I'd argue you're picking the two least favorable variables to come up with your comparison group then. I'd bet that out of those 13 teams, only a few returned almost the entire running game (O-line plus RBs). There will be regression, but it will be somewhat cushioned.
So I'm going to admit I don't have a good intuitive understanding of your rating system.
Can you clarify the 85% figure?
MSU loses 6 of 11 offensive starters: 5 of 6 skill position starters but only 1 of 5 offensive linemen. (It's only 5 of 11 starters if you count Sims as a starter since he was listed as a co-starter for the the bowl game.)
MSU loses 68% of yards gained from scrimmage by backs and receivers.
Definitely rebuliding on the offensive side of the ball, but I think we've got more back than 15%. Again, I agree MSU is getting a little overhyped right now, but I don't buy the "cratering" prediction.
I don't disagree that MSU might a be somewhat overrated at the moment, but to say they're going to have "one of the worst offenses in the country" is over the top.
Football Outsiders' projections say the offense will be the 37th best in the country. (16th on defense, 24th overall)
They return a lot of experience on the offensive line, and Andrew Maxwell has as much experience in the program as you could ask for as a new starter at QB.
Michigan's average is pulled down by sophomores. MSU's is pulled down by freshmen. So there is a difference there, given the jump in efficiency normally experienced between the first second years of a player's career.
Haven't had a chance to look at this in detail, but there actually is a stat called defensive rating. It looks at the available individual defensive statistics (defensive rebounding %, block%, steal%--the first probably being the most important) and then basically divvies up the rest of defensive team performance (the portion of field goal defense not captured in block%, in particular) into equal chunks based on minutes played (if I'm recalling things correctly).
You'd have to read "Basketball on Paper" for all the details (review linked below). Unfortunately, I don't think anyone calculates and publishes individual defensive ratings at the college level.
Short of having those ratings available--which would still be a much less reliable indicator than the offensive ratings in terms of pinpointing individual performance--i'd be very hesistant to try to create a comprehensive individual defensive metric.
Finally, if you're trying to value a steal in terms of points prevented, I'd say the correct number is opponents' effective points per possession--i.e. points scored per possession on which the opponent doesn't turn the ball over. That's usually a number around 1.2-1.3. But to make the value comparable to a replacement-level player, you'd need a measure of how many steals a replacement-level player produces.
It's quite a mess. The beauty of PORPAG is that it takes statistics other people have done the heavy lifting on and turns it into a single, more easily interpretable number.
Also: Hard for me to believe 11W doesn't rank higher than that. They've gotta be ahead of TOC. Wonder if there isn't a technical issue with their numbers.
The numbers in my scatterplot are full-season numbers (pulled from the FO Almanac; I don't think they publish conference-only numbers). Doubt it makes a lot of difference (would make the most difference for Michigan, given the 4-0 non-con record), but I am compelled by my data wonkishness to point it out.
1) Potential champion game match-ups are a key consideration, but so is overall competitive balance. As an MSU fan, I'm certainly OK with your divisions, since we'd only face 2 of the 6 historical upper-division foes on an annual basis, while, say, Minnesota would have to play 4 such teams every year. But I don't think that's the fairest approach for the league overall.
2) See the comment to my post (which your link actually goes to) on past attempts to manufacture top-notch conference championship game match-ups (Miami-FSU, anyone?) . While there's more stability in college football program performance than in other major U.S. sports, there's still quite a bit of fluctuation in where teams fall in the standings. You've seen that with every program in the league besides OSU just in the last decade.
I'd chalk that up to small-sample-size randomness. When you only have 10-12 truly international-level players, they're not always going to be nicely distributed across a standard soccer lineup.
The positive spin, I guess, is that the upside on this team is higher than it was for the traditional U.S. lineups that were solid in the back but lacking creativity up front. As untested as the forward group is, the Donovan/Dempsey/Altidore trio is oozing with scoring potential relative to what previous U.S. squads have brought to the WC.
Grant Wahl, the most esteemed of all the extremely bald men with a lot of soccer experience, thinks it's Edu over Torres and Findley over Holden (against England, at least).
Including defensive stats wouldn't help Manny's case much. His defensive rebounding percentage in Big Ten play is down from 22.4% last year to 13.5% this year.
http://bit.ly/aRjago
The fact that the deal involves playing a MAC school every year isn't news. That was going to happen (as it does at just about every Big Ten school) regardless.
The road games are the news--and what's being spun for PR purposes. It it spin? Absolutely. That's what big organizations do when they announce financial deals. I don't buy the "we're saving the state's economy" line any more than you do.
Brian implied in the post, though, that MSU was somehow forced into doing this by a lack of negotiating leverage. That does not appear to be the case.
I actually thought Hoyer got a bit of a bad rap last season. Hard to be successful when you're constantly being asked to throw on third and long and rarely being given chances on first and ten.
I think the addition-by-subtraction theory holds in the long run--by midseason, optimistically. It's just a matter of how long it takes to identify the starter and get that guy comfortable as the leader of the offense. With three seasons of eligibility remaining for both Cousins and Nichol, I think Dantonio will take his time in picking the starter, potentially sacrificing a game or two this season for stability over the next two seasons (when contention for a Big Ten title is more likely).
Oh, and I like Nichol. If it's close, you have to go with the guy with more innate talent. With Ringer gone, we need a playmaker on offense.
Thanks for the kind words about our work. Looking forward to the first football season with tne new blogging team in place.
1) I suspect that MSU barely beat out Iowa/etc. in the voting. The fact they don't release the voting details (and only go 3 teams deep) makes the result see more dramatic.
