From the FanPosts -- JD
Based on how Kansas City Chiefs' offense has responded to playing top-ranked ranked defenses this season, I thought it would be a good idea to point this out right now:
Using FO's Passing DVOA (Defense-adjusted Value Over Average) metric for the 2022 regular season, the Eagles ranked first at -15.5%. In DVOA scores, an average team gets a score of 0%. Better-than-average defenses will have negative numbers, while better-than-average offenses get positive numbers.
Philadelphia is also ranked sixth in Total Defense DVOA.
FO makes these measurements pretty carefully. Without going to far down in the weeds, here's some of what they say about their DVOA metric:
Doing a better job of distributing credit for scoring points and winning games is the goal of DVOA... DVOA breaks down every single play of the NFL season, assigning each play a value based on both total yards and yards towards a first down, based on work done by Pete Palmer, Bob Carroll, and John Thorn in their seminal book, "The Hidden Game of Football." On first down, a play is considered a success if it gains 45 percent of needed yards; on second down, a play needs to gain 60 percent of needed yards; on third or fourth down, only gaining a new first down is considered success.
I remember reading "The Hidden Game of Football" in the late 80s. It made a lot of sense to me to judge teams on progress toward first downs -- rather than just yards.
FO does more than that -- including normalizing seasons and other data smoothing techniques. Some of this data is proprietary and is only available with a paid subscription. The values that I'm writing about, however, are available with a free registration.
So what about the Chiefs?
By a wide margin, Kansas City had the league's top Total Offense DVOA at 25.2%. Unsurprisingly, the Chiefs' Passing Offense DVOA led the league at 41.1%.
And because it is value over average, DVOA shows how much above or below average a given unit performs against a known quality of opponent. So "didn't play anybody" or "played a harder schedule'" arguments for either team are almost washed out of these numbers.
This, however, does suggest one potential flaw of DVOA: recency.
For example: comparing Kansas City's defense today to what it was in Weeks 1 through 6 would be a fool's errand. We can see the vast improvement of rookies, Frank Clark and Trent McDuffie's returns to health -- and perhaps an "addition by subtraction" when Rashad Fenton (although a good dude) was traded away.
Likewise, the Eagles beating the Jacksonville Jaguars in Week 4 was less of an accomplishment than it was for the Chiefs to beat Jacksonville in the playoffs -- because the Jaguars got better.
So to try to account for the ever-changing qualities of NFL teams, FO also provides weighted DVOA values. These put more emphasis on recent games -- and less on earlier matchups.
For the Chiefs, the weighted Total Offense DVOA still ranks first. But with weighted figures, Philadelphia's overall defensive ranking drops to ninth -- a gap of 22.2 absolute percentage points.
And on the other side of the coin, Kansas City's weighted Total Defense DVOA ranking rises to 14th from 17th, while the Eagles' weighted offensive ranking falls from third to sixth -- a gap of 10.2 absolute percentage points.
For special teams, DVOA scores favored the Eagles in the regular season -- particularly late in the year. But in the postseason, Kansas City has had the edge.
This is all to say that the game will feature a good Eagles defense against the Chiefs' top-ranked offense -- and a just-above-average Kansas City defense against a good Philadelphia offense. I personally don't think all of these metrics have have yet caught up with how good the Chiefs' defense has become -- but it's also true that with all of its skill-position injuries, the Kansas City offense might be overrated, too.
So for me, it's still too close to call. Just remember that in the coming game, quarterback Patrick Mahomes and his receivers will be facing a difficult challenge against the NFL's very best pass defense.