April 17, 2021

The Only Football

Get The Latest NFL News And More!

Super Bowl LIV Preview | Football Outsiders

19 min read
Super Bowl LIV Preview | Football Outsiders

The story of Super Bowl LIV is offense vs. defense. Yes, the San Francisco offense is pretty good, and the Kansas City defense is pretty good (against the pass, at least) but the hype around this game is going to be about Patrick Mahomes against the San Francisco defense.

On one hand, we know that offense is more predictable than defense, and that would seem to favor the Chiefs. The 49ers defense has certainly regressed to the mean over the second half of the season. Yes, you might be able to explain that regression with injuries to players such as Kwon Alexander and Dee Ford, and they’ve rebounded to previous levels in the playoffs, but that’s just two games. (In fact, it’s more like one and a half games, because Green Bay was moving the ball well in the second half of the NFC Championship.) And if you want to put extra weight on San Francisco’s defensive performance in the playoffs, you also have to put extra weight on Kansas City’s offensive performance in the playoffs, which happens to represent their top two offensive games of the season by DVOA.

On the other hand, the history of the Super Bowl favors defense. In just this one game, the last game of the season, top defenses have a better record than top offenses. Top 3 DVOA offenses (of which Kansas City is one) have a 13-13 record in the Super Bowl. Top 3 DVOA defenses (of which San Francisco is one) have a 13-7 record in the Super Bowl.

This is the eighth Super Bowl where a top 3 DVOA offense has faced a top 3 DVOA defense. One of those Super Bowls involved the 1991 Redskins, who were a top-3 team on both offense and defense. But in the other six games, the better defense has a 4-2 record against the better offense.

But that’s only six games. Not exactly the biggest sample size.

Let’s dive deeper into both matchups to look at Super Bowl LIV. For those who may be unfamiliar with the Football Outsiders stats, they are explained at the bottom of the page. Scroll down or click this link. Game charting data appears courtesy Sports Info Solutions, unless noted. All stats represent regular season only, except for weighted DVOA and anything else specifically noted.

DVOA 27.9% (5) 30.2% (2)
WEI DVOA 32.2% (4) 44.1% (2)
49ers on Offense
DVOA 7.2% (7) -3.4% (14)
WEI DVOA 13.7% (6) -5.2% (12)
PASS 24.4% (8) -9.3% (6)
RUSH -0.5% (13) 4.1% (29)
Chiefs on Offense
DVOA -19.8% (2) 22.7% (3)
WEI DVOA -13.6% (4) 33.2% (1)
PASS -26.3% (2) 43.7% (2)
RUSH -12.1% (11) -1.4% (14)
Special Teams
DVOA 1.0% (12) 4.1% (2)

If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game.


Let’s start with the meeting of strength against strength, the matchup that everyone is waiting to see. The Kansas City offense wasn’t as powerful this season as it was in 2018, but it still finished the year third in offensive DVOA. Some of that was because of the games where Matt Moore played instead of an injured Patrick Mahomes, and some of that was because Mahomes himself was limited by injuries early in the year. In our weighted ratings that drop early-season games and weight midseason games less, the Chiefs are the No. 1 offense in the league. They’ve had their two best single-game ratings of the year in the two playoff games.

The Chiefs offense in also remarkably consistent, in particular Mahomes. In two years as a starter, Mahomes has finished only one game with a negative passing DYAR: Week 17 of this season gainst the Chargers.

While the Kansas City offense has improved over the course of the season, the San Francisco defense has regressed. All five of their below-average games this season have come since Week 9, although the NFC Championship gets a bit of an asterisk because Green Bay got all of its offense in the second half when the 49ers had a large lead. The decline in San Francisco’s defense does track with a series of defensive injuries, particularly to linebacker Kwon Alexander and edge rusher Dee Ford, both of whom have returned healthy for the playoffs. And San Francisco had one of its best defensive games of the year against Minnesota in the divisional round three weeks ago.

