Examining Notre Dame’s Offensive Efficiency

Earlier this season, NDTex wrote a fantastic piece that examined the 2012 team and the concept of offensive efficiency. Now that we are much further into the season, and our offense has been under quite a bit of scrutiny (mainly because Oregon and Kansas State are offensive teams) I wanted to look at the Irish offense and see how we are progressing.

For a reminder, here is a little background on offensive efficiency and how it is calculated:

For this exercise, I focused on something that came up in one of Brian Kelly’s pressers last season. In it, he mentioned that gaining loads of yards was worthless if you didn’t have any points to show for it, citing the goal should be a TD every 80 or so yards gained. Using this metric, I took a look at the first three games from the last time we started 3-0 (2002), our last two BCS seasons (2005 and 2006), and last season in comparison to the start we’ve seen thus far in 2012.

Using Kelly’s baseline of wanting a TD every 80 yards, I took each game from the aforementioned years, and applied a simple formula:

[Points Scored on Offense] / [Number of Yards Gained] = Offensive Efficiency

And to clarify, “points scored on offense” includes only TDs scored by the offense (no kick/punt returns or defensive TDs) and FGs (as it is the end result of an offensive drive). So if we use Kelly’s baseline, for every 80 yards, we should see 7 points, leading to an offensive efficiency of 0.0875. Now that mark would be the “perfect game” of sorts, so if we were looking for a “par” score in this area, that would be a FG every 80 yards or a 0.0375 efficiency.

So within the range of 0.0375 – 0.0875, you have an offense that is actually functioning as expected. Fall below that range, and there are serious issues. Going above that means some very beneficial field position via turnovers and/or special teams play.

Here are the raw 2012 numbers:

…and here’s what they look like when you plot them out:

There are several things I took away from this information. First, it is obvious that we are slowly trending downward when it comes to offensive efficiency — 4 of our 5 most efficient offensive games were in the first five games played. Additionally, we see that offense (no surprise) is the main culprit for our home struggles — in fact, our defense is playing even better (yardage-wise) at home than on the road.

Also, while Tex is correct that an offense is functioning correctly when scoring between .0375 and .0875 in offensive efficiency rating, Coach Kelly named .0875 as an indicator of where he wants and needs the offense to be in order for him to be happy — not as a mark for ultimate success. It is very possible to have a rating over .0875, particularly when your defense sets you up with good field position regularly. Bottom line, the fact that the offense is averaging out at .0633 indicates that they are not meeting Coach’s expectations.

Out of curiosity, I decided to look at our opponents on the year and see how their points allowed per game would compare on a similar line graph:

Sadly, this didn’t reveal anything except exactly what we already know: Notre Dame’s offensive is remarkably inconsistent when it comes to scoring points. While the Irish handled Miami exactly as they should have (and conversely, struggled against BYU, the best defense they faced), games vs. Purdue and Boston College should have been a field day for the offense, while Navy and Oklahoma should have provided decent tests — none of those things happened. It is mind-boggling to see the ups and downs that the offense is having.

One of the things that goes without saying is that turnovers are an absolute killer to this efficiency rating, and fumbling twice in a game like Boston College hinders what progress the offense could be making. But signs of progress can be found — Everett Golson is working on his footwork and has even noticed that he has an All-American tight end at his disposal.

There’s two weeks left, and the Irish need to showcase a higher-powered offense if they want to turn some heads in the human polls. There’s one thing that this offense has shown so far this year, and that’s that we truly have no idea what we are going to get.

Powered by
  • NDtex

    A few things:

    1) While Kelly would be happy with a .0875, that’s a rather absurd mark to hit. If you take a look at my quoted text, I flat out call that a “perfect game”. There are too many factors to take into account to expect that (field position for instance).

    2) As you say, the biggest difference in the dips are turnovers. Pitt and BC were plagued with them. Michigan and Stanford were messes too. I would say that isn’t so much an offense getting worse, but rather that the Irish really need to protect the damn ball better.

    3) Sorry, but I have to. The lowest point is BYU. Who started that one? :)

  • https://twitter.com/#!/GrantlandX Grantland-X

    I knew the BYU comment with Rees starting was inevitable haha. They also happened to be the toughest defense played by the Irish, in terms of points allowed.

    The one thing I would also like to note is that both the Stanford and Pitt games offensive numbers were skewed upwards in terms of efficiency, with 3OTs each. Obviously you don’t need to gain as many yards (if any) to score since you begin on the 25 yard line. So really the decline in our offensive efficiency is really something that we should be worried about.

    And while you cite .0875 as almost flawless, Kansas State has beat that mark in 7 of their 10 games this year, and has blown it away in 5 of those instances. Oregon has beat the mark in 9 of 10 games this season. Reality is that .0875 truly isn’t a crazy benchmark, just one that Brian Kelly believes is conducive to success. Our offense is just inefficient right now, plain and simple. Hopefully we make some big strides, and fast.

