In this series, we review the Kansas City Chiefs’ performance primarily using Next Gen Stats (NGS) — along with other advanced metrics that turn up during the season. For any questions on the statistics used in this series, please refer to our Football Analytics Glossary and Metrics page.
In this edition of Chiefs by the Numbers, we will be taking a dive into some advanced analytics to see how Kansas City’s defense is performing — and where it could get better.
One of the best resources for football analytics is Football Outsiders, which is known for its DVOA statistic. This metric describes a team’s efficiency (on offense, defense and special teams) by comparing the result of every single play to a league average that is based on situation and opponent.
DVOA measures not just yardage, but yardage towards a first down. In other words, five yards on third-and-4 are worth more than five yards on first-and-10 — and much more than five yards on third-and-12. Red zone plays are worth more than other plays. Performance is also adjusted for the quality of the opponent.
DVOA is expressed as a percentage, so a team with a DVOA of 1.0% is a little better than an average team, while one at 20.0% is a lot better. Because DVOA measures the potential for scoring, offensive and special teams units are better with larger, positive numbers — while defensive scores are better when are smaller, negative numbers.
The Chiefs are currently ranked fifth in total DVOA — which is calculated by adding together the offensive, defensive and special teams figures. Kansas City is ranked 22nd in defensive DVOA, first in offensive DVOA and 18th in special teams.
The Chiefs’ defensive DVOA
Kansas City’s defensive DVOA is 6.7%. For comparison, the Buffalo Bills rank first with a score of minus 16.7%, while the Las Vegas Raiders rank last with a score of 17.6%.
For a historical reference, here are the Chiefs’ defensive DVOA scores during the time quarterback Patrick Mahomes has been the starter.
Historically, teams that win Super Bowls tend to be in the top 15 in defensive DVOA.
When broken down further, the Kansas City defense ranks 22nd in passing DVOA and 17th in rushing DVOA.
Week- by-week defensive DVOA
One thing that stands out to me from these numbers: the Chiefs’ defense was significantly better in the first two weeks and the last three weeks than it was from Week 4 through Week 6. This directly correlates with the suspension of Willie Gay Jr. When Gay has been on the field, the team’s defensive DVOA has been 1.5%, which would jump it to 18th overall.
As the defense gets healthier with the return of Trent McDuffie and Mike Danna — and with Frank Clark’s return from suspension — we should be able to see the Kansas City defense perform at its full potential.
Defensive DVOA by down and distance
|Down & Distance||Rnk||DVOA|
|2nd & Short||23||10.5%|
|2nd & Mid||22||6.4%|
|2nd & Long||32||51.9%|
|All 2nd Downs||32||27.9%|
|3rd/4th & Short||19||4.3%|
|3rd/4th & Mid||9||-9.6&|
|3rd/4th & Long||12||-22.0%|
|All 3rd/4th Downs||14||-5.1%|
When digging into these numbers, we see the Chiefs are horrible on second down — particularly second-and-long. But they’re pretty good on third/fourth-and-long — just not so much on third/fourth-and-short.
The Chiefs are giving up way too many yards on second down — which is what leads them into those third/fourth-and-short stations where they also aren’t at their best. If the defense can figure out how to get better on second down — thereby forcing more long conversion attempts on third and fourth down — we should see the team’s defensive efficiency rise pretty quickly.
Individual player evaluations
Using Pro Football Focus premium grading system, we can see which defensive players are performing at a high level. Similar to DVOA, these grades should not be taken as fact — but they are a good way of evaluating players at their respective positions.
Ever since being drafted in 2019’s second round, Thornhill has been a fan favorite. Before his ACL injury at the end of his rookie season, he was graded the league’s 26th-best safety with a PFF grade of 71.5. This was carried by a coverage grade of 78 that ranked him in the top 15 safeties. Last year, we saw flashes of rookie-year Thornhill: he ranked 25th overall. While his coverage grade wasn’t nearly as good as it was during his rookie season, his run defense and tackling grades had improved.
This year was supposed to be the year in which Thornhill would take the next step, cementing himself as a quality safety. Unfortunately, that hasn’t happened. By these numbers, this is his worst year yet. Thornhill is now ranked 51st out of 63 qualified safeties. His overall PFF grade is 56.6 — with a coverage grade is 60. Both are the lowest of his career.
Despite Thornhill’s poor numbers, I don’t believe the Chiefs should switch to safety Bryan Cook. The rookie’s numbers aren’t much better — and come from a much more limited sample size. Still, Kansas City would definitely benefit from Thornhill improving his play on the back end.
The fourth-year defensive tackle — and self-proclaimed CEO of “Stuff Nation” — Saunders is putting together his best season. Since his rookie year, Saunders has struggled to stay on the field. Now finally healthy, PFF ranks him at 53rd among defensive tackles. His PFF grade of 61 is much better than his rotation partner Derrick Nnadi, who ranks 132rd out of 138 defensive tackles who have played at least 20% of their team’s snaps. PFF ranks Nnadi — known for his run-stuffing ability — 132nd against the run, while Saunders is ranked 39th.
Saunders and Nnadi are currently splitting snaps fairly evenly. I’d like to see Saunders get more snaps — and see veteran defensive tackle Taylor Stallworth get playing time over Nnadi.
Coming into this season, there were very lofty expectations for the first-round pick. For the most part, Karlaftis has not lived up to those statistical expectations. His play on the field, however, is what the Chiefs needed from him.
Karlaftis has played on 72% of this season’s defensive snaps — behind only Chris Jones on the defensive line. He’s recorded 23 quarterback hurries, batted down five passes and recorded one sack. PFF has given him a 45.8 grade. Except for Nnadi, that’s the worst grade for any Kansas City defensive regular. While Karlaftis’ production hasn’t been pretty, he’s provided snaps in a position where the Chiefs desperately needed them.
There are only five rookie defensive ends who have played more than 50% of their team’s snaps. Of those, three went within the top five picks of the draft: Travon Walker, Aiden Hutchinson and Kayvon Thibodeaux. The other player is Atlanta Falcons’ EDGE Arnold Ebiketie — whom PFF grades out better than Karlaftis in every one of its stats. That raises some concerns about evaluating Karlaftis’ selection.
The hope is that Kansas City can use Karlaftis’s energy, durability and work ethic to mold him into a productive No. 2 edge rusher who can play around 80% of the snaps. I believe in him — and I think that in 2023, he will have a big year. Until then, he will continue to eat up snaps — and hopefully, begin to produce more big plays.
For what it’s worth, PFF thinks Karlaftis turned in his best performance on Sunday. He was given a game grade of 68.5 — his best so far.
Where the Chiefs’ defense is headed
After going over the advanced analytics for Kansas City’s defense, I believe that it is on an upward trajectory toward being an average defense. They have the talent to reach that goal. They just need a few players to bring better performances — and a few to start playing more often. If defensive coordinator Steve Spagnuolo can figure out how to fix the unit’s second-down issue, the unit could become one of the better defenses of Mahomes’ career.