After reading other people's thoughts about last week's performance, I am confused by the willingness of some to give the defense a passing grade because they held Michigan State to "only" 20 points. Overall, I agree that the offense did a worse job than the defense, especially considering they are the stronger of the two units. That said, this was most definitely a team loss. I don't find compelling the argument that "if you would have told me before the game that we would hold MSU to 20 points, I would be ecstatic."
When I evaluate defensive performance, I look at how well the defense has accomplished two "universal" objectives. While some coaches may make field position or turnovers a high priority, I think the majority would find these two objectives critical...
1. Minimize Points Scored per Drive Except in rare end-of-game situations, you never want your defense to allow the opponent to score. The total points given up by the defense is a factor of how often the defense sees the field (total # defensive "drives") and how well the defense performs each time it sees the field (def pts allowed per drive). Other factors like field position and offensive performance (protecting the defense) contribute as well. Of these, the defense has the greatest control over the defensive points allowed per drive - DEF PPD. In my view, a defense that gives up 24 points in 12 possessions has performed better than a defense that gives up 24 points in only 8 possessions.
But what I really want to talk about today is goal #2... 2. Minimize Defensive Time of Possession (TOP) per Drive Quite simply, get the ball back to your offense so they can score. No coach wants the other team to hold onto the ball. RR has stated this is one of the key objectives of his defenses. Also, defense, heal thyself - get off the field. The defense has equal share in winning the overall TOP as the offense. You can't reasonably expect the offense to win the game if the defense is not giving them sufficient chances to succeed.
Metric: Defensive TOP Per Drive (DEF TOP/D). This measures the average time the defense leaves the opposing offense on the field per possession. Here is how defensive TOP/drive correlates to outcomes over the years 2001-2007, 2009. I excluded the 2008 season because no useful data can be extracted from that historically bad team other than "we sucked."
Avg. DEF TOP/drive >2.5 min... 11-15 (0.423) Avg. DEF TOP/drive 2.4-2.5 min... 11-5 (0.689) Avg. DEF TOP/drive <2.4 min... 46-5 (0.902) In Saturday's loss, the defense left MSU's offense on the field for 3.6 minutes per possession in regulation. This defensive TOP/drive was the highest I found in my research going back through 2001 (even in the excluded 2008). No other game was close. In the first half, the defense averaged over 5.5 minutes per drive. Even if they held MSU to only 10 points in that half, saying they did a decent job is only telling half of the story.
Here is the raw data:
DEF TOP/D (min) -- Games (losses in bold italics) 3.6 - 09 MSU (L) 3.2 - 05 OSU (L), 02 Iowa (L) 3.1 - 09 EMU, 02 OSU (L) 3.0 - 03 CMU 2.9 - 07 Wisc (L), 05 Wisc (L), 04 Ind, 03 Ore (L) 2.8 - 06 MSU, 05 NIU, 05 PSU, 02 PSU 2.7 - 06 USC (L), 05 Minn (L), 04 NW, 04 Tex (L), 01 Wash (L) 2.6 - 09 ND, 06 Ind, 06 OSU (L), 04 OSU (L), 03 USC (L), 01 WMU, 01 MSU (L) -------------------------- 2.5 - 07 OSU (L), 06 NW, 05 MSU, 03 Ind, 02 Wisc 2.4 - 07 Ore (L), 07 EMU, 07 Minn, 06 MSU, 05 ND (L), 03 Minn, 02 ND (L), 02 Wash, 02 Pur, 01 Wisc, 01 Tenn (L) -------------------------- 2.3 - 07 PSU, 07 NW, 07 Ill, 07 Fla, 06 Iowa, 04 MSU 2.2 - 07 AppSt (L), 07 ND, 07 MSU, 06 Minn, 06 Vandy, 05 Iowa, 03 NW 2.1 - 09 WMU, 06 BSU, 03 Ill, 03 OSU, 02 WMU, 01 MU 2.0 - 09 Ind, 07 Pur, 05 EMU, 04 MU, 04 Iowa, 04 Ill, 02 Ill, 02 MSU, 02 Minn, 01 Pur, 01 Minn, 01 OSU (L) 1.9 - 06 ND, 06 Wisc, 04 SDSU, 03 ND, 03 MSU, 03 Iowa (L), 01 PSU, 01 Iowa 1.8 - 06 CMU, 05 Ind, 05 Neb (L), 03 Hou, 03 Pur 1.7 - 04 ND (L), 02 Utah, 02 Fla 1.6 - 04 Minn, 04 Pur, 01 Ill 1.4 - 05 NW
Going to lead with the expected points rankings since the boss seems to like the metric and I am a suck up. Expected Points is one of the few metrics I use that is not opponent adjusted.
