Can we agree to stop using "Total Offense" as our go-to metric?

CP87

Well-Known Member
Stop reading now if you have no interest in Offensive Efficiency metrics, this is a very TLDR post (@IowaLawWasRight -esque)

Do you know how to tell someone has an agenda when evaluating Iowa's offense? Their go-to stat will be "yards/game."

With Iowa's style of play, which includes emphasizing time of possession (hence fewer overall possessions) and field position (hence ideally working with shorter fields), Iowa almost always does worse in yards/game than they do in the much more important metric of pts/game.

I was looking at data over the past 13 years (this is how far back "efficiency" metrics are available). Over that time, Iowa's national rank in pts/game is on average 16 places higher than its national rank in yards/game. Over the last 5 years (New-Kirk era), the pts/game national rank is on average 31 places higher than its national rank in yards/game.

So who loves to use yards/game? Nebraska fans. Because they can fools themselves into thinking that Frost is some kind of offensive wunderkind. Nebraska kind of stunk the joint up on offense this year, but they can still say, "We were 55th in yds/game, and Iowa was 99th." And they would be right. But what if we look at pts/game?

Nebraska was 72nd nationally in pts/game, Iowa was 89th (it was a terrible year for Iowa in the red-zone). So Nebbie still has the upper hand on that side of the ball. But, what about accounting for the fact that Nebraska has way more possessions based upon their style of play, and they turned the ball over like a sieve?

Well, to account for that, we need to look at Offensive Efficiency. Football Outsider is a great resource to use for this. They essentially look at how many points a team gets out of each possession, adjusting for starting field position and quality of the opposing defense. ESPN also has efficiency metrics, but they do not control for quality of opposing defenses.

If we look at Efficiency metrics, we see that Iowa outpaces Nebraska with a national ranking of 62nd vs. 80th (Football Outsiders) or 52nd vs. 67th (ESPN). Nebraska also ranked 96th in the country in turnovers per possession, Iowa ranked 28th.

Furthermore, we can see that the Nebraska fan-base's disdain for their D-coordinator is probably misplaced, as Nebraska's defensive efficiency was substantially better than their offensive efficiency (44th vs. 80th).

Don't get me wrong, in general it is better to have more yards than fewer yards, it's just that yards by themselves are so incomplete. On Iowa's schedule alone this year, the losing team out-gained the winning team 4 times (ISU, PSU, Minn, USC).

So hopefully we can catch up with basketball fans, who have long since eschewed pts/game for pts/possession metrics, and we can ditch "Yards/game" as a go-to metric to encapsulate offensive quality.
 
If you are interested, here is how Iowa has fared over the last 13 years in national rankings in pts/game, yards/game, and offensive efficiency (from Football Outsiders). I have also included the B1G rank in offensive efficiency in parentheses:

upload_2020-1-9_9-29-58.png


A couple things stand out: BF is slightly better than KOK in efficiency, roughly equal in pts/game, but substantially behind in yds/game.

GD is substantially behind both BF and KOK in efficiency and pts/game.

BF regressed heavily in 2019. That ranking actually jumped about 10 spots after the offense scored 5 TDs in the bowl game. That backslide was mostly due to a very poor showing in the red-zone, which dropped their % of drives ending in TDs from a 60ish national rank the last two years to a 93rd national rank this year. Their redzone TD% of 57% (83rd in the country, and again this took a big jump after the bowl game) was substantially behind the last 2 years (69% in 2018, 25th in country; 65% in 2017, 44th in country; data from cfbstats.com).
 
Save your breath. Some people are too dumb to understand Iowa's strategy. It has been the same for 20 years. It is very simple: Limit possessions in a game. It substantially inflates the value of the defense and sacrifices the offensive stats. It sometimes puts Iowa in a position to lose to an inferior team like Northwestern, but it also regularly puts Iowa in one possession battles with teams it should probably lose to like PSU and Michigan. We have had mediocre o-line play the last few years, but if that gets shored up, the squad will be really good.
 
