This makes me feel much better about the upcoming game
*note: this is a reproduction of post on my blog: productivesweatpants.com
Indiana fans are pumped up about their 3-0 start to the season. And well why not? It's exciting to win. But more importantly they have revenge on their minds after last year's game in which Tate Forcier did Tate-like things: thing 1 and thing 2.
With a very similar start to the 2010 season as in 2009, we need to know things will be different this year, right? Well let's take a look at these two team so far.
I came across a couple of interesting statistics that I think will help give us a picture of the two teams. Instead of looking at the total and scoring offense numbers, or total and scoring defense numbers, we are going to look at how efficient the offenses and defenses are. We will also look at strength of schedule and Sagarin rank of each team to get an idea of what we can expect.
To find out how efficient an offense is we are going to take the total yards and divide it by points scored. It gives us a yards per point score. This will give us an indication of how efficiently an offense scores. It's good indicator of total team success.
Michigan - 13.64 yds/point good for 55th in the nation
Indiana - 10.09 yds/point good for 6th in the nation
That might be a little surprising for some given the performance of the Michigan's offense to date. But when we look at Time of Possession we see that Michigan is holding the ball longer at 31:37 TOP/game vs IU's 28:44. Also, consider that Michigan is averaging 25.8 first downs per game to IU's 22. And Michigan is averaging two more third down attempts per game than Indiana. It makes me believe that Indiana, being a passing team, is scoring on longer plays where Michigan, being a running team, is going on long drives.
Similar to the offensive efficiency number, we take the total yards given up and divide that by the number of points surrendered to give us a Yard per point given up, or a defensive efficiency number. In the converse to the offensive number the larger the number here means that offenses have to work a lot harder to score against that team. This is a great way for teams that play a bend-but-don't-break style (like Michigan) to measure up to other teams. For example: Alabama is ranked first with a 30.28 yds/point (they are number one by almost 4 yards. They're good). Let's look at UM and IU.
Michigan - 17.39 yds/point for 41st in the nation
Indiana - 17.5 yds/point for 39th in the nation
Virtually the same. I was shocked to see that Michigan although 93rd in total defense, fared so well by this metric. This my friends, gives me hope.
Looking at the efficiency numbers alone, it would seem that Indiana should be primed to take down the Wolverines. But, "Not So Fast My Friend!"
Let's take a look at the strength of schedule for each team.
Michigan - 66th ranked schedule
Indiana - 120th ranked schedule
That's right dead last. Indiana's schedule is ranked the worst in the FBS. "It's early, everybody schedules cupcakes." Sure I can give you that. But let's look at the particular flavor of cupcake Indiana has been munching on:
Towson - 1-3
OEff - 15.71 - 90*
DEff - 9.25 - 120*
SOS - NA
Sagarin - 195
Western Kentucky - 0-4
OEff - 16.69 - 99
DEff - 10.15 - 119
SOS - 93
Sagarin - 144
Akron - 0-4
OEff - 15.17 - 84
DEff - 12.02 - 106
SOS - 96
Sagarin - 175
Holy Schnikies that is awful. The only victory any IU opponent has was over Coastal Carolina in five OTs (for the record Coastal Carolina's Sagarin ranking is 185). Look at those DEff numbers. These teams are revolving doors. It's no wonder IU ranks 6th in OEff. Now, I will say IU pasted Towson, but they only beat the other two by 15 and 16 respectively.
By comparison let's look at Michigan's foes:
UConn - 2-2
OEff - 11.82 - 27
DEff - 16.26 - 51
SOS - 51
Sagarin - 69
ND - 1-3
OEff - 18.53 - 110
DEff - 15.63 - 58
SOS - 22
Sagarin - 64
UMass - 3-1
OEff - 15.23 - 84*
DEff - 15.17 - 67*
SOS - NA
Sagarin - 84
Bowling Green - 1-3
OEff - 11.67 - 94
DEff - 13.5 - 85
SOS - 77
Sagarin - 94
Of the cupcake variety? Maybe, although it is good for the 66th ranked schedule. UConn and Bowling Green have pretty efficient offenses. BG is ahead of Nebraska and Nevada. And, hell, at least all the teams have a win. Even our FCS snackcake is ranked higher by Sagarin than Indiana (85). The only team that is not ranked lower than any team on IU's schedule is Bowling Green and you saw that outcome.
So what has this little exercise taught us? A few things I think:
1) None of Indiana's statistical rankings stand up. I'm not sure you can take any of them seriously given the competition.
2) Michigan's defense efficiency is better than any of us expected. And they've been tested by some decent, offensively efficient teams.
3) Indiana's defensive efficiency is basically the same as Michigan's but against offensively efficient teams ranked near dead last in the nation.
4) I'm not nearly as nervous as I was before I did this little exercise.
*this is where they would rank among FBS teams although I didn't include all FCS teams.
This makes me feel much better about the upcoming game
Enjoyed this but a bit misleading because holding down BGSU would be one thing but doing against a backup QB is another and is worth mentioning.
...greatly effects the metric for offensive and defensive efficiency.
Michigan's defensive efficiency metric is inflated because we don't punt or turn the ball over much, therefore our opponents have a long way to go to score.