Expected Points - Offense
|Off||G||Expected Points||Actual Points||Diff||Drives|
Michigan is second in both expected points (offense getting good field position) and value added. The drives number is only qualifying drives, non garbage time non half ending kneel downs.
Expected Points - Defense
|Def||G||Expected Points||Actual Points||Diff||Drives|
Reasonable showing so far for Michigan although there is a big gap between the top and everyone. Michigan appears to be vying for the crown of best of the rest for the Big 10.
Offense - Rush+
Minnesota is an interesting case study here, last in the league in yards/game but 3rd in the league with value added. Haven't dug into it but a possible explanation is that they don't run in high volumes or for high yardage but they are effective in picking up first downs in short yardage plays, giving them a modest but still 3rd best value of nearly a point a game.
Offense - Pass+
A comment on the individual ranks yesterday was understandably surprised to see Juice at #5 in the QB rankings. We determined that a lot of it was due to the adjustment for opposition. You can see here, as well, when you take out what he does on the ground (included in QB ratings but not on team passing) and include sacks, Illinois is now dead last in the Big 10 in passing, even accounting for opposition.
Offense - Season+
Michigan third in the Big 10 in offense, going to face off with Iowa who is 9th.
Defense - Rush+
Michigan is down at 10th against the run, costing the team about 1.7 points per game.
Defense - Pass+
Iowa and Ohio State are clearly dominating here, both teams saving about a TD game due directly to their pass defense. Michigan is a solid fourth at a little over a FG a game benefit.
ST - All - Per Game
Michigan is a resounding first in the Big 10 in special teams, with the combined groups being worth nearly 2 points a game. Kicking, kick return and punting have are all at or near the top of the conference while punt return has been average/non-existent and kickoff coverage has been the lone weak spot.
Team - Season
The all-in numbers show Iowa, despite a below average offense is just above Ohio State in performance to date. The Big 10 is clearly lacking a dominant team this year as its top-ranked team only 17th nationally. Michigan is nearly tied with MSU for 4th overall and there is a clear separation between the top 5 and the rest of the conference.
The overall conference rankings show that its not just that the Big 10 doesn't have a top team this year, but that the conference as a whole is not as strong as past years.
No detailed Iowa preview this week
Will be traveling and without my computer, here is a quick summary of what the unit breakdowns look like.
Run Offense: -3
Pass Offense: -7
Rush Defense: even
Pass Defense: +4
Special Teams: +2
Home field: -3
Michigan 18 Iowa 26
The numbers aren't pretty so you can all it 17-24 or 17-27 but I've got it pegged right at the Vegas line and just below the under.
1) Do not vote until Sunday. Right now we can only accommodate so many people on the ballot (I believe that it is around 140), so please do not waste a vote by voting before all the games have been played.
2) Leave the bias out of it.
3) I promise not to change your votes, unless you are ranking teams in a way that is just crazy. It is hard to monitor over 100 people, so help me out with this.
4) You do not have to put your name on there, but it would be cool if you did so that I know who has been voting. I appreciate those that do and would like to give people a shout out if they have been doing really well.