The San Francisco 49ers play primarily zone defense in a scheme influenced by the Legion of Boom-era Seattle Seahawks. San Francisco used zone on 61% of passes according to Sports Info Solutions, which ranked fourth in the league. (They also led the league with 18% of coverages marked as “other,” which is mostly screens and a lot of combo coverages with man on one side and zone on the other.) Mahomes truly excelled against zone coverage in 2019, with 9.7 yards per pass compared to “only” 7.3 yards per pass against man coverage.

You already know about Richard Sherman and how opposing quarterbacks tend to avoid throwing at him on the offensive right side. On the left side, the 49ers have replaced Ahkello Witherspoon with Emmanuel Moseley, and Moseley had excellent charting metrics during the regular season: 63% success rate (10th) and 6.3 yards allowed per pass (14th).

Mahomes specializes in throwing passes to two areas: the short middle, and anywhere deep. Mahomes had the league’s best DVOA on passes to the short middle, but this is also a strength of the San Francisco defense. The same isn’t true of deep passes, 16 or more yards past the line of scrimmage. Mahomes ranked sixth in the league in DVOA on these passes, but a tight grouping at the top of the league puts him pretty close to No. 1. San Francisco’s defense is 12th against deep passes, compared to second against shorter passes. And a particular problem for the San Francisco defense was yards after the catch on these deep passes. San Francisco allowed an average of 10.4 yards after the catch on deep passes, which was last in the league. No other defense was above 7.1. This is not a good weakness to have against the likes of Tyreek Hill and Mecole Hardman.

One interesting aspect of Kansas City’s explosiveness is that the Chiefs’ offense declined relative to the league as the field got shorter. The Chiefs had the No. 1 offensive DVOA on their own side of the field, but ranked 10th in the “front” zone (from the opposing 39 to the 21) and 19th in the red zone. However, that Chiefs decline in the red zone hasn’t been there since their Week 12 bye. In Weeks 1-11, the Chiefs had -14.0% DVOA in the red zone, which ranked 21st. In Weeks 13-20, including the postseason, the Chiefs have had 53.6% DVOA in the red zone, second behind Baltimore.

Incidentally, the San Francisco 49ers’ defensive DVOA was also not as good in the red zone as it was overall. For the season, the 49ers ranked ninth in the red zone. Since Week 12, including the postseason, they’ve ranked 18th in the red zone. However, there’s really no consistency for red zone defense separate from overall defense, so this red zone decline isn’t the best guide to how we should expect San Francisco to play in the red zone in this game.

Contrary to the idea that the Chiefs should spread things out, the 49ers defense was better this year against 11 personnel than against 12 personnel (the main two personnel groups they faced, and the main two groups Kansas City uses). The yardage difference was small, 5.0 to 4.7, but because teams use 12 personnel more on earlier downs, the DVOA difference was -27.6% DVOA (against 11 personnel) compared to -4.0% DVOA (against 12 personnel). Then again, 12 personnel for Kansas City is often spreading things out since Travis Kelce often moves away from a conventional tight end position.

Kansas City’s receivers performed well no matter where they lined up but the slot might be the right place to attack the San Francisco defense. Against receivers lined out wide, San Francisco was third in yards allowed per play and second in defensive DVOA. Against receivers lined up in the slot, San Francisco was ninth in yards allowed per play and 18th in defensive DVOA.

(Note: “Receivers” here means any player, including tight ends and running backs. And “receivers in the slot” includes tight split except a tight end position, even if the receiver is actually the widest receiver on his side of the formation.)

The one type of receiver that San Francisco didn’t have a problem with in the slot? Tight ends. Against tight ends in the slot, San Francisco ranked sixth in DVOA. This was part of San Francisco’s overall great defense against tight ends. One of the most interesting stories of this game is that Kelce and George Kittle are probably the two best tight ends in the NFL and San Francisco and Kansas City were two of the best defenses against tight ends by DVOA. Against all tight end passes, San Francisco ranked second in DVOA. The 49ers allowed tight ends a league-low 33.5 yards per game (after adjusting for opponent) and only four tight ends had at least 50 yards in a game against them this year. Because the 49ers play so much zone, there’s no defender to particularly spotlight for this excellent play; tight ends were covered by linebackers, safeties, and even cornerbacks.