It's like I always say, I don't care if Everytime we start a drive it's at their 10 yard line. 50 yards equals 35 points..... I'm ok with that.
What is important to me is the eyeball test. If everyone is on the same page, if things are smooth. You can't take that. Think about it this way, everyone praises a wr for making amazing catches, but nobody considers that if the ball was placed better the catch wouldn't look so great. Same with blocking and a long run gets busted.
When everything is smooth and flowing you end up with games like the OSU one.
 
If you want to dig further into the data, a few of the metrics really tell the story of how Iowa wants to play:
  • OFD (essentially % of drives for which the offense gets at least one 1st down, important for time of possession and keeping the defense off the field)
  • OBD ("busted drive rate", % of drives that gain 0 or negative yards, which is obviously a killer for field position)
  • OTO (turnover rate, % of drives ending in a TO)
These are basically "protecting your defense" metrics.

The other telling metric is OED, or "explosive drive rate." This is % of drives that averages at least 10 yds/play. Iowa tends to lag here, as choosing the higher risk calls to create explosive plays increases risk of 3-and-outs, busted drives, and TOs.

Here is how Iowa fared in these categories over the last 13 years:
upload_2020-1-9_10-5-13.png

It should be noted that 2007 is really an outlier among those 5 KOK years. If we look at median instead of average, his numbers are much better (same goes for the data I presented in the post above).

Also, 2010 was just a monster offensive year. The defense was good, too (top 20 in defensive efficiency). It really came down to situational playcalling/execution, they just couldn't make the handful of plays that would have made the difference between 7-5 and 10-2 or better. Also, both those units really took a dive late in the season secondary to injuries (A. Rob, interior OL, and LBs) and lack of DL depth.

BF was horrendous at playing "KF Football" when he took over in 2017. Tons of 3-and-outs, tons of negative drives, not good with TOs. Overall the offensive efficiency numbers were okay because when they weren't famished, they were feasting (big outputs vs. ISU, Illinois, OSU, Neb). He has improved substantially at protecting the defense, but he has gotten worse at scoring points. Team performance has improved along the way.

Before the Wisconsin game, in the face of the offensive regression this season, I said that if the offense laid an egg vs. Wisconsin, it was probably time to give up hope on BF. They did lay an egg (more or less). But then he goes and reels me back in with solid game-plans to open the game vs. Minn, Illinois, Neb, and USC. I am not sure what to think now. But given that it is much more fun to be optimistic than pessimistic, I think I will use the improvements noted above, combined with an assumed return to characteristic red-zone performance, as reason to retain hope going forward. Hope is a good thing, maybe the best of things.
 
Last edited:
Okay, these things have been rattling around in my brain for the last few weeks, and I had to dig into some numbers and put fingers-to-keyboard to clear the mechanism. Now I will try to get some actual work done. Carry on.
 
Stop reading now if you have no interest in Offensive Efficiency metrics, this is a very TLDR post (@IowaLawWasRight -esque)

Do you know how to tell someone has an agenda when evaluating Iowa's offense? Their go-to stat will be "yards/game."

With Iowa's style of play, which includes emphasizing time of possession (hence fewer overall possessions) and field position (hence ideally working with shorter fields), Iowa almost always does worse in yards/game than they do in the much more important metric of pts/game.

I was looking at data over the past 13 years (this is how far back "efficiency" metrics are available). Over that time, Iowa's national rank in pts/game is on average 16 places higher than its national rank in yards/game. Over the last 5 years (New-Kirk era), the pts/game national rank is on average 31 places higher than its national rank in yards/game.

So who loves to use yards/game? Nebraska fans. Because they can fools themselves into thinking that Frost is some kind of offensive wunderkind. Nebraska kind of stunk the joint up on offense this year, but they can still say, "We were 55th in yds/game, and Iowa was 99th." And they would be right. But what if we look at pts/game?

