’s Gator Bowl and whether I think it’s right or wrong.

Vols-Hoosiers

From the perspective of Tennessee

Tennessee’s points:

  • Tennessee scoring offense for the season: 24.3
  • Indiana scoring defense for the season: 24.5

The Indiana scoring defense is most similar to the following prior Tennessee opponent(s) (FBS only):

  • BYU 24.4
  • South Carolina 26.1

Early in the season, Tennessee scored 26 points against BYU, but got 41 against the Gamecocks as the Vols began to find their stride. Taken together, that 133% of what those teams usually give up, so the SPM estimates 32.6 points for the Vols against Indiana.

Indiana’s points:

  • Tennessee scoring defense for the season: 21.7
  • Indiana scoring offense for the season: 32.6

The Indiana scoring offense is most similar to the following prior Tennessee opponent(s):

  • Georgia State 32.4
  • Florida 33

Florida scored 34 points against the Vols, and Georgia State scored 38. That’s slightly more than what those teams scored over the course of the season, and the SPM translates that into 35.9 points for Indiana against Tennessee.

Estimated score: Tennessee 32.6, Indiana 35.9

From the perspective of Indiana

Indiana’s points:

  • Indiana scoring offense for the season: 32.6
  • Tennessee scoring defense for the season: 21.7

The Tennessee scoring defense is most similar to the following prior Indiana opponent(s) (FBS only):

  • Michigan State 22.7
  • Northwestern 23.6

Indiana scored 31 points against Michigan State and 34 against Northwestern, 141% of what those teams usually give up. That makes the estimated points for Indiana against the Vols 30.6.

Tennessee’s points:

  • Indiana scoring defense for the season: 24.5
  • Tennessee scoring offense for the season: 24.3

The Tennessee scoring offense is most similar to the following prior Indiana opponent(s):

  • Maryland 25.3
  • Purdue 25.8

Maryland got 28 against Indiana, and Purdue got 41, meaning Indiana’s defense allowed the two closest comps 135% of what they usually score. Estimated points for Tennessee against Indiana: 32.8.

Estimated score: Indiana 30.6, Tennessee 32.8

SPM Final Estimates

Throw it in, cook it up, take a peek:

SPM Final estimated score: Tennessee 32.7, Indiana 33.2

SPM Final estimated spread: Indiana -.5

Difference between the SPM and the Vegas opening spread: 1

Those numbers are the SPM’s way of warning you not to put too much confidence in the outcome of this game one way or the other.

Eyeball adjustments

Here’s the thing. I think the estimates of Tennessee’s points are too low and the estimates of Indiana’s points are too high. From the Vols’ perspective, the comps include a BYU game that was really early in a long season of continued improvement. Just using South Carolina as the comp would make the estimated points for the Vols from this perspective 38.

And I feel like Indiana’s estimated points are too high because the comps are Georgia State and Florida, two games very early in the season before the defense began to improve. Against teams not named Alabama, the Tennessee defense allowed an average of only 13.5 points against teams that together averaged 23.6 points. If you use that data, the estimated points for Indiana from the Vols’ perspective would be only 19.

All of that would make the Vols’ perspective spit out Tennessee 38, Indiana 19. I don’t think the machine is quite as wrong as that, but the difference is enough to make me think this isn’t the pick ’em everybody and everything thinks it is.

So, my eyeball-adjusted prediction is Tennessee 35, Indiana 28. The SPM is imitating Vegas’ shrug on this one, but I like the Vols by a touchdown.

Other predictions from other systems

As I said before, Indiana opened as a 1.5-point favorite, and it has since shifted to the Vols being a 1.5-point favorite. With an over/under of around 49, that translates to something like Tennessee 25, Indiana 24.

ESPN’s FPI gives the Vols a 52.3% chance of winning.

I haven’t seen SP+ bowl picks yet, but when I see them I’ll add the Tennessee-Indiana prediction here.

Bottom line

The SPM gives only half a point to Indiana in this game, and it’s about as far from being confident against the spread as it can get. I think the eyes see something the machines can’t, and so I like the Vols by a touchdown.

  • Vegas (current): Tennessee, -1.5 (~Tennessee 25, Indiana 24)
  • SP+: (TBD)
  • SPM: Tennessee 32.7, Indiana 33.2 (Vols don’t cover)
  • Me: Tennessee 35, Indiana 28 (Vols cover)

What do y’all think?

0 0 votes
Article Rating
2 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Evan
Evan
4 years ago

Interesting stuff!

What happens if you were to chop off the pre-florida numbers and run the machine for comps from the remaining part of the season?