Don’t expect Damien Williams to play a big role as a receiver in this contest. San Francisco also had the best defense in the league against running backs in the passing game. And while Andy Reid has always been considered an excellent designer of screens, but the Chiefs did not do well on screen passes in 2019: just 5.0 yards per pass with -6.2% DVOA. Still, we’re likely to see a few of them to slow down the San Francisco pass rush; the 49ers faced more screen passes than any other defense this year (5.2 yards per play, -20.5% DVOA).

We haven’t talked about the pass rush yet. When it comes to pressure rate, both the San Francisco defense and the Kansas City offense likely rank lower than you expect. The San Francisco defense ranked No. 1 in adjusted sack rate but just 13th in pressure rate. And the Kansas City offense ranked fourth in adjusted sack rate but just 23rd in pressure rate allowed. They get away with this because Mahomes was very good under pressure this season. Kansas City’s offensive DVOA with Mahomes under pressure was -3.7%. Being close to zero is phenomenal when the league average under pressure was -64.4%.

The 49ers succeed despite a lower pressure rate because they’re bringing that pressure with only four guys. The 49ers blitzed on only 19.5% of pass plays this year, 26th in the league. Mahomes is awesome against a four-man rush, 8.4 yards per pass. But he’s even better against a blitz: 9.1 yards per pass.

Part of his success against pressure is that Mahomes is a very good scrambler. Including the postseason, Kansas City has 123.8% DVOA on Mahomes scrambles — only Ryan Tannehill is higher — with 9.9 yards per carry. Mahomes’ scrambling ability was limited earlier in the season by a nagging ankle injury, but he scrambled more in the second half of the year and has ten scrambles in the two previous playoff games, eight of which converted for first downs.

San Francisco was actually pretty good against scrambles this year, but quarterbacks ran against them a lot. San Francisco faced 70 total quarterback carries during the regular season (not counting kneels), which led the league. They allowed 5.3 yards per carry, a bit below the NFL average of 5.9 yards per carry. On just scrambles, they allowed 5.2 yards per carry and ranked fourth in DVOA. But ranking fourth in defensive DVOA against scrambles still means you’re allowing 17.0% DVOA. It’s just that scrambles are so efficient that the NFL average is 46.1%. So Mahomes is probably going to scramble a few times and those are likely to be strong gains even though the 49ers are relatively good on these plays.

Kansas City doesn’t use the conventional running game much and that’s fine because the running game is clearly a weakness compared to the passing game. In fact, it’s a bigger weakness than the basic DVOA stats suggest. The Chiefs ranked 14th in run offense during the regular season but that rating is propped up by Mahomes’ scrambles. Remove those plays, and Kansas City’s run offense DVOA drops further to -7.9% (18th). 

Kansas City’s favorite rushing plays are outside zone and inside zone, but the Chiefs weren’t particularly good at them in 2019. They had 4.7 yards per carry on outside zone because of a couple of big gains, but just a 34% success rate which led to -21.7% DVOA. On inside zone, the Chiefs averaged just 3.0 yards per carry and had -14.4% DVOA. San Francisco’s defense was also very good against these plays: -30.0% DVOA against inside zone (3.2 yd/carry) and -13.8% DVOA against outside zone (4.1 yd/carry).

Kansas City’s struggles on the ground extend to short-yardage plays, where they ranked 23rd in short-yardage run conversions while the 49ers defense was No. 1 in preventing such conversions. Surprisingly, even Kansas City’s pass attempts in short-yardage situations (3rd/4th down, 1-2 to go) only converted 57% of the time, basically league average.


Let’s move to the other side of the ball, to the more underrated units on each team. Normally, we start these playoff previews with a look at the passing game. After all, passing is generally more important than running in the modern NFL. But for the 49ers, everything starts with the running game, and that running game has been phenomenal over the last two months.

