Nebraska was decent on third down in 2019. Or maybe “better” is the better word to use here. The Huskers were better on third down in 2019 than they were in 2018. Nebraska ranked 50th in conversion percentage, up from 86th the previous season, and 32nd in conversions per game, up from 106th.
Of course, the problem with third-down stats is that they are just one piece of the puzzle. An important piece––it's one of the box-score stats, like rushing yards and turnovers, that can give you a good idea which team won a game without knowing the score––but still just a piece. Nebraska converted more third downs in 2019, at a higher rate, but it also faced the 19th-most third downs per game in 2019. In 2018 the Huskers ranked 104th in third downs faced per game, and in this case lower in the rankings was a good thing.
To get a better sense of Nebraska's offensive rhythm, and where it lacked last year, we need a little more context. We need first and second down and a way to compare all three.
Predicted Points Added, as calculated by collegefootballdata.com, offers a way to do that. It's simply taking predicted point values of a drive based on the scenario––say, first-and-10 from the opponent's 25-yard line––calculating the same value on the next play––maybe second-and-3 from the opponent's 32––and subtracting the former from the latter. If you have a positive PPA on that calculation (and you would) the predicted value of the drive has increased; a negative value (loss of yardage, sack, turnover) means the opposite.
On average in 2019, Nebraska generated .5761 PPA on third downs which ranked 33rd nationally. Scott Frost's first Husker team ranked 97th in the same category in 2018. That's good, but it was also a problem.
The 2019 Huskers basically had to generate a high PPA on third down because they were below average on first and second down, and what's interesting about that is that it was a total reversal from 2018 where the Huskers were better-than-average on first and second downs but slightly below average on third.
Here, it's easier just to look at it.
As that chart shows, third downs are indeed high-stakes plays. The average PPA for college football on third down was higher than that on second down which was higher than first downs. It's always that way. Has to be because third downs are often do-or-die; either the drive continues with a first down (and a new, higher predicted points for the drive) or it goes to zero when the ball goes back to the other team.
But a team can't get by just by being good on third down because there are, by definition, more of the other downs. In 2019, 45.6% of all offensive plays were on first down, 33.5% on second down and 20.9% on third down. Philosophically, if you were a coach with a magic wand to wave and say, "we're going to be the best here," where is "here?" Do you want the relatively low-value, high-frequency of first down or the high-value, low-frequency of third down?
If you want to see this thought experiment at work, I included the four Playoff teams from 2019 compared to the national average in the second tab of the chart above. While all four were significantly above average on all three downs––that's why they were Playoff teams––their relative successes are interesting.
LSU, the eventual national champion, generated a ton of added value on first down, dipped back closer to average on second down and then was proportionally better (for a Playoff team) than average on third.
You could have called Clemson a "second-down team" in 2019, which is a strange thing to say but it's where the Tigers were generating the most value. Oklahoma had basically a straight line from first to third down, the Sooners were good in first down, a little better compared to average on second and a little better again on third.
Ohio State, however, is perhaps the most interesting example. I thought the Buckeyes deserved the 1-seed in the Playoff because I thought they were the best team in the country. With the benefit of hindsight, however, Ohio State's PPA by down might've indicated one of the few cracks the Buckeyes had compared to the other Playoff teams. Ohio State took its biggest jump on third down. It’s a little like a basketball team confident in its ability to hit contested 3s late in the shot clock. It might work for a while, but when the talent is level it’s rarely the preferred path.
And, in the Buckeyes' semifinal loss to the Tigers, we saw that impact. From a PPA perspective, Ohio State "won" third down and second down, the latter being a nice win given how Clemson had succeeded there all season. But the Tigers dominated on first downs, generating a .3846 average PPA to the Buckeyes' .0239. That eventually takes a toll even if Ohio State is making up ground later in series. Ohio State had a better third-down conversion percentage in that game, but it also faced four more third downs. When the margins are thin, as they always should be in the 2-3 Playoff game, that can be the difference.
Now back to Nebraska's chart for a second. If you gave Frost the magic wand, what would he want the line to look like in 2020? You could start with just connecting the first down and second down PPAs from 2018 with the third down PPA from 2019. That would basically be the 2019 Oklahoma path, in trajectory if not magnitude, and might be good enough for eight or nine wins. Though if you wanted to get really particular about it, you'd probably ask for a higher PPA on first down and live with slightly diminished returns (but still good returns) later in a series. (The four Playoff teams ranked first, third, fourth and fifth in first-down PPA. Wisconsin ranked sixth, for what it’s worth.)
Why did the Huskers' first- and second-down production fall off last year? That's probably a post of its own, but you can start with a diminished run game that the Huskers leaned on more heavily than it did the year before. Nebraska ranked ninth nationally in rushing PPA in 2018 but dropped to 49th in 2019. Despite that, Nebraska ran the ball more often last season particularly on third down.
The Huskers ran the ball 60.4% of the time on third down in 2019. The only teams with a higher rate were Air Force, Army, Navy and Georgia Southern––i.e. option teams. Nebraska ran the ball 47.8% of the time on third down in 2018.
Why the jump? That's a tough question to answer succinctly. Maybe it was because the expected increase in quarterback efficiency wasn't there. Maybe that was due to injury, which led Nebraska to play three different guys there. Maybe the receivers were part of it. Football always makes tracing the error difficult in this way.
What should be clear as the start of spring practice looms is that last year's high-stakes game on offense was good for one more win and two fewer points per game. Like Nebraska's down-by-down performance in 2019, it's not really an equation that adds up.
The Grab Bag
- Jacob Padilla looks at the talent Nebraska basketball can potentially land in his latest Padding the Stats column.
- Kristian Hudson is only averaging five minutes per game for the Huskers, but her impact on the team is much larger.
- Greg Smith on the recruiting impact of the proposed transfer reform.
Today’s Song of Today