Nebraska was 72nd nationally in pts/game, Iowa was 89th (it was a terrible year for Iowa in the red-zone). So Nebbie still has the upper hand on that side of the ball. But, what about accounting for the fact that Nebraska has way more possessions based upon their style of play, and they turned the ball over like a sieve?

Well, to account for that, we need to look at Offensive Efficiency. Football Outsider is a great resource to use for this. They essentially look at how many points a team gets out of each possession, adjusting for starting field position and quality of the opposing defense. ESPN also has efficiency metrics, but they do not control for quality of opposing defenses.

If we look at Efficiency metrics, we see that Iowa outpaces Nebraska with a national ranking of 62nd vs. 80th (Football Outsiders) or 52nd vs. 67th (ESPN). Nebraska also ranked 96th in the country in turnovers per possession, Iowa ranked 28th.

Furthermore, we can see that the Nebraska fan-base's disdain for their D-coordinator is probably misplaced, as Nebraska's defensive efficiency was substantially better than their offensive efficiency (44th vs. 80th).

Don't get me wrong, in general it is better to have more yards than fewer yards, it's just that yards by themselves are so incomplete. On Iowa's schedule alone this year, the losing team out-gained the winning team 4 times (ISU, PSU, Minn, USC).

So hopefully we can catch up with basketball fans, who have long since eschewed pts/game for pts/possession metrics, and we can ditch "Yards/game" as a go-to metric to encapsulate offensive quality.
First, I want to know, “did you evaluate where the Nebraska program is at” before deciding to compare programs?

Second, it’s almost like you’re saying that the metric you’re using equates to or has similar factors as “complimentary football.”

Things that make me go, hhhmmmm

Third, did you evaluate IowaLawIsWrong before comparing yourself to him.

Fourth, good job.
 
One more thing: 2011 was an under-rated Iowa offensive year. Their defensive efficiency was uncharacteristically worse than their offensive efficiency that year (62nd vs. 32nd), which led to an overall disappointing result (6-6).
 
One more thing: 2011 was an under-rated Iowa offensive year. Their defensive efficiency was uncharacteristically worse than their offensive efficiency that year (62nd vs. 32nd), which led to an overall disappointing result (6-6).

IIRC, 2011 was Coach O'Keefe's final year. JVB tossed 25 TDs against 7 picks. Plus Coker had almost 1400 yards and 15 TDs. The offense was pretty good that year. Then, the Greg Davis experiment began. 7 TDs against 8 picks in 2012. My God that was painful to watch.
 
Here’s a stat for yah. Fifteen consecutive years without a conference title. (drop mike and walk off stage).

That is a statistic, and that same stat applies to >70% of the B1G (everyone save OSU, PSU, MSU, and Wisc, the only 4 teams to have one a conference title over the past 15 years).

A statistic is hardly worthy of a mic drop. How about you do some analysis and draw some conclusions off your stats? Do you think Iowa should clean house based upon your statistic? If so, why?

Do you think a new staff is more likely to clear that bar than the current one? That would be the only reasonable argument, but I think it is a tough one considering the trajectory of the Iowa program (recruiting gradually going up, number of draftable players gradually going up, efficiency metrics gradually going up, team record gradually going up, final ranking gradually going up), along with the fact that we know KF is capable, he has done it before.

Or do you want to extract your pound of flesh, punishing KF and staff for failing to clear that bar? Fanbases that call for moves on that basis quickly end up in a coaching carousel that NEVER comes out positively for a program like Iowa. Alabamas and Texases and USCes can mess up over and over, and then fix things with the right hire. Teams that are outside of that top tier can do damage that is very hard to overcome.

I am not saying don't take chances. If there is evidence of stagnation (which there was pre-2015), then consider moving on, as long as you have a clear vision of how and why the program will improve with that move. But to make a move just to give fans their pound of flesh when the program is clearly trending up will get us in lots of trouble.
 
