Michigan drops 2 spots to #7 in S&P+
Last week Michigan was #5 overall, 4th in defense, 24th in offense and 42nd in special teams.
This week Michigan is #7 overall, 4th in defense, 27th in offense and 9th in special teams.
Overall S&P had a relatively favorable view of our offensive performance since we remained fairly stagnant despite only scoring 20 points. This probably has to do with the fact that we gained almost 400 yards on over 4 YPC rushing and 8 YPA passing against the S&P+ #31 defense in Northwestern.
OSU actually dropped a spot from #2 to #3 overall. Their offense is #3 but their defense is down to #23 and their special teams down to #50. PSU also dropped 2 spots down to #8. MSU is 21st overall with the 47th offense and the 17th defense.
Maryland is currently ranked 56th with the 76th offense, 50th defense and 12th overall special teams.
September 30th, 2018 at 2:02 PM ^
I am not surprised investors are "shorting us" in the S&P 500 after last night. SELL SELL SELL!!!
September 30th, 2018 at 2:16 PM ^
Mortimer, your brother’s not well...
Fuck him!
September 30th, 2018 at 2:42 PM ^
here’s you $5 bonus.
thank you Mr. Mortimer. Maybe I’ll go to movies...alone.
the $5 was from both of us!
September 30th, 2018 at 2:34 PM ^
Yeah, but their Done and Bradstreet shot waaaay up, since they iced that loudmouth...
September 30th, 2018 at 4:18 PM ^
That's why I buy index funds instead of investing in individual stocks
September 30th, 2018 at 8:24 PM ^
Fully expecting clarence beeks to comment here. Come on buddy.
September 30th, 2018 at 2:10 PM ^
We can't talk about statistics until we beat OSU and MSU!
September 30th, 2018 at 2:18 PM ^
No talking or posting until we beat Wisconsin, MSU, PSU, OSU, Alabama, Georgia, Clemson, and Oklahoma.
i just want to establish that the goalposts will always be moved.
September 30th, 2018 at 5:22 PM ^
Do all of those wins have to be in the same season?
September 30th, 2018 at 6:30 PM ^
In consecutive weeks.
September 30th, 2018 at 8:48 PM ^
In consecutive weeks on the road...
September 30th, 2018 at 5:23 PM ^
Also, I think you forgot Notre Dame.
September 30th, 2018 at 3:21 PM ^
This is just dumb.
September 30th, 2018 at 2:11 PM ^
Good summary. S&P+ is fascinating to me. I watch a game like last night, shrug, and then say well I guess it was a win. Emotionally I feel worse because I was expecting a blowout. S&P+ is only looking at data. So the data they mined says our offense was ok? Very interesting.
September 30th, 2018 at 2:13 PM ^
The offense was quite good statistically. It was just penalties and stalling in the red zone that almost killed us.
September 30th, 2018 at 2:22 PM ^
Basically it uses yards per play and a few other factors to tell you how a team performed and will perform going forward because that is a better predictor than points scored.
Bill C does 5 factor box scores that come out later today that include post game win expectancy and Michigan’s from yesterday will be above 80% because they statistically dominated the game.
September 30th, 2018 at 2:55 PM ^
Exactly. Some details of S&P+ can be found here, but 5 factors are looked at to determine the ratings. The factors are efficiency, explosiveness, field position, finishing drives, and turnovers. According to Connelly's numbers, "efficiency is by far the most replicable and least random aspect of football" so his success rate number (which is based on efficiency) is the biggest contributor.
In terms of explosiveness, I don't remember exactly how Connelly defines it but we had 3 runs of 15+ yards and 4 passes of 20 yards+, so 7 explosive type plays.
Neither team turned the ball over so that was a wash.
Our drive finishing wasn't great, 4 trips into the red zone with 2 FGs, but we got points every time down and got inside the 10 each time. Connelly defines that number by giving extra weighting to success inside the 40 so just getting inside the 10 and finishing with points each time is a net positive for us probably.