5) Do not kill each other if you do not agree with the rankings. There will hopefully be enough people to balance out the few really high/ really low ranked teams.
6) Have fun with this. I think that our poll is a more true example of what is going on in the world of college football, because we can leave the pre-season stuff out of it. Vote based on WHO HAS THE BEST RESUME SO FAR. Do not try to predict the future.
Week 2 was interesting. We have been able to add some things, like most first place votes for our poll, and we increased our voters by a couple this week. Again, please vote, the more of you we get the better.
Here is the link- enter your information at anytime Link- http://spreadsheets.google.com/ccc?key=0AmMe19eG6tLTdHVQOXdWYzlyVHJENTJP...
Great Work Again by Bleedin9Blue.
OK on to the Poll for this week
|MGoBlog||Coaches Rank||AP Rank|
|#||Team||Points||PVS||PVS in ()||PVS in ()|
|1||Alabama (18)||397||1||3 (3)||3 (3)|
|2||Florida (3)||378||2||1 (1)||1 (1)|
|3||Texas (1)||372||3||2 (2)||2 (2)|
|4||Virginia Tech.||330||4||5 (6)||5 (6)|
|5||Boise State||328||5||6 (5)||6 (5)|
|6||LSU||325||7||4 (4)||4 (4)|
|7||Cincinnati||296||6||10 (11)||8 (10)|
|8||USC||279||10||7 (7)||7 (7)|
|9||Iowa||263||8||14 (11)||12 (13)|
|10||Miami||254||15||11 (21)||11 (17)|
|11||TCU||237||13||9 (10)||10 (11)|
|12||Ohio State||229||12||8 (9)||9 (9)|
|13||Oregon||216||17||17 (25)||13 (16)|
|14||Penn State||172||16||12 (13)||14 (15)|
|15||Auburn||169||NR||19 (NR)||17 (NR)|
|16||Kansas||145||20||15 (16)||16 (18)|
|17||Nebraska||121||23||22 (24)||21 (23)|
|18||Oklahoma State||95||14||13 (12)||15 (14)|
|19||Wisconsin||87||NR||25 (NR)||NR (NR)|
|20||Missouri||85||25||18 (23)||24 (NR)|
|21||South Florida||83||NR||24 (NR)||23 (NR)|
|22||Georgia Tech.||72||24||23 (NR)||22 (25)|
|23||South Carolina||51||NR||NR (NR)||25 (NR)|
|24||Brigham Young||46||21||20 (21)||18 (20)|
|25||Mississippi||40||NR||16 (18)||20 (21)|
LOVE 'EM (Avg. Spots higher than the polls)
#19 Wisconsin (6.5)
#17 Nebraska (4.5)
#9 Iowa (4))
#15 Auburn (3)
Wisconsin gets the most "love" this week from our voters. They crack our Top 20, despite being ranked 25 and NR in the two polls. At 5-0 it is tough to see them ranked below a 2-2 Oklahoma team in the AP and Coaches
HATE "EM (Avg. Spots lower than the polls)
#25 Mississippi (7)
#24 BYU (5)
#18 Oklahoma State (4)
#12 Ohio State (3.5)
Mississippi is an average Top 20 team, but we barely feel that they are a Top 25 team. Mississippi ranks in at 16 in the Coaches Poll, a full 9 spots higher than our poll
ON THE BRINK (Teams 26-30)
FWIW last week 4 of our 5 teams On the Brink made an appearance in the Top 25 this week. The most likely to appear next week are Stanford and Houston. They both play big road games this week. Stanford is at Oregon State, and Houston is playing a non-conference game on the road against Mississippi State. If Notre Dame were to beat USC, not a snowballs chance, they would assuredly be a Top 25 team. Also, the Maize and Blue could jump back in with a win over Top 15 Iowa.
