Continued from Harbaughffense Part II(a). Apparently I either write too much or embed too many graphics.

In this section, we're going to Harbaugh the Harbaughffense the way that Harbaugh Harbaughs the Harbaughffense.

Points Per Drive

Here we go. Let’s do it. Let’s measure that Harbaughffense the way that Harbaugh (reportedly) measures the Harbaughffense.

I haven’t looked at the stats yet…but I’m guessing that this isn’t going to be as good of news as we’d hope. We know that Ohio State has the #1 scoring offense kinda a lot. We know that we’re comfortable leaning on our defense to win games instead of scoring oodles of points.

Side bar, this frustrates me every single year, because if you don’t practice scoring a lot – and quickly – when you can, you won’t be able to score a lot – and quickly – when you must… and every single year there’s at least one game where I end up screaming at the TV “you don’t have all day to come back from this deficit!”

Long, methodical drives are a great strategy to build on a lead and break the other team’s spirit, but only if you actually pay them off with points. So how good are we at turning drives into points? What’s the gap between the Harbaughffense and the top performing team every year?

I’m also suuuuuper interested in seeing how the Harbaughffense performed in the NFL, where his total points were depressed (from where his offenses hover in college football) and the number of rushing TDs was abysmal­ by Harbaugh standards.

Well. There it is. At its best, the Harbaughffense can hang with the big boys. This has happened exactly 3 times. The rest of the time, there’s a material deficit in the Harbaughffense’s ability to put up points.

At Michigan, the closest that Harbaugh has gotten to a true “top offense” was…last year. But let’s not discount 2021 either: exceeding 3 points per drive is a HUGE accomplishment, because it basically means you can average a score (a field goal or better) every time your offense trots out onto the field.

It’s important to note that this data is “non-garbage drives” in FBS-to-FBS games only (the NFL data doesn’t do this kind of excising). So even though it’s still skewed by quality of opponent, there’s a lot less cupcake-skew than there could be.

And I’m done punishing myself by looking at 2017. Every metric we’ve looked at is telling me that 2017, not 2020, was Harbaugh’s worst year at Michigan. At least from an offensive point of view (and believe me, it was offensive). A combo of Drevno and Hamilton at OC, plus our Three Stooges at QB*, made for a disaster of a year.

* Ok, that’s harsh to Wilton Speight, John O’Korn, and Brandon Peters, and I’m sorry guys. Let’s look at context here. In 2016, Speight was definitely showing promise before the injury, then he got hurt again in 2017 (twice, IIRC). Brandon Peters got his head taken off too. Twice, IIRC.  QBs can be broken. O’Korn was always a long shot, and we’ll always have THAT game against Purdue, but he also…well…he tried against Ohio State. My god if the dude could have completed a pass we would have won. To his credit, he clearly saw the game the same way and felt terrible about it. I’m done ribbing them. I’m done ragging on 2017. I’m moving on. Until Part III of this analysis, where I’ll look, by year, at how effectively the Harbaughffense actually wins games.

Also, I think the NFL years are an interesting little tidbit here! Turns out the NFL isn’t (wasn’t?) a passing league; it’s a defense league.

Let’s take the next click down and look at whether drive length affects our scoring. We’re just going to compare the Stanford and Michigan Harbaughffenses here; FCS data isn’t available at all, and NFL data…well…I’d have to do the data slicing myself and I’m feeling a little lazy after having to piece together the San Diego offense for previous charts because the normal online databases are either wrong or incomplete...

“Long drives” are 80 yards-to-go or greater (i.e., starting inside the 20), “medium” are 60-80 yards-to-go, and “short drives” are 60 yards-to-go or fewer.

Ok this is a lot to take in, so let’s do some quick hits:

• Long drives
  • The Harbaughffense is BUTT on long drives, except for the Andrew Luck years. I don’t think people appreciate the Andrew Luck years nearly enough.
  • 2022 Michigan was the best long-drive team Harbaugh has had since 2010.
  • HUGE indictment of the 2019 Michigan team, this definitely supports the “could not sustain drives” narratives from that year.
• Medium drives
  • Once again, the Luck years were SPECIAL special. And once again, 2021 and 2022 show the Harbaughffense on the upswing.
• Short drives
  • Here we go! The 2022 Michigan team is the best short-drive team that Harbaugh has ever had, surpassing 5 points per drive, which appears to be the level that “elite” teams operate at.
  • It makes sense, right? If your defense and special teams give you an opportunity, how good are you at paying it off with points? The best teams are VERY good, and now Michigan is right there.

Overall…this really shows how close Harbaugh is at Michigan to recreating his Stanford success. But it also underscores how incredible those Stanford offenses were in 2009 and 2010.

What does it all mean??

It means that the Harbaughffense lives and dies on the rushing attack and the quality of the FG kicker. If it takes “NFL-ready Andrew Luck”-level performance to break open the passing attack, well…let’s not hold our breath.

On the other hand, when armed with a great OL and a couple great RBs, watch out because big running plays are inbound. Even so, all of this data suggests that Harbaugh’s offense, taken as a whole, hovers between “ok” and “good but not great” in any given year. Except for the Andrew Luck era, which is either “once-in-a-career magic” or “hopefully-repeatable magic given OL development and QB recruiting.”

I’m also seeing the bigger picture here, because you don’t need the best offense in the nation if you have a defense you can lean on. And Harbaugh has definitely had some really good defenses…but not to the point where an offense which can best be characterized as “a step or two behind the leaders” …is “good enough.” I’ll explore this more in Part III, which will be the first acknowledgement that Defense exists and is worthy of statistics. Respect will be paid. Overall, it’s simple enough, right? If you take Points per Drive scored and take away Points per Drive allowed, you’ll get a pretty darn good indication of whether the team is winning games at an elite level.

Hilariously, these stats seem to imply that the best way to beat an Harbaughffense is to beat them at their own game. Make them go on long drives. Reduce the number of drives to increase the leverage of any given play. Make them take field goals instead of TDs. Sit on a lead. Turns out that Harbaughffense is NOT good at long, methodical drives. Pretty decent at medium methodical drives. Elite at short methodical drives.

So do I still think that 2018 could have been great had it not been for Pep Hamilton? Mmm. Maybe. Yardage was definitely looking good, but the points-scored just doesn’t bear it out. Michigan was good that year, but comparing it to 2015 and 2016 (years with identical records), scoring was actually pretty well in line. 2019 was the disappointment year tbh. 2017 didn’t exist.

We’re getting closer to Stanford-level success at Michigan. Yardage is there and then some. The rushing attack is there. The field goal game was there-and-then-some for the past few years, but is now a question mark again. Last pieces of the puzzle are 1) receiving TDs, and 2) sustaining drives.

Join us again for Part III! Will it show that Michigan comfortably wins games despite a lack of gaudy scores? Will it show that Michigan lives one play away from disaster at all times? Tune in and find out!