The 49ers only ranked 13th in run offense DVOA for the regular season, but that overall number hides an extraordinary split. From Week 6 to Week 11, the 49ers were missing fullback Kyle Juszczyk and/or tight end George Kittle, two of their top blockers. In those weeks, San Francisco ranked 31st in run offense DVOA and averaged just 3.75 yards per carry. But since Week 12, with both Juszczyk and Kittle in the lineup, San Francisco has ranked No. 1 in the league in run offense DVOA and averaged 5.69 yards per carry. There’s a similar, though smaller, split in San Francisco’s passing numbers.

San Francisco Offense by Week, 2019
  Rushing Passing
Weeks DVOA Rank Yd/Att DVOA Rank NY/P
Weeks 1-5 11.0% 5 5.57 38.7% 5 7.65
Weeks 6-11 -26.4% 31 3.75 15.6% 15 6.70
Weeks 12-20 28.9% 1 5.69 27.1% 8 7.73

Yes, the San Francisco running game really stood out in their dominant win against Green Bay in the NFC Championship Game. But that running game has been playing well for weeks now. In three of the last six games, San Francisco’s single-game rushing DVOA has been over 50% with at least 6.0 yards per carry. Since Week 12, San Francisco’s DVOA has been better running the ball than passing the ball. That’s really rare for a good offensive team. Usually, if DVOA is higher for the run than the pass, it’s only because the team in question has a terrible quarterback.

This matters in particular because run defense is such an obvious weakness for the Kansas City Chiefs. Kansas City’s run defense was 29th in the league in DVOA and 28th in adjusted line yards. It’s a little better in weighted DVOA including the playoffs but still bad (22nd).

San Francisco’s best direction by adjusted line yards was left end (third), which was also the Kansas City defense’s best direction by adjusted line yards (12th). But San Francisco’s ALY numbers were also strong to the right, where the Kansas City defense was very weak.

San Francisco of course loves its outside zone. Based on SIS charting, that was the 49ers’ best run play (20.6% DVOA, 6.0 yd/carry) followed by power (13.2% DVOA, 4.8 yd/carry). DVOA lists Kansas City’s defense as below average with 6.1% DVOA against outside zone and 6.6% DVOA against power.

San Francisco also likes to use more than just running backs in the running game. If we look at runs by wide receivers and tight ends (wildcat plays and fake punts removed), the 49ers ranked first in the league with 195 yards and 109 DYAR. They had 23 such carries, third in the league. Deebo Samuel has added another three for 49 yards in the postseason. We don’t know much about how Kansas City deals with runs like this because they’ve only faced six of them all season.

There is one statistical split where the Kansas City defense does well against the San Francisco offense: the Chiefs excelled this year when opponents had a fullback in the game. The Chiefs had -20.4% DVOA against 21 personnel, allowing just 4.2 yards per play. The first number ranked eighth in the league, the second number fifth. And it’s a much smaller sample of just 29 plays, but the Chiefs also allowed just -16.0% DVOA against 22 personnel. Of course, these personnel groupings with Juszczyk on the field are even bigger strengths for the 49ers offense, whether they are passing the ball or running the ball. San Francisco ran more plays from 21 personnel than any other team and had the best DVOA (26.4%) of any team that used this personnel at least 50 times. That includes 11.2% DVOA running and an even more impressive 48.1% DVOA passing. San Francisco also had the best DVOA out of 22 personnel (42.3%) of any team that used this personnel at least 30 times.

The 49ers primarily have a short passing game based on gaining yards after the catch. Garoppolo’s average depth of target was 6.67 yards downfield; only the two New Orleans quarterbacks were lower. But San Francisco led the league with an average of 6.6 yards after the catch, while Kansas City’s defense was 30th with an average of 5.9 yards after the catch allowed. The gap between the two units is particularly notable on short passes, 1-15 yards past the line of scrimmage. On these passes, the 49ers were first in YAC (5.2) while the Chiefs were 31st in YAC allowed (4.7).