Field position we struggled with the first 3 drives but after that it was even or in our favor. That was the main reason our defense gave up points, so my guess is our defense didn't get overly knocked for the scores in his rankings accordingly.
From a box score/S&P+ view point, the reasons we didn't win by more were the penalties and the early field position issues. The penalties are indirectly addressed by efficiency and drive finishing while the field position is directly addressed and one of the negatives from that game. But on the whole, S&P+ looks at this as us outplaying NW relatively comfortably. In its eyes, it's just the nature of football that even with all that the score can be much closer than the box score would indicate. Which is why Memphis lost to Navy earlier this year in a game where they had a 99% win expectancy.
You can view that how you want. S&P+ obviously can't look at why some of our drives stalled (drops, incorrect routes by receivers, bad throws, bad play calling, etc.) and it doesn't know why we committed the penalties we did or how legit those are. It either ignores those reasons because it doesn't have the capability to use them, or it treats them more as random variables that don't correlate from week to week. So it may be inherently undervaluing some of our major issues, ones that if you watch Michigan you know are very concerning, and overrating us . But it could also just be valuing us exactly right and us being overly pessimistic and realizing the same thing is probably happening to every team, but we just don't watch all of their games.
September 30th, 2018 at 3:30 PM ^
I like your summary. I went and read his five factors and it mirrors just what you said. It was interesting to look at the graphs he presented in his description of five factors. The graphs show that other than a few outliers, it is pretty predictive of success.
I do think though that Northwestern is one of those teams that just breaks any analytical model. They seem to be able to operate outside of predictions because their team is so weird, if that makes any sense. I could see them beating Michigan State next week and then follow that up with a loss to Nebraska.
September 30th, 2018 at 3:43 PM ^
Yeah Bill C will even say that his method obviously isn't perfect. Some teams (like MSU and K State) tend to generally our perform their S&P+ projections and similarly some consistently under perform. I'm not sure where NW falls there but it wouldn't surprise me if the generally out perform their projection. Even now though, they're 1-3 but a top 60 team so still well regarded by S&P+.
Like I said, S&P+ can't tell us exactly why we're failing in some senses. It may be underrating our pass game issues, play calling issues and penalty issues. But when he comes out with his breakdown of offensive and defensive specific ratings after 7 games, we'll be able to see exactly where we stand in his system. 27th means we're a good offense but flawed, which I agree with. The main flaws in his system right now probably being drive finishing and explosiveness after this past week.
Also keep in mind he has pre-season projections still in his rankings until after 7 games are played. Each week they're phased out more and more, but until 7 games in we won't know exactly how our O has been based on this season alone or how OSU's D has done, etc.
September 30th, 2018 at 5:43 PM ^
He actually is keeping the pre-season projections in all year this year. He said that they make the model more accurate if there is some preseason projection left.
September 30th, 2018 at 5:01 PM ^
So true! Fitzgerald's teams drop weird games like Akron. But he has staying power at NW in part because he always gets a signature win or two every season.
September 30th, 2018 at 3:22 PM ^
S&P uses success rate, not YPP. Success rate is defined as 50% of yards on 1st down. 70% on 2nd. 100% on third. YPP is highly correlated with that obviously, but it's more accurate because a Ben Mason FB 1 yard touchdown run is more valuable than a 3 yard run on 1st and 10 from your own 25.
September 30th, 2018 at 2:30 PM ^
You need to lobotomize the first 3 possessions (Offense and Defense) from your memory. After that the game is only slightly frustrating; red zone offense could be better but other than that nothing much to complain about. You would walk away feeling NW was never in the game.
September 30th, 2018 at 5:03 PM ^
The defensive line is setting the tone right now. So much quality depth. Backups like Dwumfour, Paye, Uche, Hutchinson and Marshall are starting to flex their muscles. It'll get real interesting when Solomon returns. All hail Don Brown and Greg Mattison!