Save your breath. Some people are too dumb to understand Iowa's strategy. It has been the same for 20 years. It is very simple: Limit possessions in a game. It substantially inflates the value of the defense and sacrifices the offensive stats. It sometimes puts Iowa in a position to lose to an inferior team like Northwestern, but it also regularly puts Iowa in one possession battles with teams it should probably lose to like PSU and Michigan. We have had mediocre o-line play the last few years, but if that gets shored up, the squad will be really good.

I think most understand Iowa's (KFz's) strategy fine. They also have seen its limitations. I'd suggest the fewer possessions theory is somewhat of a myth. It's lost us as many games against "inferior" teams as it has won against the leaders. (looking at where we've won most of those, home field advantage would seem to be a big factor too)

Open it up more against inferior teams and put some points between us. Use ball control when it supposedly benefits us most and demonstrate the advantage. (which would arguably be even greater if teams thought they had to play honest after seeing us attack in other games and not load the box). Don't tell me the two are mutually exclusive and can't be done with the same personnel, we've seen it.
 
Open it up more against inferior teams and put some points between us.

There are a lot of years where I would honestly have to ask what it would mean to "open it up." With the exception of right now and 2009-2011, we have had some pretty big limitations on the outside most years.

Tressel was about the only guy I can recall in the modern era who could really evolve the offense from week to week exploit the current opponent. But he had OSU talent. One week he could roll out there and just pound the hell out of a team like Northwestern and the next week he could go 4 wide and run a hurry up to scorch Iowa or Wisconsin's base 4-3 by abusing the linebackers with slot guys. Unless you have really good personnel, you are pretty much wed to one team identity.

Iowa is definitely in a spot talent wise right now where I think they can be a lot more flexible on looks and packages at the skill positions, but that is not the norm in the KF era. And even now, there are some limitations with the line. I don't care who the o-coordinator is, if you can't block long enough for the QB to at least get a half second look once he sets his feet or to allow the RB to get at least back to the LOS, you are going to have major offensive limitations.
 
I think most understand Iowa's (KFz's) strategy fine. They also have seen its limitations. I'd suggest the fewer possessions theory is somewhat of a myth. It's lost us as many games against "inferior" teams as it has won against the leaders. (looking at where we've won most of those, home field advantage would seem to be a big factor too)

Open it up more against inferior teams and put some points between us. Use ball control when it supposedly benefits us most and demonstrate the advantage. (which would arguably be even greater if teams thought they had to play honest after seeing us attack in other games and not load the box). Don't tell me the two are mutually exclusive and can't be done with the same personnel, we've seen it.

Some might actually argue the opposite. Use the Goliath strategy (low risk, low variance) against the weaker opponents, and use the David strategy (higher risk, lower floor but higher ceiling) strategy against superior opponents.

I suppose it should probably be a combination. Try to minimize possessions against superior foes, but try to take calculated chances in the highest leverage situations.

Maximize possessions vs. inferior opponents by emphasizing tempo, but still play fairly close to your vest. Play with an acceptable amount of risk, but keep it conservative when a big mistake could cost you dearly.

Purdue is an interesting team to study. It seems like they play all-David-all-the-time. They have come up with some big wins that way. They also have some absolutely head-scratching losses.

There is probably an ideal combo somewhere that has to be informed by experience and gut-instinct. I don't think Iowa has necessarily found the right balance. They do really seem to play a lot better each year once they get to the "nothing-to-lose" point. Does something shift for them, or is it just random?
 
There are a lot of years where I would honestly have to ask what it would mean to "open it up." With the exception of right now and 2009-2011, we have had some pretty big limitations on the outside most years.

Tressel was about the only guy I can recall in the modern era who could really evolve the offense from week to week exploit the current opponent. But he had OSU talent. One week he could roll out there and just pound the hell out of a team like Northwestern and the next week he could go 4 wide and run a hurry up to scorch Iowa or Wisconsin's base 4-3 by abusing the linebackers with slot guys. Unless you have really good personnel, you are pretty much wed to one team identity.