Despite their general dependence on short passing, the 49ers can make big gains downfield. Remember how Patrick Mahomes was one of the NFL’s best quarterbacks on deep passes? Surprise: So was Jimmy Garoppolo. Garoppolo ranked No. 2 in passing DVOA on passes of 16 or more air yards, trailing only Drew Brees. But he was far behind Mahomes in DYAR (total value) on these plays because he just didn’t throw many of them: only 64 for the entire regular season, compared to Mahomes throwing 115 despite losing two and a half games to injury. This tweet from the NFL’s Next Gen Stats group points out how: San Francisco receivers had the best average separation of any offense on deep passes. Play-action is a big part of these shot plays, as the 49ers used play-action on 31% of passes (second in the league) and averaged 9.7 yards on these plays (third in the league).

San Francisco struggled to throw to receivers who were lined out wide, although only two offenses (Oakland and Baltimore) threw less often to receivers lined up wide. The 49ers had 6.2 yards per pass on these plays (28th) and -27.4% DVOA (32nd). Meanwhile, Kansas City excelled against receivers lined up wide, with 6.8 yards allowed per pass (sixth) and -14.3% DVOA allowed (third).

Receivers in the slot were a different story. San Francisco led the league with 9.9 yards per pass to receivers in the slot, and finished fourth in DVOA. Kansas City’s defense was fourth with 7.2 yards per pass allowed to receivers in the slot but tenth in DVOA allowed.

Despite the presence of Frank Clark and veteran late-season addition Terrell Suggs, the 49ers should have the advantage over the Chiefs along the line of scrimmage. Kansas City was only 20th in pressure rate on defense and San Francisco was fourth in pressure rate allowed on offense. Of course, this is in part schematic, with all those designed short throws. San Francisco was all the way down at 26th in ESPN’s Pass Block Win Rate. (Kansas City was 19th in Pass Rush Win Rate.) However, Sports Info Solutions’ blown block charting smiles upon the 49ers offensive line. Sixty-six linemen in the league were listed by SIS with at least 20 blown blocks this season. None of them were on the 49ers.

Kansas City will blitz more often than the 49ers do; they were above average with blitzes on 26.8% of pass plays. Garoppolo dropped a little bit against blitzes, but not much, going from 8.8 yards per pass against four pass-rushers to 8.0 against five or more. And the Kansas City defense this year allowed slightly more yardage when blitzing compared to a standard pass rush, 5.9 to 5.7.

First down was a significant advantage for the 49ers against the Packers defense. It’s less of an advantage against the Chiefs, but still an advantage. San Francisco’s offense was best on first down, ranking fifth in DVOA: fifth passing and sixth rushing. Kansas City’s defense was 14th on first down: 10th against the pass and only 24th against the run.

There’s a general impression that the 49ers might have a harder time coming back from a late deficit because they’re not as explosive as the Chiefs. Numbers from this season suggest this is not true. San Francisco was the No. 1 offense by DVOA when losing by more than a touchdown, and ranked third on defense. When tied or losing by a touchdown or less, San Francisco ranked seventh on offense and third on defense. And Jimmy Garoppolo led four fourth-quarter comebacks during the regular season.


The special teams matchup here is closer than the regular-season numbers show, since San Francisco’s special teams have been better than Kansas City’s during the postseason. As with the NFC Championship, San Francisco’s biggest disadvantage on special teams is probably not really a disadvantage. Including the postseason, Kansas City’s Harrison Butker has been worth 7.9 points above average after adjusting for weather while San Francisco’s Robbie Gould has been worth 4.1 points below average. However, we know that 1) field goal kicking is very inconsistent from year to year; 2) Gould has good career numbers, which are probably more trustworthy than one-season numbers; and 3) Gould has positive value since he returned from a quad injury in Week 13, including a 54-yarder he hit against Green Bay in the NFC Championship. So it’s hard to believe that the 49ers should really be trusting Gould any less than the Packers trust Butker just because Gould missed a handful of kicks early in the season.

Kansas City’s other advantage comes on kickoffs, where the Chiefs were better than the 49ers on both kickoffs and kick returns. Mecole Hardman had a touchdown against the Chargers but also fumbled away two kick returns this season. Ten of the 26 standard kickoffs he chose to return did not make it as far as the 25. Richie James had neither lost fumble nor touchdown for San Francisco. Six of the 18 standard kickoffs he chose to return did not make it as far as the 25.

