The weekly edition of my ongoing diary evaluating the B1G using the advanced stats of the Football Outsiders (http://www.footballoutsiders.com/stats/feiplus). Here is the chart:
Following suggestions, I've rotated the chart 45 degrees so that the better teams are at the top of the chart and worse teams are at the bottom. Teams that are better on defense are on the right (e.g., Michigan State), teams that are better on offense are on the left (e.g., Ohio State).
- Michigan is migrating towards the clump of okay B1G teams: Iowa, Northwestern, and Nebraska. These games are certainly winnable but they are looking less and less like gimmes (per the advanced stats, of course).
- The B1G ten is currently dominated by Ohio State, Wisconsin, and Michigan State.
- Purdue is worse than Akron.
- That Penn State loss is not looking any better.
What does cute baby have in common with our offense vs. MSU? They both suck!
Al Borges is easily the most hated man on the Michigan staff. Darrell Funk may be gaining some ground in that department, but he has a long way to go. Even Brady Hoke--the man who has delivered Michigan's best recruits in over a decade--is facing some heavy criticism.
I get it. Saturday was painful in so many ways. It's hard to watch your QB run play action from under center and get sacked almost before he turns around. It's hard to watch your team's championship hopes vanish in game that wasn't a contest. It's hard to know you can't beat the best, or even the very good.
Every story is made better with a villain. This is our national mode of thinking now--in politics, in business, and in sports. It's just so much easier to blame someone than to actually look at the complexity of an issue.
And to be sure, Al Borges does deserve blame. As does Funk, and, yes, even Hoke. Coaches are responsible for the product on the field. Period. Our product sucked with the fury of a black hole (yes nerds, I know they don't actually "suck") on Saturday, and it's natural to want bastards to pay for that.
But, just for a moment, let's look at how we could have attacked that MSU defense, instead of focusing on our complete inability to do anything resembling offense on Saturday.
This is right off of Brian's picture pages. It is a very typical alignment for MSU, although the field (don't want to argue about strong/free) safety is a bit deeper than usual at ten yards; it's often closer to eight. Michigan is in a pretty standard 3WR Shotgun look. Sometimes, that slot LB is actually a LB, sometimes it's a LB/S hybrid. Doesn't really matter.
How do you attack this defense? Let's examine a few possibilities:
- Quick pass/extended hand-off. Outside press coverage eliminates this possibility for the X and Z (outside receivers). You could look to the H (slot), but that LB or LB/S thingy is definitely close enough to tackle almost immediately. As soon as DG turns his head there and Funchess turns to catch the ball, that LB is in KILL! mode. He knows he has safety help over the top if he's wrong, so there's very little risk form him. He would attempt decapitation. Even if Funchess can shake him, that safety is there--and this assumes the TE blocked the middle LB. So basically, this option sucks.
- Slants/hitches/crosses. We're going to assume this is man coverage, since that's what MSU runs most of the time. First of all, it's terribly difficult to beat press coverage on a three-step drop, especially when the corners are as accomplished as Waynes and Dennard. This is made even more difficult by the fact that their corners are playing "inside technique," meaning they are a half-step inside the WRs, trying to force them to the outside. But let's assume, just for fun, that our guys beat their guys. The WR on the left side is now running right into that slot LB, who is reading the H. A common antidote to this would be to send the H on a flat route, attempting to pull the LB with him. That might actually work. The problem, though, is that MSU's safeties use a technique called "pattern reading." Basically, they are watching the first few steps of those slot/TE types, and reacting. If the slot/TE dudes don't stretch the field, the safety crashes down to try and help in coverage. MSU's safeties will make this ready quickly. A best-case scenario is a perfectly executed slant that gets thumped by the safety after a five or six yard gain. But this is a high-risk play: if the LB reads the QB's eyes and sinks, it could be a pick-six. I didn't mention the hitch, but it's real tough against press man, and the crossing routes are even MORE susceptible to LB INTs, which MSU does well.