HEAD SCRATCHIER (Team where people had no idea where to rank them)
Last week's was Oklahoma, as you can see they were well worth the ranking, as they dropped from 8 in our polls to being unranked. This week's are Ohio State and Oklahoma State. Both had some rankings in the Top 10, but were ranked in more than one poll in the low 20s. This is what you have when you have teams with one loss bouncing all over the board.
Source Material: Original Post, Definitions, 2002 Class, Problems, 2002 and 2003 Classes.
The generally low level of activity on the last couple McBean posts is because the season is upon us. Or the posts aren’t very good. My ego and the advice of wolfman81 are sure it is the former; it appears this project is ideal for the off-season when football filler is welcome.
The plan, therefore, is to shelve for a few months the massive posts that rank the players class by class in exchange for amusing ourselves with surgically precise mgoboard posts that ask about individual players. Each recruiting class has a good handful of players on the bubble, and it is my goal to increase the survey participation beyond the current group of hard core McBean aficionados. This will take the form of an mgoboard post that has a short preamble, a link to source material and a survey about a single player.
My goal will be to get to 25 or 30 votes on each bubble player. If the mgoboard post falls off the front page before that total, then I will repost it at a later date. This should allow us to get to a statistically useful number of opinions about borderline players for when we finish the project in the off season.
Here is an example:
Preface: In August, I launched the McBean Rating System and asked the mgocommunity to help me rank every Michigan recruit at the end of their career as a point of comparison to their initial rating (using Rivals)*. Jake Long was a four-star recruit coming in and, after his career, he was a five-star going out. Kevin Grady was a five-star recruit coming in and will likely be a three-star going out. This subjective rating system depends heavily on definitions designed to maintain the same relative number of players in each Rivals rating bucket. The goal of the project: to develop a “collaborative, ongoing post-recruitment rating system that will allow us to determine if, in the Rich Rodriguez era, perfect-fit three-stars are more desirable than random four-stars.” In other words, to answer the Pat White question once and for all.
Player: David Harris
All American: No
All Big 10: Senior year
Drafted: 2nd round (47th)
Bubble Question: Was David Harris a four-star or a five-star player at the end of his career at Michigan? (Please review the definitions.)
Good idea? Bad idea? Additional information needed? Board? Diary? Do you think this will be productive?
*Since that time, I can no longer take ownership of this project as there have been significant contributions from several mgobloggers; this is a mgoblog community project now.
This week I've introduced a new stat category that is hopefully easier to understand. Basically I just put in a logical operator in the Conversion Efficiency stat that returns a 1 if the score is positive and a zero if it's negative. No one likes negatives. This now makes it a true efficiency that can just be understood as the percentage of times that you throw at a receiver where he is contributing to converting the first down. And so I've renamed the old stat to Conversion Score.
Some things that stand out:
- It's kind of amazing that there are such few redzone passing attempts. This probably speaks well to the talent we have at running back. We're not forced to pass it down there very often.
- Nearly every player has a Conversion Efficiency above 60%, which speaks well about Tate's ability
- The notable exceptions are Mathews, who has gone three games without a catch despite getting 10 balls thrown his way (yes some of those were rain effected), and Grady (19). In Grady's place, Odoms and Roundtree have been getting more looks.
- Other than Carlos, our yac numbers have not been good. And his numbers are inflated due to catching 60 yard swing passes. But even without that he's got the third highest conversion efficiency.
Critical and objective observers of last weekend’s game vs Michigan State undoubtedly realize something sucky for Michigan—State was able to put together lessons learned from the previous four games to completely shut down the offense; crappy secondary and all. State showed the rest of the schedule a good first draft blueprint on how to beat Michigan: play keep away (only 10 possessions all game) and stop our bread and butter (the run).These aren’t novel observations and its said before every game, but now it has been done. Opposing Coordinators now only have to implement/tailor a proven game plan. So the imminent question now is, can Iowa execute this plan, and more importantly, is there anything we can do about it?