Both the 49ers punting and the Chiefs punt returns finished seventh during the regular season, but the 49ers punting had more positive value. We have punter Mitch Wishnowsky as below-average in gross punt value but the 49ers’ punt coverage team was excellent. The same goes for Dustin Colquitt and the Chiefs’ punt coverage team. Hardman and James are also the main punt returners.

Overall, the Kansas City special teams are more high variance than the 49ers. They’re more likely to have a big play that changes the game, whether that play is good or bad for the Chiefs. The playoff win over Houston, for example, saw both sides of that variance.


The more I’ve analyzed this matchup, the more I’ve picked out places where the 49ers seem to have a good advantage on the Chiefs. The performance of the San Francisco running game since Week 12 is obviously a huge issue. Kansas City has to stop that running game, but it’s a very different style of running than the Tennessee Titans running game that the Chiefs stopped two weeks ago. Passing may be more efficient than running, but running is certainly plenty efficient enough when it’s gaining 5.7 yards per carry. The Chiefs may want to dedicate more assets to stopping the run, or perhaps use Tyrann Mathieu as a rover who will be there to both key on running plays and help take away short crossing routes.

The 49ers can also hit deep passes. But they’re only going to try a handful of those each game. Those are high variance plays: concentrate on San Francisco’s running game and short passing attack and hope the 49ers miss when they do take a shot downfield.

When Kansas City is on offense, San Francisco should be able to pressure Patrick Mahomes more than most of the defenses he’s faced this season. They may be able to contain Travis Kelce using Fred Warner, who is an excellent coverage linebacker. And yet, I keep coming back to the idea that the passing game is the biggest determinant of who wins games in the NFL. Offense is more consistent and predictable than defense. Offense is on a greater scale than defense: the best offenses are better than the best defenses. We see that here, where Kansas City’s pass offense DVOA is nearly 20 percentage points stronger than San Francisco’s pass defense DVOA even though both units ranked second in the regular season.

We’ve been lucky to have a string of very close Super Bowls with matchups that looked very close going into the games, and this one is no different. But I give the slight edge to Kansas City. I think San Francisco will be able to have offensive success running the ball, but their defense is not going to go out and make Patrick Mahomes look like Kirk Cousins looked three weeks ago. Calling for a high-scoring game didn’t end up working out for me last year but I’m calling for a high-scoring game again this year. I also think it will be close, but the Chiefs are the favorite with the better chance to come out ahead.


DVOA (Defense-adjusted Value Over Average) breaks down each play of the season and compares it to the NFL average based on situation and opponent. You’ll find it explained further here. Since DVOA measures ability to score, a negative DVOA indicates a better defense and worse offense, and a positive DVOA indicates a better offense and worse defense.

Team DVOA numbers incorporate all plays; since passing is generally more efficient than rushing, the average for passing is actually above 0% while the average for rushing is below 0%.

SPECIAL TEAMS numbers are different; they represent value in points of extra field position gained compared to NFL average. Field goal rating represents points scored compared to average kicker at same distances. All special teams numbers are adjusted by weather and altitude; the total is then translated into DVOA so it can be compared to offense and defense. Those numbers are explained here.

Each team is listed with DVOA for offense and defense, total along with rush and pass, and rank among the 32 teams in parentheses. (If the DVOA values are difficult to understand, it is easy to just look at the ranks.) We also list WEIGHTED DVOA (WEI DVOA), which is based on a formula which drops the value of games early in the season to get a better idea of how teams are playing now (explained here). Weighted DVOA ratings include the playoffs.

Each team also gets two charts showing their performance this year, game-by-game, according to offensive and defensive DVOA. In addition to a line showing each game, another line shows the team’s trend for the season, using a rolling average of the last five games. Note that the defensive chart is reversed so upwards is a more negative defensive DVOA (which is better).


Leave a Reply

Your email address will not be published. Required fields are marked *