- MANBALL. Run the ball, damn it! Well, even if we had a good O-Line, this is tough. A give to the RB means our six blockers are up against their six plus"two halves" in the box. That LB over the slot and the boundary safety are both on the edge of the box, ready to pounce on a running play. Even if we leave a backside guy unblocked, MSU will have numbers in the box very quickly, and our best-case is a three-yard gain (which we actually mustered a few times). BUT, even this will be inconsistent, because sometimes those LBs blitz and that safety crashes even before they know if it's run or pass, and in that case the slower-developing run from the shotgun is dead to rights. This, by the way, is why AB and Hoke want an under center game: your RB can immediately get to the business of running when the ball is snapped, rather than waiting for the snap to get to the QB and for the QB to handle the snap and then hand it off to a guy who is right next to him. From under center, the RB can (and does) start to move at the snap, and gets the ball closer to the LOS and with a some steam. It's not a huge difference, but it is significant and it is why RB runs are often more effective from under center.
- Option run. Yeah. We tried a few of these. You saw what happened. MSU makes quick, aggressive reads on almost every play. They blitz frequently, and they slant, stunt, and twist all the damn time. If even one of these guys gets free--which they almost always did against us--you're looking at a five yard loss. Even in a well-designed play (which Brian covered in the picture pages) the MSU defense collapses fast, gets off blocks, and gobbles-up your RB or QB. They get nine guys in the box (the safeties crash fast) as quickly as any team I've ever seen. Their DE's are also athletic enough to trouble the option on their own. All that said, a finely-tuned and perfectly-blocked option scheme would give them trouble, as it would anyone.
- Deep passing game. Without even the slightest doubt, this is where MSU is susceptible. IF Ohio beats them, it will be with big plays in the passing game, IMO, and perhaps some read/option stuff with probably the best running tandem QB and RB this side of Oregon. But, schematically, there is no doubt that this defense is susceptible to the deep pass, especially off of play action. You can get single coverage on your H (slot), your Y (TE), and your X and Z (outside WRs) against this defense. There are two problems with this: MSU's one-on-one coverage is really good; and MSU's blitzes and pass rush are even better than their coverage. I believe Denicos Allen is the best blizting LB in the country, and Bullough isn't far behind. Your QB needs a 7-step drop for these routes; your line needs to hold-up; and your RB probably needs to block (since that's one of the only ways to shut down the double-A blitz). We had open receivers against MSU. Other teams have also had wide open dudes. But when the QB's face is in the turf--or he's worried about his face being in the turf--it's awfully hard to read the defense and make the perfect throw required to beat the coverage (see Gardner's INT). Max protect should give you the time you need, but you still have to execute against an excellent blitz and stellar coverage, and MSU's safeties do a good job of reading three-man routes and helping their CB's where they can.
- Play action. For the record, I hated the play action calls. Having Gardner turn his back when his line can't block just sucks. But I understand why Borges calls them. MSU does read-and-react, and their LBs and safeties do hold a beat on play action. The run/pass conflict is there. That said, they are often blitzing a LB no matter what, which has the potential to blow-up your play before it gets started. So there's a bit of roulette there. But if you can block it, the opportunities for deep passes will be there, and pop passes to the slot from the shotgun or pistol (which we did) can be effective. That said, it's a slow-developing play that requires more blocking, and MSU is betting that you can't beat them in that department. Yet another reason to run play action is to set-up the run. Yep, you read that correctly. When a defense gets burned on a PA pass, they naturally slow their reaction to the run. This is why you might run a PA pass before you run the base running play. There is yet another reason Borges calls these PA passes and under center runs: he's preparing the team for the future. If we don't run under center until we're good at it, we'll never be good at it. Borges would get (rightfully) bashed for bringing out the 2015 team and expecting them to run under center stuff if they'd never done it before. Like it or not, it's part of our future, and we need experience doing it some, even if we suck at it.