2) The fact we don't have to play OSU was probably the tie-breaker for a lot of voters.
I'm also a soccer fan who's only emotionally vested in the USMNT. Would love to have a place to keep up with and converse about US team developments without having to wade through all the club-related stuff.
Recent Comments
A large portion of MSU tailgating occurs on grass, without killing it.
This is, of course, because we are primarily a farming/turfgrass institution.
I'd argue you're picking the two least favorable variables to come up with your comparison group then. I'd bet that out of those 13 teams, only a few returned almost the entire running game (O-line plus RBs). There will be regression, but it will be somewhat cushioned.
So I'm going to admit I don't have a good intuitive understanding of your rating system.
Can you clarify the 85% figure?
MSU loses 6 of 11 offensive starters: 5 of 6 skill position starters but only 1 of 5 offensive linemen. (It's only 5 of 11 starters if you count Sims as a starter since he was listed as a co-starter for the the bowl game.)
MSU loses 68% of yards gained from scrimmage by backs and receivers.
Definitely rebuliding on the offensive side of the ball, but I think we've got more back than 15%. Again, I agree MSU is getting a little overhyped right now, but I don't buy the "cratering" prediction.
I don't disagree that MSU might a be somewhat overrated at the moment, but to say they're going to have "one of the worst offenses in the country" is over the top.
Football Outsiders' projections say the offense will be the 37th best in the country. (16th on defense, 24th overall)
They return a lot of experience on the offensive line, and Andrew Maxwell has as much experience in the program as you could ask for as a new starter at QB.
I asked Mackinder on Twitter about that. Here's what he said:
Michigan did switch on some screens. Can't remember exactly, but I think Novak was "on" Kearney & Thornton on some made 3s.
Makes some sense as Thornton was at the 4 spot some late in the game.
Michigan's average is pulled down by sophomores. MSU's is pulled down by freshmen. So there is a difference there, given the jump in efficiency normally experienced between the first second years of a player's career.
Also: Trey Burke does not count as a freshman. :)
FYI: A week ago, the stupendous Dan Hanner calculated you guys would have been #14 in the nation if you only counted games against top-100 teams.
http://basketball.realgm.com/article/218239/At_Their_Best_Against_the_B…
Also points out the problems with restricting the sample to just games vs. quality opposition.
Haven't had a chance to look at this in detail, but there actually is a stat called defensive rating. It looks at the available individual defensive statistics (defensive rebounding %, block%, steal%--the first probably being the most important) and then basically divvies up the rest of defensive team performance (the portion of field goal defense not captured in block%, in particular) into equal chunks based on minutes played (if I'm recalling things correctly).
You'd have to read "Basketball on Paper" for all the details (review linked below). Unfortunately, I don't think anyone calculates and publishes individual defensive ratings at the college level.
http://spartansweblog.com/2008/07/08/book-review-basketball-on-paper/
Short of having those ratings available--which would still be a much less reliable indicator than the offensive ratings in terms of pinpointing individual performance--i'd be very hesistant to try to create a comprehensive individual defensive metric.
Finally, if you're trying to value a steal in terms of points prevented, I'd say the correct number is opponents' effective points per possession--i.e. points scored per possession on which the opponent doesn't turn the ball over. That's usually a number around 1.2-1.3. But to make the value comparable to a replacement-level player, you'd need a measure of how many steals a replacement-level player produces.
It's quite a mess. The beauty of PORPAG is that it takes statistics other people have done the heavy lifting on and turns it into a single, more easily interpretable number.
Inside the Hall is the biggest IU blog (basketball-only site). Looks like they'd be #4 on the list above (bumping my site down a notch).
http://www.quantcast.com/insidethehall.com
Also: Hard for me to believe 11W doesn't rank higher than that. They've gotta be ahead of TOC. Wonder if there isn't a technical issue with their numbers.
The numbers in my scatterplot are full-season numbers (pulled from the FO Almanac; I don't think they publish conference-only numbers). Doubt it makes a lot of difference (would make the most difference for Michigan, given the 4-0 non-con record), but I am compelled by my data wonkishness to point it out.
FWIW
1) Potential champion game match-ups are a key consideration, but so is overall competitive balance. As an MSU fan, I'm certainly OK with your divisions, since we'd only face 2 of the 6 historical upper-division foes on an annual basis, while, say, Minnesota would have to play 4 such teams every year. But I don't think that's the fairest approach for the league overall.
2) See the comment to my post (which your link actually goes to) on past attempts to manufacture top-notch conference championship game match-ups (Miami-FSU, anyone?) . While there's more stability in college football program performance than in other major U.S. sports, there's still quite a bit of fluctuation in where teams fall in the standings. You've seen that with every program in the league besides OSU just in the last decade.
Making a decision about a job offer that defies Delvon Roe's dad's gut-level prediction does not equal "doing a Billy Donovan-esque pullout."
I'd chalk that up to small-sample-size randomness. When you only have 10-12 truly international-level players, they're not always going to be nicely distributed across a standard soccer lineup.
The positive spin, I guess, is that the upside on this team is higher than it was for the traditional U.S. lineups that were solid in the back but lacking creativity up front. As untested as the forward group is, the Donovan/Dempsey/Altidore trio is oozing with scoring potential relative to what previous U.S. squads have brought to the WC.
Grant Wahl, the most esteemed of all the extremely bald men with a lot of soccer experience, thinks it's Edu over Torres and Findley over Holden (against England, at least).