Iowa is definitely in a spot talent wise right now where I think they can be a lot more flexible on looks and packages at the skill positions, but that is not the norm in the KF era. And even now, there are some limitations with the line. I don't care who the o-coordinator is, if you can't block long enough for the QB to at least get a half second look once he sets his feet or to allow the RB to get at least back to the LOS, you are going to have major offensive limitations.

Good points. I'm not suggesting you totally refit the offense every week but you can open it up some against lesser teams. There are many points along the spectrum between run,run,run into a 9 man front and 4 wideouts every play. (seems like when ever you mention more aggressive offense the knee jerk reaction is "WHAT, you think we should run a spread and throw the ball all over the place?!").

We've seen IOWA do it under KFz and it's worked. So it can be done. Other times he's stubbornly stuck to ball control when it was obvious to everybody watching the game...that it was benefiting the opponent more. (see NDSU, ISU, etc etc). Like it or not, the term playing-not-to-lose is on the money.

I'd suggest that some of the talent thing has been a self-created phenom. You don't throw downfield to your wide receivers you're not going to get great wide receivers to play for you.
 
I'd suggest that some of the talent thing has been a self-created phenom. You don't throw downfield to your wide receivers you're not going to get great wide receivers to play for you.

I like that last point. And in 2019, Iowa definitely did throw down the field (3rd most 25+ yard passes in the B1G behind Minn and Ind). It is also the reason I would like to see them use more tempo against the lesser teams. More touches, more "sizzle", keeps guys happy, catches the recruits eyes.
 
Last edited:
Before the Wisconsin game, in the face of the offensive regression this season, I said that if the offense laid an egg vs. Wisconsin, it was probably time to give up hope on BF. They did lay an egg (more or less). But then he goes and reels me back in with solid game-plans to open the game vs. Minn, Illinois, Neb, and USC. I am not sure what to think now. But given that it is much more fun to be optimistic than pessimistic, I think I will use the improvements noted above, combined with an assumed return to characteristic red-zone performance, as reason to retain hope going forward. Hope is a good thing, maybe the best of things.

Data is important because it can provide relative information. But, maybe the end of year performances are a measure of Brian's learning curve. Next year they'll need to lean more on the offense, I believe. They'll have the weapons in place, minus an experienced QB, it will be a challenge, but it's not as if Petrus doesn't know the system. 4 years is enough of a run to know whether he's got the goods to deliver consistent game plans, and I think more importantly, the ability to effectively shift gears on the fly when Plan A isn't working in any given week.
 
One thing I know is that we must use whatever metrics make junior look the best. If goal posts have to move year to year, that's what we'll do.
 
Good points. I'm not suggesting you totally refit the offense every week but you can open it up some against lesser teams. There are many points along the spectrum between run,run,run into a 9 man front and 4 wideouts every play. (seems like when ever you mention more aggressive offense the knee jerk reaction is "WHAT, you think we should run a spread and throw the ball all over the place?!").

We've seen IOWA do it under KFz and it's worked. So it can be done. Other times he's stubbornly stuck to ball control when it was obvious to everybody watching the game...that it was benefiting the opponent more. (see NDSU, ISU, etc etc). Like it or not, the term playing-not-to-lose is on the money.

I'd suggest that some of the talent thing has been a self-created phenom. You don't throw downfield to your wide receivers you're not going to get great wide receivers to play for you.

I think a lot of it is a talent problem, though. Most of the real head scratchers are early in the season, which makes sense in part due to the fact that is usually when we play cupcakes (Western Michigan excluded). Iowa's offense can be pretty bad early. I think Kirk is trying to get fundamentals down and ease new guys in. Guys don't have timing down yet. The staff doesn't know how the new guys will perform in game situations. As much as I hate to say it, there have been a lot of losses to bad teams where we just had a bad squad. Western Michigan was a good example. There wasn't some panacea of play calling that was gonna keep us from getting our ass kicked that day. We just weren't very good.

If Ferentz gets the line consistently back to 2003 or 2009 levels (2002 is an unattainably high goal) and we have Brian calling plays like he has in our decent games this year, this ball club is going to be really, really good.
 

Latest posts

Top