Henry was not the same kind of risk Chris Barnett was [Bryan Fuller]

Fliers actually a good thing?

You mentioned in the last UV that "If Michigan hangs onto 8-10 guys
they could add a few fliers and be fine. The guys they hang onto are
actually touted recruits instead of the mess that was Rodriguez's last
class."
It seems like a large percentage of the big names on the team today
were fliers in the Hoke/RichRod class and Hoke's first class. Names
that immediately came to mind as late offers that panned out better
than expected are Norfleet, Morgan, Taylor and Henry. I wonder if
taking a few more chances on last-minute fliers wouldn't pay off for
this recruiting class?

-Jon

There's a difference between late fliers you take after scouting them in secret for a while and the kind of late fliers Michigan took after Brady Hoke was hired and they needed to cram ten guys into Rodriguez's battered final class. Morgan was a Rich Rodriguez add after extensive scouting; Henry was a Hoke add after the same; Norfleet was a highly touted spread guy Michigan had room for on Signing Day; he was well known.

Here are the guys Michigan added after The Process concluded in early 2011 (minus Chris Bryant, because Bryant was on the verge of committing to Michigan for months beforehand):

• Chris Barnett (flamed out before fall camp)
• Frank Clark (check)
• Thomas Rawls (never played, now CMU feature back)
• Russell Bellomy (third string QB)
• Antonio Poole (pec injury forced retirement)
• Matt Wile (kicker)
• Keith Heitzman (backup to AJ Williams)
• Raymon Taylor (check)
• Tamani Carter (transferred after one year)

They got two players out of eight swings and they got one of those because Frank Clark went from 210 to 280 like guys who get drafted from MAC schools. That's not a great hit rate, and that hit rate was about as expected. Only Taylor, one of the two hits, had a recruiting profile even on the 3/4 star borderline. All others were fliers picked away from Vandy, Minnesota, Purdue, etc.

Now combine that with the rest of the class, which featured four more guys who didn't make it through year one (Greg Brown, Chris Rock, Kellen Jones, Tony Posada) and that's a 20 commitment class in a year you could have taken 25 that has way too many washouts. 

This year is different. A guy coming in at the same time Hoke did last year would only need to add four or five guys and the guys already in the class aren't particularly likely to flame out, because that's the thing Hoke has been terrific at. They would not desperately need the the late flier guys to work out, and that's a good thing because they would not be likely to.

It shouldn't matter in a class that looks like it'll top out at 15. So I'm just sayin' if it's January 1 and Michigan has just installed a new AD I wouldn't necessarily think Hoke is safe.

[After THE JUMP: anti-mascot concept art]

Fancy metrics re-introduction.

Can you put short descriptions of FEI and S&P in this week's mailbag? I've seen a bunch of misinformation and confusion on the board recently. It could be time for a helpful reminder on how to interpret these numbers.

-eschaton811ydau

All right. They're both advanced metrics that try to account for pace of play and schedule strength when ranking teams. FEI is drive-oriented. From the description on Football Outsiders:

All drives are filtered to eliminate first-half clock-kills and end-of-game garbage drives and scores. A scoring rate analysis of the remaining possessions then determines the baseline possession efficiency expectations against which each team is measured. A team is rewarded for playing well against good teams, win or lose, and is punished more severely for playing poorly against bad teams than it is rewarded for playing well against bad teams.

This means than any 75 yard touchdown drive that isn't in garbage time means the same thing, as long as it's against the same level of defense.

S&P is play-oriented. It's based on "success rate," primarily. Success rate varies by down but it's pretty intuitive. If you get five yards on first down that's a success. If you get five yards on third and ten it's not. I prefer FEI most of the time because I like the idea that a point is a point is a point no matter how you get there, but I do understand the argument that blowing defenses up consistently is more predictive.

Both spit out some weird results from time to time. I don't mind because standard metrics also do that and I like the ability to control for tempo and opponent. FEI also has a special teams component that's really useful for determining what bits of a team's kicking components are any good—its main problem is that return touchdowns are so rare and distorting that they throw things out of whack.

The main things to keep in mind are:

They are schedule adjusted. Since standard stats aren't if you finish 30th in something you're probably pretty good. Being 30th in FEI or S&P means you're about average amongst power conference teams. Michigan checking in at 67th in FEI is abominable, but all you have to do is look at #68 Florida to know that.

They are tempo adjusted. Surprised that Michigan's seemingly good defense is ranked a bleah 44th in FEI despite being ninth in total yardage? Don't be: we're amongst the slowest teams in the country. Meanwhile, Oregon's "horrible" defense is 100th nationally in yards per game… and 27th on FEI.

They dump garbage time. "Why is Michigan ranked at all then?" you waggishly inquire, you wag you.

They're not game based. This is good and bad. It's good if you're trying to use them to predict things; if a team ends up losing on some crazy stuff but wins a box score they'll generally be higher up than they would in a results-oriented poll. The bad part is that by discounting events that are generally pretty random they can miss teams like, oh say this year's Michigan team.

There is not much data. These systems do have a lot more input than the old dumb BCS computers that weren't even allowed to take final scores into account, but even seven games into a season there's a lot of wobble, and single very strong performances can overwhelm what looks like common sense. Arizona's currently #4 in FEI despite narrow escapes against UTA, Nevada, and Cal thanks in large part to their win over #1 Oregon. FEI in particular feels like it can overrate single games against top teams—IIRC Miami was way up on the offense list one year in a non-intuitive fashion, and the best I could guess was that one thunderous performance against VT was the reason.