Our offensive efficiency metric looks artificially bad, because we haven't had great field position since our defense and special teams have fallen down on the job. However, our offense has proven that it doesn't matter too much where we are on the field, more yards to go is more Denard stats and more vertical stretch of the field.
So, you make some good points but I'd question the nature of the metrics you use. I don't know how you would do it, but I think you need to factor in starting position on drives into your metrics. Time of possession also plays a complex role when rating an offense.
By your current metric, a team that had the best defense and special teams in the world that started every drive from the opp 35 yard line could have great offensive efficiency if they scored on 1 of every 3 drives but turned the ball over one way or another on the rest. Similarly, a team that only gained yards on big TD plays that it only got as a result of good field position would also be considered to have good offensive efficiency.
It is an interesting study - I like punishing teams that get a lot of yards for no points, which makes bend-but-dont-break defenses look better - but it definitely has its flaws too.
TL:DR You are right, Indiana is fucked, Go Blue!
You're right. And truly you should look at the offensive and defensive numbers together. Our defensive numbers are better because our offense wins the field position battle for us 9 times out of 10. And our offense suffers because of our defense.
I'll search for some field position metrics and see how to work that into the fold.
My only problem with this analysis is the following assumption:
Offensive efficiency. . .it's a good indicator of total team success.
While this may often be true (who knows, the merits of the metric are largely beyond the scope of this diary), I'm still not sure it applies. If we have to drive 95 yards to score 7 points, our offensive efficiency looks pretty crummy. But that really doesn't matter if nobody can stop us from driving 95 yards...
mostly because I think it's not based in any sort of reality. Maybe it says something about field position (ie special teams) and turnover margins - but success comes from scoring points no matter how far you go to get them. That is not to say it doesn't correlate.
"it's not based in any sort of reality"
Someone around here has some sense.
Nittanywhiteout did an evaluation of "How Far Big Ten Offenses Have to Go to Score" (see link). This was done before the BGSU game, but Michigan is going the farthest to score and putting up a lot of points, which means that they are coverting on a lot of their drives deep in their own territory. I really don't think your yards/point metric is a "fair" evaluation of how successful Michigan's offense is.
The ND game killed our offensive efficiency as we marched to midfield several times and then punted. I'd say it was "that first game on the road" thing. Also due to M's horrible special teams, our starting field position has been pretty miserable...adding to higher yardage total. Furthermore, we haven't had any short fields due to turnovers (except the ND game but we choked on that drive).
The defensive numbers seem to carry more weight. These could keep going up as we continue to use the "bend don't break mentality" and try to let teams march down the field (bc they will) and then try to force FGs in the RZ.
Also, the ND game featured a high number of drive-killing penalties. There were several times when we gained more than 10 yards on a set of downs but had to punt because of a penalty.
Doesn't work if you don't inhale...Hold on to it...
hold on to it...
how about now?
You are right, when i read about Indiana's opponents and their rankings (oeff and deff) it really shows that IU is not a "legit" 3-0 team. Yeah they have the wins, but like you said your opponents only win came against Coastal Carolina.
It would probably be difficult to put together a schedule easier than Indiana's through weeks 1-4. Their SOS is ranked 200th by Sagarin. That's worse that 80 FCS teams. It's worse that half of the Ivy league. And they already had a bye week.
The fact that they're not top 20 in every offensive and defensive category should give us comfort.
There's no total points or total yardage in the calculation?
Offensive Efficiency is total yards divided by points scored.
Team A gets the ball at their own 20 ten times and scores touchdowns on every drive for efficiency of 800/70 = 11.43
Team B does not gain a yard for the first 59 minutes of their game, three and out every posession, but in the final minute, on 1st and ten from their own 5, their TE gets behind the covering linebacker, a freshman free safety reacts poorly to the ball and a 95 yd TD completion is the result. Efficiency of 95/7 = 13.57
Team B is the more efficient?
You want fewer yards per point. This would mean more of your drives are going for points. You want more yards per point for your opponent.
I believe the ideal metric would be points per possession, removing all drives which end d/t end of half/game.
our defense doesn't stop anyone, we get the ball back from kickoffs on the 20
offense very rarely has a short field, thus the higher yards per point
have to look at avg. starting position
Your example was great until Team B's lone touchdown. Make it a fumble recovery on the opponents 5, and an effeciency of 5/7 = 0.71, which makes it insanely better than Team A with this odd metric.
To the OP: I admire your effort and dedication, but please employ the use of the thing between your ears. Your offensive "efficiency" metric is flawed at best, but frankly is downright stupid. I have two words for you: Field. Position. I could go on, but i'm sure others will scold you for the lack of basic logic in your idea of offensive efficiency.
Defensively, though, you might be on to something. Measuring yards allowed versus points allowed will give us a very good idea of how well our "bend but dont break" defensive philosophy is working out.
Side note: The offensive metric could become slightly more useful if you only used it when talking about plays in the red zone, or on plays on the opponent's side of the field, but still probably not too useful.
EDIT: This was meant to be a response to Gene
Hate to be a nay sayer - but I am concerned, always have been and always will be when we face a Big Ten foe.
Our defense is my concern -as I am sure it is for all of you.
Keeping my fingers crossed.
I love that we are all concerned about IU. The fact that I've never worried about IU In the past, coupled with the fact that I am a little worried this year makes me feel strangely good about this season. That, in turn, worries me. Man, I'm mental!