Blue Dawg's Weekly Sagarin Football Thread 2021

[Blue Dawg]

Sagarin wasn't really impressed by our ass-kicking at the hands of a Belch team.



I don't see any real surprises here. We often live here on the edge between winning and losing, so "8-4" can be very misleading. The percentage wins at 6.5-5.5 is much closer to my gut feeling going into this season, with the highest chance at a 6-6 season. This chart will show the % chance for us to have X wins total at each week in the season.



As usual, we have the weakest divisional opponents. I still expect ODU to beat us until proven otherwise.

[HogDawg]

Sagarin wasn't really impressed by our ass-kicking at the hands of a Belch team.



I don't see any real surprises here. We often live here on the edge between winning and losing, so "8-4" can be very misleading. The percentage wins at 6.5-5.5 is much closer to my gut feeling going into this season, with the highest chance at a 6-6 season. This chart will show the % chance for us to have X wins total at each week in the season.



As usual, we have the weakest divisional opponents. I still expect ODU to beat us until proven otherwise.

I don't understand Marshall being picked so high. #1 in CUSA? Surely not. They’ll have a brand new coach who has never coached a down in CUSA.

[Blue Dawg]

I don't understand Marshall being picked so high. #1 in CUSA? Surely not. They’ll have a brand new coach who has never coached a down in CUSA.

You say "so high" but they're #84.

In fact, having someone not in CUSA is probably a bonus.

[HogDawg]

You say "so high" but they're #84.

In fact, having someone not in CUSA is probably a bonus.

I see your point. But, if a brand new guy can come in here and win a 14-team CUSA conference title in his very first year, that would be the ultimate insult for the rest of us. I know Lane Kiffin did it, but then again, he's Lane Kiffin.

[inudesu]

I still expect ODU to beat us until proven otherwise.

Come on.

They sat out last season and didn't win an FBS game the season before that.

They upset us at their place once by three points years ago with a really good QB. Upsets happen in college football. They usually don't mean much past that game, or season at the most.

Results from other sports don't factor here.

There is no psychological edge to a team that we've only ever played once.

[inudesu]

I see your point. But, if a brand new guy can come in here and win a 14-team CUSA conference title in his very first year, that would be the ultimate insult for the rest of us. I know Lane Kiffin did it, but then again, he's Lane Kiffin.

I don't think many of the analytic rankings factor in coaching. And I'm not sure I blame them (be really hard to quantify - I've seen some efforts, but it's not great).

This stuff is built off of returning production, recruiting rankings, rolling averages, etc.

Blue Dawg, do you know if there is anything in Sagarin for head coach? Surely not, right?

[Blue Dawg]

I don't think many of the analytic rankings factor in coaching. And I'm not sure I blame them (be really hard to quantify - I've seen some efforts, but it's not great).

This stuff is built off of returning production, recruiting rankings, rolling averages, etc.

Blue Dawg, do you know if there is anything in Sagarin for head coach? Surely not, right?

It's possible he puts something in. The algorithm is proprietary - so while I think after a few weeks any "opinion" is factored out, I think it's much more subjective early on.

[inudesu]

It's possible he puts something in. The algorithm is proprietary - so while I think after a few weeks any "opinion" is factored out, I think it's much more subjective early on.

I know Bill Conolly (of SP+) has used his rankings to look at things like "most likely to win in close games" and stuff like that, but I don't think he puts anything in the rankings. How could you?

[Champ967]

It's possible he puts something in.

twss

[Blue Dawg]

I know Bill Conolly (of SP+) has used his rankings to look at things like "most likely to win in close games" and stuff like that, but I don't think he puts anything in the rankings. How could you?

That's inevitably a proxy for "how much media attention does this team get"

[inudesu]

That's inevitably a proxy for "how much media attention does this team get"

It's been a minute but I think he had some pieces where he took his expected results (from his formula) and put it up against actual results. I'll have to look for it. Skip basically right in the middle.

[Blue Dawg]

It's been a minute but I think he had some pieces where he took his expected results (from his formula) and put it up against actual results. I'll have to look for it. Skip basically right in the middle.

There is a metric that captures that, I haven't seen it a lot but I include it in my table as Quality. This matches the expected margin against the actual margin. 90+ is a very good game.

Skip's average last year was 55. The bowl game was 27.

Tough to find much on it these days, as the big networks have largely moved towards the FPI models - which I wouldn't be surprised to find has a .isP5() toggle

[inudesu]

This is from a couple of years ago - so it's missing Skip's 10 win season (and the .500 season last year).

But it's more of an after the fact thing, specifically not factored into the rankings, it's the rankings and results compared to the coaches.

[Blue Dawg]

This is from a couple of years ago - so it's missing Skip's 10 win season (and the .500 season last year).

But it's more of an after the fact thing, specifically not factored into the rankings, it's the rankings and results compared to the coaches.

Interesting. Thanks!

[HogDawg]

This is from a couple of years ago - so it's missing Skip's 10 win season (and the .500 season last year).

But it's more of an after the fact thing, specifically not factored into the rankings, it's the rankings and results compared to the coaches.

Remind me again why Texas fired Tom Herman? He's literally the biggest overachiever on the list. He has 3 times the overachievement rating that his replacement --Steve Sarkisian-- has.