The New England Patriots' Defense Isn't As Good As It May Seem
The axiom, "Defense wins championships," is about to have its yearly fun in the sun.
Around Super Bowl time, people whip that sucker out like a cell phone when you're on the toilet, and dissenting from it is grounds for scorn.
This would seem to lead to the conclusion that the New England Patriots are primed to unleash an epic whoopin' on the Atlanta Falcons. No team allowed fewer points per game during the regular season than the Patriots at 15.6, and neither of their two playoff opponents have topped 17 points. The Falcons, meanwhile, were down in 27th in points allowed for the regular season. Sub-Gucci.
The strange thing about a 16-game NFL season, though, is that you're leaving yourself up to a lot of variance. It's a small sample, so a team's point total allowed in that span can depend heavily on things such as matchups and luck that aren't sustainable from a forward-looking perspective. This is where we start to meet some resistance to the assertion that the Patriots should simply stymie one of the best offenses in quite some time.
As such, we need a little help in determining the true abilities of the Patriots' defense, and we can get most of it from numberFire's Net Expected Points (NEP). This is the metric we use to track the efficiency of both teams and players with the team totals being weighted for strength of schedule, helping account for one of the complicating aspects in a 16-game schedule. By tracking the expected points gained or lost on every play, we'll also get a more clear read on a team's true abilities than we would by just looking at total points allowed.
The Patriots may have allowed the fewest points in the regular season, but does that mean they fit the mold people look for in championship teams? A look at the advanced metrics will give significant reason for pause there.
A Beneficial Schedule
A 16-game schedule isn't necessarily robust, but it's also a pretty solid number of games. You'd assume that -- somewhere in that span -- a team would face an elite quarterback on the opposing sideline. Not so much for New England.
Of those 16 games, 13 came against passers who finished the year with at least 100 drop backs. Here's where those quarterbacks ranked in Passing NEP per drop back, numberFire's metric that shows the expected points a quarterback added or subtracted each time he dropped back during the season.
Overall, 39 quarterbacks hit the 100-drop back mark. None of the top 10 faced the Patriots.
There be a whole lotta nastiness up in here. Four of these 13 games came against passers who were 32nd or worse in Passing NEP per drop back. Only one was against a top-15 passer.
This is why we should be putting value in stats that track efficiency and adjust for schedule. If you never face a top-10 quarterback throughout the entire season, it should be expected that you allow few points. The Patriots did that, but it clearly paints in incomplete picture.
When we turn, instead, to Adjusted Defensive NEP per play -- which includes a team's abilities both against the pass and the rush -- the Patriots finished the regular season ranked 11th overall. This is still above average and well ahead of the Falcons in 25th, but are they the best defense in the league? Most likely not.
The obvious rebuttal you could provide is that the Patriots did face a top-10 passer just last week. Ben Roethlisberger finished seventh in Passing NEP per drop back, making him easily the best passer the team faced this year. They held Roethlisberger to 314 yards on 47 attempts with one touchdown and one interception en route to a 19-point victory. That should help bolster the defense's reputation a bit.
If Roethlisberger weren't so unique, this would be true. That conference championship game was in Foxboro, meaning the Pittsburgh Steelers were on the road. Roethlisberger was a radically different passer on the road this year than he was at Heinz Field, as you can see below with his advanced metrics in the regular season and postseason combined. Success Rate is the percentage of drop backs that result in positive NEP.
|Venue||Drop Backs||Passing NEP||Passing NEP per Drop Back||Success Rate|
If we use just Roethlisberger's Passing NEP per drop back at home, he would have easily led the league in this stat for the season. His number on the road would have ranked 24th, right between Matt Barkley and Eli Manning. The Patriots' defense was impressive last week, but Roethlisberger's general struggles on the road taint that accomplishment a good chunk.
All of this is simply a roundabout way of saying that the Patriots haven't faced a true test defensively yet this season. The Falcons will certainly provide that. Matt Ryan led the league in Passing NEP per drop back (Tom Brady was second). The Falcons held -- by a wide margin -- the top Adjusted NEP per play in the regular season while finishing first in passing efficiency and third on the ground.
If the Patriots want to claim they're the best defense in the league, they'll need to stop the Falcons first, and it's not an assertion they can viably make based on what we've seen thus far.
If we're measuring up the two defenses that will square off in Super Bowl LI, it's clear the Patriots have the superior unit to the Falcons. That doesn't mean they live up to how they're being billed, though.
An 11th-ranked defense can do plenty of damage against mediocre competition, and that's what New England did throughout the regular season. The Super Bowl will be a different task, though, facing Ryan and his scorched-Earth band of efficiency savants. New England should put up a tougher fight than some of the recent teams Atlanta has faced, but it's no lock to shut them down.
If the Patriots come out and dominate defensively in the big game, then it's possible the metrics were underselling them. But until we see them step up against a big-time foe, we should be having faith in these numbers, and they say it's still a defense against which the Falcons can move the football.