The best course of action with these stats is to use them in conjunction with traditional stats and common sense. I didn't buy the Miami offense that one year but I do buy that Oregon's defense is a lot better than conventional statistics give them credit for. Etc.

Kill for Kill?

Obviously this came up in the press today, but I've been vaguely wondering for a while.

Why don't we take Jerry Kill seriously as a candidate for the nonexistent coaching opening?
Most importantly, he has succeeded 4 times in 4 places. He's 53 years old with 20 years as a head coach - good numbers. We could probably get him.

Why isn't he more noticed in general? Well, he's coached in small places, he isn't an aggressive showman, and seems kind of pleasantly/won't-get-arrested boring.

Aren't these good things? Aren't they exactly the below-market-value features we should be looking for? Is he the John Beilein of football?
Yeah, I went there.

Jeremy

The first and most important reason we cannot hire Jerry Kill is that it would be wrong to separate him from Minnesota and thus break up the closest match between coach and mascot in the history of college football*. There are lines men should not cross. This is one of them.

Kill does have a quality, Beilein-ish resume. He's been a head coach since 1994 at five different stops, finding success at Saginaw Valley, SIU, and NIU; he's also got Minnesota in great position for being Minnesota.

I'm not entirely sold, though. He has a Mullen thing going on with his wins. Last year's 8-5 record featured a win over #25 Nebraska and no other ranked teams; they played four horrible nonconference teams last year; the only quality nonconference game this year was a 30-7 shellacking against TCU. The difference: Mullen has been keeping his historically awful program's head above water much longer in a much tougher conference, and oh yeah he's got the #1 team in the country this year. Minnesota just beat Purdue by a point.

And then there is the seizure thing. After the Michigan win, Kill earnestly thanked a doctor from Grand Rapids for "saving his career." There was some discussion in the comments about whether it was fair to disqualify a guy based on that. I think it clearly is, because Jerry Kill just flat out said if things didn't get under control he'd have to retire. They are under control for now; the possibility of a recurrence is there.

If Kill had a truly gangbusters resume I would say it might be worth the risk. Since he's about on par with a bunch of other guys it's not.

*[Unless Ole Miss had a really racist coach for an uncomfortably long time.]

[jim mora playoffs voice] HOPE?

Hi Brian,

The last time Michigan football team beat both OSU and MSU was in 2003. Since then we've gone 8-12 against them (4-8 since 2008; soon to be 4-9, 4-10...). I can't recall any major FBS school did that poorly against its two major rivals within this 10-year period.

With that being said, what will be the next time Michigan beat both of them? Realistically I am looking at 2017. This is because, if we have a new head coach in two months, he ain't gonna beat MSU in 2015 since no Michigan HC ever beat MSU in his first year; and in 2016 both games will be on the road. So that is a whole freaking lot of despair between now and 2017.

Kefeng from Indianapolis

Despair? I will not despair if Michigan splits with two teams that are amongst the best in the league.

My despair goalposts are moving all the time. I no longer despair at the fact that we're 17 point underdogs to Michigan State. I despair at the possibility this state of affairs will not result in the swift excommunication of all adult-type substances involved with the impending face-punchin'. You have to dig through layers of tar to find my despair goalposts, and then actually kicking something through them requires an enormous drill, like an Ocean's 11 drill.

Also: basketball.

Media does not respond to stimuli

Hey Brian,

Was wondering your opinion on why Hoke is so, for lack of a better word, horrible to so many media members? Why does he choose to almost completely dismiss injury and other questions altogether as opposed to saying something as simple as "Player A is having some elbow pain, and we're keeping him out for precautionary reasons. Not sure on his prognosis yet but we'll keep you posted."?

It seems like in these types of positions (especially for someone who is obviously on the hot seat), where their perception is to some degree determined by media write-ups, that he'd want to be as respectful as he can.

Dan

It doesn't matter either way. Being super nice to the media didn't help Rich Rodriguez one iota, so to some extent they've brought this on themselves. Michigan was much looser under RR and the only thing that got him was guys in the department telling Snyder and Rosenberg which embarrassing documents to FOIA, plus avalanches of concerned columns about how RR was too mean to his players.

Hoke could spend every press conference throwing his own poop at the media and the only one who would notice is poor Nick Baumgardner. Hell, even after the incompetent handling of Shane Morris you had more local(-ish) guys piping up to chide fans for thinking Brady Hoke's a bad person—an assertion I literally did not see anyone with a platform make—than wondering if Hoke was too incompetent to be Michigan's coach.

And Michigan's done a standard job of answering questions without actually saying anything, so media members look petty if they complain. They either leave the beat as fast as possible or suck it up and get on with their jobs.

Thank you for the helpful label

Brian,

Why didn't we avoid the Noid?  Was it the handsome suit jacket that threw us off?

Andrew

Class of 2000

This is a mascot I would support for Michigan. It could be our anti-mascot. Everyone would boo it and throw marshmallows at it. The cheerleaders would shame it publicly and maybe hurl it into the goalposts. #AntiMascot4Michigan