    • NDtex

      I’d be interested to see what their average starting field position is on top of that. We already know that both of them haven’t exactly faced great defensive units.

      I’d wager in the rare games K-State faced a defense that number went down. And Oregon…well, they haven’t exactly faced a functional defense yet.

      It’d definitely be an interesting metric to take a look at the rest of CFB and compare ND. Might be a good project for the offseason when I get my ranking code/database working right.

      Either way, ND definitely needs some work, but I think as long as we reduce turnovers we are in rather good shape.

  • http://- Tim

    Another interesting metric to look at would be points per offensive possession. Especially since ND’s offense has been eating up a lot of clock lately. I’d be interested to see how this measures up with Oregon, Kstate, Bama, etc. And if you wanted to get really in depth break that down further depending on starting field positon. (Starting in own territory or opponent territory for example).

    • Big Red

      This. Would love to see how ND compares efficiency-wise to its fellow contenders.

      • NDtex

        I’ll keep this in mind. I’m off work next week and have a lot of free time on my hands to do research.

  • Davie Dave 40

    Along with Tim’s comment, any thought to adding time of posession into all of this. If my math and sources are correct, ND leads the country in TOP. I think holding on to the ball and grinding down to the red zone and getting 6 is pretty big deal.

  • trey

    What this data tells me is that Offensive Efficiency is a terrible metric to use when gauging final success and that the pollsters are(as we all pretty much knew already) morons for simply picking the best two offenses and ranking them 1/2 in the polls. I jump directly to two data points, fMich and Fredo: only a .01 difference in efficiency score yet produced decidedly different results outside of the W. Were you nearly as nervous last week as we were against the Skunks? Were you biting your fingernails expecting Chase Rettig to catch fire and light up our defense? Of course not. It was a nice and comfortable win on the road. Juxtapose that with the nerves we had facing Shoelaceless in almost an identical scenario. It doesnt matter how efficient your offense is, what matters is that you control the game youre in for each opponent. Same point can be made for the PU game compared to these two. This measure is useless in determining the final strength of the team.

    • NDtex

      I created a terrible metric? :(

      /kicks rocks

      • trey

        No, not that at all. As usual, you’re insight is astounding and enlightened. Im just saying that trying to argue that offensive efficiency is the biggest contributor to success is flawed. That’s why we are in the boat we currently sit, because the amateurs at the WWL think the best offense means best team.

        • NDtex

          I should definitely do a chart of the the opposing offensive efficient. Those numbers would be fun to look at.

    • https://twitter.com/#!/GrantlandX Grantland-X

      I don’t know if I would describe it as a “useless” stat — I mean Kelly himself referenced it as very important to him. Does it not tell the entire story for the Notre Dame team this season? Absolutely, but our defense is #1 in the country, and an outlier compared to college football teams in general. That masks a lot of problems that really do exist on the offensive side of the ball, in the same way Oregon’s offensive efficiency likely hides some real defensive shortcomings.

  • Beej

    Great read. I was looking at Football Outsiders just this morning – FEI Offensive Stats. Based on their calcs, Our Raw Offensive Efficiency (the unadjusted efficiency of the given team’s offense, a measure of its actual drive success against expected drive success based on field position) ranks 54th in the nation, or slightly above average. HOWEVER, once adjusted for opponents’ defensive efficiency, we rise to SIXTH IN THE NATION. This blew me away. Through all the trials and tribulations you note above, this seems to suggest that our offense is considerably underrated, if not the most underrated offense out there. Other stats that jumped out at me: 23% of our drives are considered “methodical”, which means they last for 10 plays or longer, which ranks 5th in the nation. The strength of schedule our offense has faced thus far ranks 2nd in the land. And finally, our “Value Drives” percentage, the percentage of each offense’s drives beginning on its own side of the field that reach at least the opponent’s 30-yard line, is 48%, good for 28th in the nation. Looking ahead to SC (so sue me), we’ll be facing the 34th ranked (opponent strength adjusted) defense, considerably behind the likes of Michigan State (3), Stanford (4), Oklahoma (12), BYU (13), the list goes on. Wake’s D ranks 43rd. Go Irish.

  • Kegs

    Great stuff.
    As far as 0.0875 being an “absurd mark to hit” or a “perfect game,” this year’s golden standard for offense, Oregon, has a season average of about 0.0910. Indeed, their offense is absurd, but this is what we’re going up against in perception as we jockey for human poll votes. Kansas State appears to be about 0.902. I agree with the above comments about grading on a per-possession basis, as they may make our stats more comparable. The fact remains, we could sure use a blowout of Wake. A victory at USC is all we can hope for.