TL;DR - MSU's defense is really good, and very difficult to attack. When you can't block their blitzes, you are left with few or no good options. This is not an excuse for the coaches--they need to get their kids to block. But the two best strategies against this defense require good, sustained blocking, and they are the only way to open up the running game.
We simply aren't good enough to execute against this defense. And that falls on the coaches just as much (if not more) as it falls on the players. Better play-calling couldn't have helped much.
Synopsis: Michigan's TOM for the game was 0.0 for the second game in a row and for the year remains at – 2 (– 0.25 per game) which improved slightly to #75. Turnovers were not a primary factor in determining which team won the game but the interception by Ramon Taylor with 2:18 left in the third quarter with the score just 16-6 presented a golden opportunity. I was freezing my butt off in Spartan Stadium with the rain/sleet pelting me and was shocked that Dantonio was still throwing the ball so much. I figured the only way Michigan had a chance to come back would be through turnovers. I thought with the two score lead, Sparty would just run the ball and punt. But, Michigan could not do anything with the turnover and the Spartans dodged a bullet (16-13 would have been mighty interesting).
Gardner did throw an interception but, under the weather conditions, I expected more turnovers. Gardner, Gallon, and the Team (that horrible snap) all had fumbles but all were recovered by Michigan. The Spartans did not have a fumble.
Michigan had 12 giveaways in the first 4 games and has cut that in half in the last 4 games with just 6 giveaways.
For the year, Michigan and Nebraska are virtually identical for turnovers. Nebraska has 2.0 giveaways per game ranked #82 and 1.70 takeaways per game ranked #55 with a –0.30 TOM per game ranked #81. But, Nebraska is –2.0 TOM for the last 3 games and is –1.5 TOM for away games. Michigan has a 0.0 TOM for home games and has a +0.3 TOM for the last 3 games.
National Rankings: All rankings include games between two FBS teams ONLY and are from TeamRankings except for forced fumbles which is from CFBStats. The four columns with *** show the best correlation to offense and defense (per Advanced NFL stats).
This chart shows Expected Points for various yard lines.
This chart shows the basis of EP calculations for each turnover.
Look, I'm not naive. This is a Michigan blog in the middle of football season, and I understand that the vast majority of you are here to read about football. That's ok, I come here for that too. Having said that, I'm guessing most of us feel a lot like this .gif (which Ace posted on Twitter Sunday night). Most of us probably have visions of the interior offensive line standing by, stationary, like cops by the side of the freeway (Hey, that guy's going by us way too fast. He's gonna cause an accid...see? See, that's what I was saying).
Forget about all that for a minute and get lost in the glorious charts of a Michigan team ranked #2 in the latest USCHO poll. That's right, there's a Michigan team that's climbing up the polls! If they continue playing the way that they have (read: willing to battle for puck possession, realtively intelligent passing, good forechecking) then there are going to be a lot of people who find a more enjoyable way to spend their Saturday nights once football season is over.
Friday, October 25 vs. Boston University (W, 2-1)
Saturday, October 26 vs. UMass Lowell (L, 1-2)
Friday, November 1 vs. Michigan Tech (W, 3-2 OT)
I am Mario and MGoBlueTV is the castle. Someday I shall breach those daunting castle walls, but last Friday was not that day. In other words, no charts.
Saturday, November 2 vs. Michigan Tech (W, 2-1)
- Winning the Corsi battle correlates with winning in the three games charted. It does not, however, correlate with scoring margin. The caveat is the small sample size, which is why I'm charting everything I possible can; we'll see how this plays out over the season.
- Michigan has only won once (against Boston College) when they have had a worse possession percentage than their opponent in the first period. Maybe there's something to be said for starting strong.
- Michigan has looked flat in the third period too often this year. It cost them against UMass Lowell and it could have cost them against Michigan Tech on Saturday. This "flatness" is reflected in the possession numbers above.
- I'm collecting power play data too, but my plan is to publish that around the midseason point in a longer post.
