Photo by Eric Francis
Nebraska Football

Hot Reads: Something to Think About When It Comes to QB Play

April 19, 2019
3,171

One of my big offseason projects has been to look at how the score in a game relates too turnovers. I've written a little bit about it to this point, but I'm saving most of it for the 2019 Hail Varsity Yearbook (subscribe today!).

Here's another little piece of that larger discussion––Adrian Martinez.

Nebraska's freshman quarterback threw an interception on 2.31% of his pass attempts last year. That's pretty good for any quarterback, but particularly a true freshman. Martinez's interception rate ranked 43rd nationally among quarterbacks with at least 300 attempts and fourth among freshman in that category.

But here's something else to think about. Based on my look at the data over the past three seasons of college football, interceptions are more likely when the team with ball trails. Makes sense. A team is trailing, needs to get back in the game, they throw more often and might take more risks with those passes.

Here's a general sense of the scale. Over the last three years of college football an offense has thrown an interception on 0.59% of all plays* when leading by 15 or more points and been picked on 1.93% of all plays when trailing by 15 points or more. The graph for interceptions as a percentage of all plays ticks up every step of the way as you move from having a big lead to having a large deficit.

*I'm using all plays because I'm also looking at fumbles and turnovers as a whole, neither of which are covered here.

The interesting thing about Nebraska and Martinez is that the Huskers didn't have a lead very often in 2018, just 30.8% of the time which ranked 91st nationally. Martinez threw six of his eight interceptions (75%) on the season with the Huskers leading or tied. It's the opposite of what you'd expect for an average quarterback.

There are a couple of ways to look at that. One, Martinez needs to value the ball a little more with a lead. His interception rate (again, on all plays) was higher than the expected average when the Huskers led by 15-plus points (the INTs that hurt the least), when they led by one score or less (1-to-7 points, the INTs that hurt the most) and when the game was tied.

Two, Martinez was good at avoiding interceptions when trailing. This is sort of a counterpoint to the above paragraph. If a QB throws an interception when up more than two touchdowns, the team can probably afford that. Up one score? Less so. But when Nebraska was trying to get back in a game, when the pressure is highest and the risk the greatest––and, again, that was 70% of NU's offensive snaps last year––Martinez was good at avoiding interceptions. To me, that seems like the hardest part.

Three, if you're just selecting the best possible scenario for Nebraska's offense and Martinez in the years ahead it's sort of inevitable that most of his interceptions will come when Nebraska is in the lead because it means Nebraska was in the lead a lot. In three years as the starter at Oregon, Marcus Mariota (a lofty comparison, I know) put up some crazy numbers. He threw 14 career interceptions, a pick on just 1.2% of his total throws. (Pretty crazy.) Nearly 86% of those 14 picks came when Oregon had a lead (sounds crazy) but that's because Oregon almost always had a lead. Just 16.5% of Mariota's throws came with the Ducks trailing (really crazy). He was just flat good at protecting the ball.

Martinez's first year in the system (as a true freshman) wasn't quite as good as Mariota's (as a redshirt freshman) in that regard, but Martinez was within striking distance and did it while trailing more than two-thirds of the time. Martinez's numbers, for a true freshman, get more remarkable the more you look at them. That makes areas of improvement tough to identify, but interception expectancy could be one of them.

Either his interception rate with a lead comes down a little or Nebraska has the lead a lot more often. Either would work.

The Grab Bag

Today’s Song of Today

 
×
Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.