Not sure why a highly ranked DL (that has the academics as well) would go anywhere else. Well, maybe OSU, Bama or Clemson, but still. Our DL is the gift that keeps on giving.
September 30th, 2018 at 9:45 PM ^
If they care about academics - according to you - they may go to Alabama, OSU and Clemson ? What ? There is no equal to Michigan if a DL wants developed plus academics. OSU Alabama and Clemson aren't even in the same stratosphere academically. OSU actually proved via Urban fiasco that they don't come to okay school.
September 30th, 2018 at 2:39 PM ^
Reposting this from elsewhere - These were our drives yesterday:
3 plays, 1 yard (punt)
3 plays, -1 yard (punt)
6 plays, 21 yards (downs) - McKeon's drop
7 plays, 79 yards (TD)
4 plays, 42 yards (punt) - false start prevented us from going for it on 4th down
3 plays, 7 yards (punt) - holding penalty killed the drive (we actually gained 17 yards on the three plays)
11 plays, 69 yards (FG)
11 plays, 56 yards (FG)
3 plays, -6 yards (punt) - the phantom hold on Higdon
11 plays, 67 yards (TD)
6 plays, 16 yards (punt)
From the third drive on, Northwestern really did not prevent us from moving the ball (although they got two red zone stops). A key drop and penalties of varying degrees of legitimacy stopped us.
September 30th, 2018 at 2:58 PM ^
Our offense was okay. You’re getting caught up in the “we only scored 20 points factor.” Michigan was able to move the ball effectively against a team ranked in the mid 30’s defensively.
Some penalties hurt Michigan drives, but when you look at it we completed 62.5% of our passes at 8.2 yards per attempt and ran for 180 yards on over 4 yards per carry. That’s an efficient offense.
It wasn’t explosive or anything. It didn’t result in a lot of points here. But you can’t really look at the performance and say it was anything less than okay.
September 30th, 2018 at 2:13 PM ^
This is ranking is so ridiculous
September 30th, 2018 at 2:17 PM ^
Maybe, but I don't feel like there are a ton of great teams this year. We have arguably the best defense in the country, and we will be a tough out for any team.
If the offense can sort some issues out, we definitely could be a top 10 team this year.
September 30th, 2018 at 2:19 PM ^
I love when people have no concept of what it is measuring say things like this.
its not only not ridiculous it is consistently one of the strongest predictors
September 30th, 2018 at 2:25 PM ^
Right. It's a very good measurement of how good a team is, but people don't understand that it really isn't meant to be a be all end all determination of the rankings. In a vacuum, this tells you how well a team is playing, but often times the subtle nuances, such as random turnovers, home field advantage, fumbled punt snaps, will determine the outcome of a game.
September 30th, 2018 at 3:24 PM ^
55% against the spread this year. Consistent >50% every year. It's valuable. It's better than random internet posters feelings.
September 30th, 2018 at 2:20 PM ^
Seems about right to me.
September 30th, 2018 at 2:23 PM ^
"Objective analysis makes my brain hurt."
September 30th, 2018 at 2:45 PM ^
I wouldn't get too caught up in the absolute rating --- I find it more interesting to look at teams in tiers/groups:
Tier 1 --- Alabama. They stand alone. They're a full 5 points ahead of
Tier 2 --- Clemson, Georgia, Ohio State. All bunched within 1.5 points of each other, and nearly 2 full points ahead of
Tier 3 --- Michigan, Oklahoma, PSU, Washington. All bunched within 1.2 points of each other, and nearly 2 full points ahead of
Tier 4 --- Notre Dame. Nobody else particularly close to them. They're another 1.5-2 points ahead of
Tier 5 --- Auburn, Oklahoma State, UCF, Wisconsin.
Then everybody else from there.
Those groups feel about right to me.
September 30th, 2018 at 2:55 PM ^
Tiers is the proper way to look at S&P+. The only note I would make is that with Ian Book, Notre Dame seems more like a Tier 3 or even Tier 2 team. Wimbush probably hurt their offensive rating.
