SOG Predictive Model - Final - #1 Alabama vs #18 Mississippi State

Lol that’s an awesome Gif

quote:

---

So what are the best-case and worst-case predictions?







No, I don't care if you have a life with responsibilities and such. You must feed my need for entertainment! Dance, monkey, dance!

---

LOLOLOLOL - Best case we win 50 to -7 worst case we win 18-9.

Yes I just made that up but it sounds reasonable.

quote:

---

State shoved the ball right up our gut last year.

---

They did, but they did it slowly and methodically. They never gashed us, they just kept getting 3 yards, 4 yards, 3 yards, 5 yards, 2 yards, 4 yards. Mullen was insanely patient and it worked to keep us off the field and give them a chance. They ran for 3.5 YPC and had 4.52 YPP. They also had 39 minutes of TOP compared to 21 for us. They were 8-15 on 3rd downs.

I don't think they'll be able to do that this year. Our front 7 is much better right now than it was that game last year (hell, our 2 MLBs were Holcombe and true freshman no idea what to do Dylan Moses). If Fitz can't beat us down the field a few times they are gonna be fvcked.

Basically, compare Rush and Pass Yards Per Attempt against opponents and compare that to the opponents averages vs P5 teams. Get an average of team performance vs opponent average for a +/- vs opponent average. Do that for both offense and defense for both teams. Figure out average # of runs and passes and points per yard, extrapolate out.

For instance

Say Alabama averages 10.7 Pass Yards Per Attempt v P5 teams. Our P5 opponents average giving up 8.9 Pass YPA. So we are +1.8. State gives up 6.7 Pass YPA ends up -0.7 vs opponent pass average on defense.

So I'd do (10.7+8.9)/2 then (1.8-0.7)/2 and net them together. Gives us an expectation of Alabama having 9.80+0.55=10.35 YPA. Then you figure out Alabama averages about 56% of plays are passes. Average out our average plays vs States average plays against. Multiple by Pass %. Then you get your # of passes and a pass yardage number.

Do that for rushing too and the other team. Then you have a Points Per Yard number that is multiplied by the yardage.

As I've said before, the Rush and Pass YPA generally come out pretty close. They are most predictable. The big variations usually show up when

(A) a team throws or runs more than usual
(B) turnovers
(C) big fluctuation in # of plays



Also, those numbers I used were just for example, they aren't the actual Alabama and MSU numbers.

I think the offense will be fired up regardless because they love to score.

quote:

---

Basically, compare Rush and Pass Yards Per Attempt against opponents and compare that to the opponents averages vs P5 teams. Get an average of team performance vs opponent average for a +/- vs opponent average. Do that for both offense and defense for both teams. Figure out average # of runs and passes and points per yard, extrapolate out.

For instance

Say Alabama averages 10.7 Pass Yards Per Attempt v P5 teams. Our P5 opponents average giving up 8.9 Pass YPA. So we are +1.8. State gives up 6.7 Pass YPA ends up -0.7 vs opponent pass average on defense.

So I'd do (10.7+8.9)/2 then (1.8-0.7)/2 and net them together. Gives us an expectation of Alabama having 9.80+0.55=10.35 YPA. Then you figure out Alabama averages about 56% of plays are passes. Average out our average plays vs States average plays against. Multiple by Pass %. Then you get your # of passes and a pass yardage number.

Do that for rushing too and the other team. Then you have a Points Per Yard number that is multiplied by the yardage.

As I've said before, the Rush and Pass YPA generally come out pretty close. They are most predictable. The big variations usually show up when

(A) a team throws or runs more than usual
(B) turnovers
(C) big fluctuation in # of plays

---

quote:

---

So do you average out just our total plays with their total plays against, or do you average the percentages as well?

i.e. do you average Alabama's 56% pass plays with an average percentage pass plays run against Miss State or not?

---

I average our average offense plays per game (total) with State's total average plays allowed (total) to get an expected total number of plays for us (and same for them).

I do not average out the % breakdown by run/pass with other teams, because that's generally much more specific to how an opponent plays. Our pass/run ratio is likely to stay relatively similar week to week, so it wouldn't make much sense to average our run/pass ratio with how Florida's run/pass ratio vs state worked out.

quote:

---

I do not average out the % breakdown by run/pass with other teams, because that's generally much more specific to how an opponent plays. Our pass/run ratio is likely to stay relatively similar week to week, so it wouldn't make much sense to average our run/pass ratio with how Florida's run/pass ratio vs state worked out.

---

I was just thinking along the lines of how a team is likely to attack a certain defense. So if a team has a weak secondary and 2 1st round interior d lineman, we would probably expect to see more passes.

I could be wrong, but my guess would be that teams pass more against us than their average because they're often trailing and that it's sort of a general belief that you have to pass against Bama.

At the same time though, I see why you don't want to just take that number because a team like Kentucky is going to run a lot regardless of who they play, and that would skew the average type of play run against Miss State's defense.

What would be interesting is if we took the average percentage type of plays run against a team above what that team's opponents run on average.

As example say Oregon plays Washington State. Washington State averages 70% pass plays, but against Oregon they run only 55% pass plays. If that trend of less pass plays being run against Oregon continues, we could probably conclude that coaches think they can get after Oregon's run defense.

quote:

---

I was just thinking along the lines of how a team is likely to attack a certain defense. So if a team has a weak secondary and 2 1st round interior d lineman, we would probably expect to see more passes.

I could be wrong, but my guess would be that teams pass more against us than their average because they're often trailing and that it's sort of a general belief that you have to pass against Bama.

At the same time though, I see why you don't want to just take that number because a team like Kentucky is going to run a lot regardless of who they play, and that would skew the average type of play run against Miss State's defense.

What would be interesting is if we took the average percentage type of plays run against a team above what that team's opponents run on average.

As example say Oregon plays Washington State. Washington State averages 70% pass plays, but against Oregon they run only 55% pass plays. If that trend of less pass plays being run against Oregon continues, we could probably conclude that coaches think they can get after Oregon's run defense.

---

Yea, I think for specific games that is worth checking out. It just kind of depends because certain teams are never going to deviate from a 60/40 unless they are trailing late or whatever.

And, the other thing is, certain YPP numbers would probably fall the more a team ran those plays.

Bottom line, lots of variables to figure out total yards or points. And lots of things that can't really be predicted. The thing I like is that it gives me a pretty good idea of expectation level for running and throwing the football and visa versa.

The specific points stuff is just kind of a guess (though almost always matched to Vegas' guess).