The Defensive Line

The latest newsletter from KCSN Sports Data Scientist, Joseph Hefner.

It’s a truism that “football is won in the trenches”. Offensive and defensive lines can put their stamp on a game in ways that the other positions just can’t. Defensive lines can completely wreck an offensive game plan by dominating up front. Steve Spagnuolo’s defensive line did just that in the 2008-2009 NYG vs NE Super Bowl.

This week, we’re going to talk about the defensive line. I’ve separated this into EDGE players and interior defensive line. Now, I need to caveat that by saying that positional designations between EDGE and IDL, and even linebackers, can be weird. For example, our own Carlos Dunlap is listed as a LB, but he was an EDGE rusher for us last year. Additionally, many players take snaps at both IDL and EDGE (Jones and Danna both switched for us often last year).

What I’m saying is that these buckets will be mostly accurate, but there are going to be some random odd players that show up where you wouldn’t expect. EDGE players should include all players listed as a defensive end or outside linebacker. IDL will be all defensive line players that aren’t in the EDGE bucket.

I also want to say that, in much the same way that approximate value (AV) struggled somewhat with offensive line play, it also struggles with defensive line play, and with defensive play in general. AV (and many public metrics) is heavily reliant on counting stats, so positions that don’t rack those up will be punished by this metric. Disruption gets you no AV if you don’t also get the tackle/sack/fumble/etc.

AV is also designed on defense for a base defense, with four defensive backs on the field, and divides up the “defense” AV points based on that assumption. However, teams play base somewhere around 25% of the time now, so this is slightly favoring the front seven (though it's much more often the front six, with five DB’s) over the defensive backs.