September 30th, 2018 at 2:56 PM ^
ND with their new QB might be tier 3 now. They now are looking more like a top 10 team.
September 30th, 2018 at 4:12 PM ^
Tough loss last night. Great environment and McSorely was other-wordly.
Too bad, it was there for the taking. I do question some of Franklin's decisions.
I'm sure the Penn State boards are as calm and rational as we are on this board after a tough loss. Or even after a win.
September 30th, 2018 at 6:46 PM ^
PSU boards are a fun place today. It's a fairly even split between (1) the Franklin believers and (2) the "Franklin just burned through 2 years of goodwill in 1 night" folk.
Here's a statistic: PSU is 1-10-2 ATS under Franklin when coming off of a straight up loss. I find that statistic a tangible sign of how he and the team haven't really rebounded well off of adversity or stress.
I feel that PSU's game vs. MSU in 13 days will be a bellwether for not just the rest of 2018, but perhaps the next 2-5 years of PSU football. Could be very much like the PSU/Minnesota game in 2016. FWIW, that game is one of the 2 ties in the 1-10-2 statistic above. It was darn near a loss, of course, except for McSorley's last minute of regulation heroics.
September 30th, 2018 at 2:57 PM ^
1) What everyone else said.
"It is not a résumé tool. If you’re looking for something transitive — Team A beat Team B and has more wins than Team C, so Team A should be ranked ahead of Team B and Team C — you aren’t going to find it here.
S&P+ is designed to track overall team efficiency.
It can be used to make predictions, similar to the analytics systems Vegas uses.
It is, at its heart, a tempo- and opponent-adjusted measure of what college football teams can most consistently do to win football games."
September 30th, 2018 at 2:20 PM ^
Wait, I thought our defense was 5th and we were 10th overall last week. Am I getting my weeks confused?
September 30th, 2018 at 2:29 PM ^
I posted the same info last week so I'm pulling last week info from there. It's hard to track S&P+ week to week data, the site will tell you what the change from the previous week's overall S&P ranking was but not anything else. So I can say for sure we were 5th last week, but the defense/offense/special team rankings can only be found from my post on this last week.
September 30th, 2018 at 2:27 PM ^
Don't want to start another thread, but can anyone explain to me how that last touchdown isn't an illegal formation with too many guys in the backfield?
September 30th, 2018 at 2:48 PM ^
"B10 refs are incompetent" should suffice as an answer to any officiating question whether in our favor or not.
September 30th, 2018 at 2:56 PM ^
Because they owed us a makeup call after killing the previous drive with that “hold” on Higdon.
September 30th, 2018 at 3:01 PM ^
Looked at the video: not completely sure. Beyond B1G ref incompetence, which is probably the reason, the only thing I could think is that they're saying Collins at the top of the formation is lined up on the line. The tackles on the line are lined up about a yard off the ball which is generally fine. Looking at the yard markers on the top of the screen I could make the argument that Collins is also a yard off the ball and thus on the line. He's covering up Bell there so Bell would be ineligible, which doesn't matter because that play was a run or screen to Collins, but that's the best I got.
September 30th, 2018 at 2:33 PM ^
We dropped all the way to 7th? The horror! Our season is over, may as well forfeit all remaining meaningful games
September 30th, 2018 at 2:51 PM ^
Not that it really matters where we're ranked right now in the poll but how do you justify our only loss of the season coming at Notre Dame by a touchdown and Stanford stays in front of us in the poll by getting blown out to that same team? And this on the heels of last week's game win if Oregon just kneels and runs out the clock Stanford loses that one
September 30th, 2018 at 3:11 PM ^
It's not the same team. The QB that beat Michigan has been benched for a dude that's putting up ridiculous numbers so far. Plus they have other playmakers back on offense. So no, not the same team.
September 30th, 2018 at 3:20 PM ^
Because Stanford beat Oregon on the road.