The Tide opened as a 29.5-point favorite over Tennessee, and as of Thursday morning, it’s moved to 34.5. Will the Vols cover against Alabama?
The SPM’s performance so far in 2019
The SPM went 27-22 (55.1%) last week on all games. Above the pre-chosen confidence level, it was 8-6 (57.14%) and within the pre-chosen confidence range it was 4-4 (50%). That’s two weeks in a row that it hasn’t done as well in the confidence range than it has just over the confidence level. ¯\_(ツ)_/¯
As expected, the SPM is beginning to find its legs with more games and comps to choose from. It’s still playing catch-up a bit from earlier in the season, but after two solid weeks, it’s now 159-165 (49.07%) overall, 74-64 (53.62%) above the confidence level, and 37-24 (60.66%) within the confidence range.
Fortunately for Vols fans, the SPM got last week’s Tennessee game wrong despite being really confident that Mississippi State would not only win, but cover.
So what does the SPM have to say about the Tennessee-Alabama game? Let’s have a look.
Vols-Tide
From the perspective of Tennessee
Tennessee points:
- Tennessee scoring offense for the season: 23
- Alabama scoring defense for the season: 17
The Alabama scoring defense is most similar to the following prior Tennessee opponent(s) (FBS only):
- Florida 14.1
- Georgia 12.3
Tennessee scored 3 points against Florida and 14 points against Georgia, so against the two best comps, Tennessee scored only 64% of what those teams usually give up. That puts the estimated points for Tennessee against Alabama right at 10.9.
Alabama points:
- Tennessee scoring defense for the season: 25.7
- Alabama scoring offense for the season: 51
The Alabama scoring offense is most similar to the following prior Tennessee opponent(s):
- Georgia 38.5
- Georgia State 35.5
Note first that those two comps aren’t really even close to what Alabama’s regularly putting on the scoreboard. For whatever it’s worth, Georgia State scored 38 points against Tennessee, and Georgia scored 43. So, the Vols allowed the two best-but-not-very-good comps 109% of what they usually score. That puts the estimated points for Alabama against Tennessee at 55.6.
Estimated score: Tennessee 10.9, Alabama 55.6
From the perspective of Alabama
Alabama points:
- Alabama scoring offense for the season: 51
- Tennessee scoring defense for the season: 25.7
The Tennessee scoring defense is most similar to the following prior Alabama opponent(s) (FBS only):
- Mississippi 27.4
- Duke 23.2
- South Carolina 23.2
Alabama scored 59 points against Mississippi, 42 against Duke, and 47 against South Carolina, so against the three best comps, Alabama scored 200% (double!) of what those teams usually give up. That puts the estimated points for Alabama against Tennessee at 51.4.
Tennessee points:
- Alabama scoring defense for the season: 17
- Tennessee scoring offense for the season: 23
The Tennessee scoring offense is most similar to the following prior Alabama opponent(s):
- New Mexico State 19.6
- Mississippi 27.1
Alabama allowed Mississippi 31 points and New Mexico State 10, which means that, together, those teams got only 88% of what they usually get when they played Alabama. That puts the estimated points for Tennessee against Alabama at 20.2.
Estimated score: Alabama 51.4, Tennessee 20.2
SPM Final Estimates
When you throw all of that in the stew and crank up the heat, here’s what’s for dinner:
SPM Final estimated score: Alabama 53.5, Tennessee 15.6
SPM Final estimated spread: Alabama -37.9
SPM Confidence level: 8.4
That 8.4 confidence level is just below the confidence range that the SPM favors the most this season, meaning this is not one of its favorite picks this week.
Eyeball adjustments
The SPM is not designed to account for things like player substitutions or teams navigating learning curves associated with youth or new coaching staffs, and Tennessee has all of that. For that reason, I’m expecting the SPM to adjust to improvement from the Vols but to be slow in doing so. It will come, but when we’re playing a good team like Alabama, the SPM’s not going to look at the most recent games but at the most comparable opponents, and for Alabama, that’s going to be Georgia and Florida. Tennessee seemed like a different team against Georgia than against Florida and is probably becoming a different team each week as it traverses the learning curve.
Still, I have a nagging suspicion that Mississippi State is just not very good and a nagging certainty that the Alabama offense is a machine lethal to defenses. So, although I sincerely believe that the Vols are improving, I’m giving a healthy dose of deference to the SPM against this week. I think the SPM is a little high on Alabama’s points but pretty close on Tennessee’s. With that, my eyeball-adjusted prediction is Alabama 48, Tennessee 17, a spread of 31. I guess that means that I don’t think the Vols will cover the opening spread of -29.5, but that they will cover the current spread of -34.5.
Other predictions from other systems
As previously mentioned, the Vegas line opened with Alabama a 29.5-point favorite, and it is now 34.5. The over/under of 62 translates to something like Alabama 48, Tennessee 14. ESPN’s FPI gives the Vols a 2.6% chance of winning.
Bill Connelly’s SP+ likes Alabama 41-14, and gives the Vols a 6% chance of winning. That’s a spread of -27.4, so it’s taking Tennessee and the points.
SP+ cooled a bit more last week, going only 26-26 (50%) overall, but it is still an incredibly impressive 57% for the year. As I said earlier, our SPM has beaten SP+ the last two weeks but is still far behind at 49.07% overall. Above our confidence level, it’s 53.62% and within our confidence range, it’s 60.66%.
Bottom line
Both I and the SPM think the Vols fail to cover the opening 29.5-point spread this week.
What do y’all think?
I am guessing JG starts and we show some energy at the beginning of the game, but begin to fizzle out by the end of the first half (similar to Georgia). The second half is all Alabama. Agree we don’t cover the spread as Tide reserves are inserted. Go Vols!
hey joel, I’m a little confused by the confidence level. i had assumed previously that when you were mentioning stats “above” the confidence level, the spm was even more confident than your predetermined confidence level. but i think I’m understanding now that it is more of a range and both below and above the confidence range means the spm is not confident? more similar to a statistical bell curve.
Hey, Jayyyy. Thanks for asking. Confidence level is just the term I’m using to measure the difference between the SPM’s projected spread and the Vegas opening spread. So if the Vegas spread is -3 and the SPM’s projected spread is -10, the “confidence level” is 7. Looking back at historical performance, I’ve found that a subset of games with confidence levels above a certain number perform better as a group than the entire set of games. But I’ve also found that that’s only true up to a certain point. In other words, if the “confidence level” is *too* high, it’s… Read more »
And if anyone’s interested, here’s the group of favorite picks this week. First team name and spread is Vegas favorite and spread, and the second is the SPM pick and spread. Clemson -22.5 Clemson -31.9 Northern Illinois -2.5 Miami (Ohio) -7.3 Oklahoma -33 Oklahoma -23.2 Texas -23.5 Texas -13.1 USC -9 Arizona -1.4 Georgia Southern -6.5 Coastal Carolina -4.2 Georgia -24.5 Georgia -13.7 LSU -18.5 LSU -29.3 Wake Forest -2.5 Wake Forest -13.4 Appalachian State -17 Appalachian State -29.3 Washington -4 Oregon -8.7 Navy -12.5 Navy -25.5 Louisiana-Lafayette -6 Louisiana-Lafayette -19.5 (Louisiana-Lafayette won last night by 17) Oklahoma State -3… Read more »