This is a pretty straightforward proposition and right in Iowa’s wheelhouse: out execute and play ball control offense. The never-ending story of a drive in the first quarter vs. MSU was absolutely huge., soaking up about 2 drives that Michigan and MSU’s opponent normally have in the 1st qtr. A lot of that had to do with exploiting the weaknesses of our defense. At QB Cousins is better, but Stanzi is probably adequate. In the running game Iowa definitely gets the nod; especially when you factor out Cousin’s scrambles last game. Stanzi can at least match what Cousin’s did on the ground against us. MSU probably has better overall receivers, but against our secondary that doesn’t matter much. Boil it all down to: yes, Iowa can play keep away from our offense if we let them.
Iowa most definitely has the ability to shut down a running game. Our guys did not have their best game on Saturday and Iowa has routinely made opposing running backs under perform this year. I adopted a rating CollegeFootball13 came up with in a previous post (like a pass efficiency rating for RBs) then tabulated the relevant data for each of the FBS running backs Iowa has faced. Its basically a clumsier version of what the Mathlete has been doing in his By The Numbers series. I split out the data into two groups, vs. Iowa and vs. others.
Each of the four FBS running backs Iowa has faced this year have under performed against Iowa as compared to their games against other opponents. What’s more is when rated in this way, Carlos Brown and Brandon Minor aren't necessarily better than Reggie Arnold (Ark. St). Ouch.
As for our most recent performance, the most obvious culprit was overall offensive execution through three and a half quarters. MSU’s defense definitely played well against us but our guys did plenty to make them look better than they are. The O-line was out played for sure and State was able to take away the running game. But, when we switched to the passing game our guys were dropping balls left and right, thus preventing any kind of momentum from getting started. Once we started catching balls and built up a head of steam at the end of the game, State had no answers. Again, I’m not saying State didn’t play well just that we didn’t. I’m taking it for granted that the offensive execution and concentration errors that killed us last game will be improved or eliminated for this week’s game.
As for Iowa most impressive victory, I think they benefited from things that are unreliable such as the blocked punt for TD and TERRIBLE QB play (Darryl Clark: 38% completion, 3 INTS). Penn State had four turnovers in the game. This is not to say that we can not possibly turn the ball over; just that turnovers are useful for the opposing defense.
So can we do anything about it? I think there are several things we can do. First, we can play defense like we did against MSU, but with better tackling. The second half specifically look more inspired to me. The play Graham made to force the field goal after the Ruby Punt Fiasco comes to mind. Also Kovacks‘s late game run stopping. If the D plays like that again, we’ll be in good shape.
The second thing we can do is short circuit the keep away strategy by coming out of the gate ridiculously aggressively. If we win the coin flip, we should take the ball then go for the jugular on the first play like Penn St. did vs. Iowa in week 4. How ‘bout opening up 5-wide shotty with Stonum, Hemmingway, Roundtree, Mathews, and Brown in the slot? Boom, Forciered. Then dial up a squib on side kick like MSU did vs. Notre Dame in week 3 on the ensuing kickoff. A failed recovery puts pressure on our D to come up with a stop from mid field, but I think they can handle that pressure especially if they know it’s coming early in the game. If we lose the coin flip, sell out for a quick 3 and out then execute the plan as previously stated. The overriding objective is to force Iowa into a shoot out and take the air out of Kinnick before anyone knows what hit ‘em. Getting Stanzi into a shootout would be a great start to the game.
The last thing, better execution on offense. No eff ups on snaps. Minimize ineffective blocks (see Huyge v.Jones @5:17 in the 2nd Qtr vs MSU). No dropped passes. None if this is magic and its all stuff we're perfectly capable of and have done before.
I think we have what it takes to beat the Hawkeyes, and we've done it before against reasonably comparable competition (Defense v MSU; Offense v Notre Dame). We just have to put it all together in a cohesive package of win.