YARDS PER PLAY AND THIRD DOWN DIFFERENTIALS: AN UPDATE
In diaries in the past, we’ve talked about how yards per play differentials and third down differentials seem to be reasonably well correlated to a team’s overall success. For the most part, win these battles and you will likely win the game. It is in looking at these numbers for Michigan’s season to date that a couple interesting things seem to appear actually.
We’ll start with third down differentials compared to win / loss margin. The summary of the last eight seasons plus this season to date is below:
It looks as if 2011 onwards has something interesting happening, right? I looked at the overall R-value for the 109 games played since the start of 2005, and it is 0.706, which is perhaps not the best but respectable as football goes. One thing that is sort of intriguing is that, if you break the first two plus seasons under Hoke into separate groups, you find:
Of course, the seasons in isolation are smaller populations, so it has to be interpreted in that light, but the trend here of third down differential becoming increasingly tied to win-loss margin under Hoke is interesting, I would think.
Here is the graph of third down differentials and score margin for this season to date:
A bit of trivia here – that -35.97% third down differential is not Michigan’s worst performance on this metric in the last eight or so years. It is the second worst actually, but it is in the end merely one game. The worst (-36.97%) actually belongs to the 2008 Ohio State game, which ended a very painful season indeed. The chances based on data from the last 109 games of Michigan having such a deficit here? About 3%, so we do have that – it isn’t likely to happen often.
The yard per play differentials this season, when compared to score margin, tell a similar story:
The thing that strikes me is that we’ve only been outgained once, although we have had three games where the differential was a yard or less. In the Penn State game, which has the smallest positive margin, we lost by three. Perhaps that is an indicator that in those scenarios where the differential is close to zero, predictability goes out the window.
To add, here is the progression so far of average plays per game, yards per play and average margin under Hoke:
AVG. NO. OF PLAYS /GAME
AVG. YPP DIFF.
The table tells a small but intriguing tale really – we are averaging more plays but not quite getting as far with them. We’ve talked about it a lot on the boards, of course, but there it is. Interestingly, the average score margin has stayed somewhat stable from last year into this one.
I don’t really have a definite conclusion here. This team is a work in progress and will be for a bit longer perhaps. I simply find these numbers / beginnings of trends (I hesitate to call them “trends” right now) kind of interesting.
I am going through the seven stages of grief after DEBACLE (not that DEBACLE). I am somewhere between the stage three and four and trying real hard to make sense of it all.
One of the common excuse for our poor performance has been that our offensive line is so young that we should have expected (historically) bad performance out of them.
But is our offensive line REALLY that young? I have yet to see anyone quantify exactly how young our offensive line is compared to other top programs. So, I figured I'd do it myself, thinking that it will help me through my grieving process...
It just ended up making it worse.
I looked at the offensive line depth (from Rival) of every team in AP Top 25 and noted how young/old they are. I then assigned point values - one for frosh, 1.5 for redshirt frosh, 2 for sophomore, 2.5 for redshirt sophomore, and so on. Using this method, Michigan boasts an average line experience of 2.8. How does that compare to other schools?
Here is the complete list of Top 25 schools with 3 or less experience value.
|Texas A&M||OT||Jake Matthews||JR||3||2.7|
|Texas A&M||OT||Cedric Ogbuehi||RJR||3.5|
|Texas A&M||OG||Jarvis Harrison||RJR||3.5|
|Texas A&M||OG||Germain Ifedi||RFR||1.5|
|Texas A&M||C||Mike Matthews||SO||2|
As you can see above, there are 9 teams in top 25 with 3 years or less average experience. A few, including LSU and UCLA have offensive line that is SIGNIFICANTLY younger than Michigan. Amazingly, even with those very young offensive line with freshmen and sophomores, they have managed not to have historically bad offenses with competent rushing attack.
Based on this data, I think blaming our offensive line woes just on experience is not correct. It does not help, but lack of experience does not automatically mean that they will bad. When you have two NFL tackles, you should be able to perform at least average, if not better.