Super Bowl LV Preview

Photo: USA Today Sports Images

One of the problems we have to deal with in writing a Super Bowl preview is repetitiveness. When we're down to only two teams to write about, it means we've written a lot about them already. We covered Kansas City and Tampa Bay plenty during the regular season. We've now written about Kansas City in two playoff previews and Tampa Bay in three. They're the only two teams we've covered for the last two weeks. So I apologize that my introduction to this Super Bowl LV preview is going to go over some familiar territory. Nonetheless, it's the elephant in the room that has to be addressed if we're talking about the top-line stats of this matchup.

Kansas City has been the Super Bowl favorite in Las Vegas for the entire season. The Chiefs were immediately installed as favorites in Super Bowl LV after Sunday's games, with the line currently at -3 or -3.5 depending on where you look. Conventional wisdom may favor the Chiefs by even more, with roughly 60% of bets currently placed on the Chiefs.

Football Outsiders' DVOA ratings, on the other hand, favor Tampa Bay. Tampa Bay finished the regular season second in total DVOA while Kansas City was only sixth (fifth if we don't count the Week 17 game where the Chiefs sat starters). When we add in the playoffs, Tampa Bay moves to No. 1 in both total DVOA and weighted DVOA (which gives more weight to recent games). Kansas City, ignoring Week 17, is now third. Our ratings have said that Tampa Bay is underrated for weeks now. Tampa Bay had not fallen out of the top three in DVOA since Week 4, even when their record fell to 7-5. In addition, at no point since Week 4 has Tampa Bay had a lower DVOA than Kansas City, enclusive of all games played to that point.

The problem is what we've been calling the Kansas City "flip the switch" theory. The idea is that the Chiefs deliberately took things easy when they had a late lead in regular-season games. This led to a number of games where the Chiefs' win expectancy never really fell below 95% but the score ended up close because of late scores by the opposition.

NBA analysts are used to basketball teams playing like this, but we're not used to it in the NFL. There really hasn't been a team that completely followed this blueprint, underwhelmed in the regular season despite winning a bunch of games and then turned it on in the playoffs with a dominant postseason march to the championship. There are some similarities between the Chiefs and the 2011 Packers or 2009 Colts, but those teams didn't win the Super Bowl.

If the "flip the switch" theory were true, we would see evidence of the Chiefs playing a lower level with a late lead, and we would see them get significantly better once we got to the postseason. And, in fact, both of those things have happened. During the regular season, the Chiefs were significantly worse in the fourth quarter with a lead of more than a touchdown. Their offense dropped to -15.2% (21st in the NFL) and their defense rose to 23.6% (20th in the NFL). In the playoffs, the Chiefs have played their best ball. Kansas City's DVOA for the AFC Championship was the team's best single-game rating of the year. The week before against Cleveland, their DVOA before Patrick Mahomes got hurt was just as high.

And so, at Football Outsiders we're stuck doing a stat-based preview of the Super Bowl while wondering if our stats for the regular season are properly capturing how good the Chiefs really are. That's a tough predicament, and one to remember as you read through the Super Bowl LV preview below. While Tampa Bay was not good enough to beat Kansas City when the two teams faced off in Week 12, the Bucs have been the Chiefs' match throughout the rest of the regular season. In certain areas, the Bucs have been even better in recent weeks. But what does it mean if our numbers have the two teams as fairly equal, if Kansas City is better than the numbers show?

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.

Kansas City at Tampa Bay

  KC* TB
DVOA 23.0% (5) 31.5% (2)
WEI DVOA 34.1% (3) 38.7% (1)
Chiefs on Offense
DVOA 24.9% (2) -14.6% (5)
WEI DVOA 27.3% (3) -13.9% (5)
PASS 50.0% (2) -5.4% (5)
RUSH -3.9% (12) -31.4% (1)
Buccaneers on Offense
DVOA 2.4% (18) 19.8% (3)
WEI DVOA -3.3% (15) 27.6% (3)
PASS 3.3% (13) 37.1% (5)
RUSH 1.2% (29) -2.0% (10)
Special Teams
DVOA 0.5% (17) -2.9% (26)
*Week 17 not included.

All readers can click here for the open in-game discussion thread. If you have FO+, you can click here to see all the matchup of DVOA splits for this game.


KC Graph TB graph

Perhaps you have heard this, but the Kansas City Chiefs offense is very good. And the main reason the Kansas City offense is so unstoppable is that they're especially extra good on third downs. Did you have a good first down result against them? Did you force them into third-and-long? You can't rest, because nobody is as good as Patrick Mahomes when it comes to converting third-and-long.

On first downs, the Chiefs ranked only sixth in DVOA this season. In fact, they were only 12th in passing DVOA on first downs this season. They ranked 10th in rushing DVOA. Those two figures combine to be sixth overall because the Chiefs pass on first down much more than the average team, and passing is so much more efficient than rushing.

But then the Chiefs ranked second in DVOA on second down, trailing only Green Bay, and first in DVOA on third or fourth down. This was the third straight year that the Chiefs ranked first in DVOA on third/fourth downs. The Chiefs have ranked No. 1 in DVOA on third-and-medium (3-6 yards to go) for three straight years, and they have ranked No. 2 in DVOA on third-and-long (7+ yards to go) for three straight years. Other teams have been better at third-and-long over a 16-game sample, but nobody is as consistently good in those situations as Mahomes.

Tampa Bay's defense worked in the opposite way this year. Tampa Bay ranked third on first down and fifth on second down, but 14th on third or fourth down, including 20th on third-and-long. In the Week 12 game, Kansas City went 6-of-12 converting third downs, including three third-and-8s and a third-and-7 at the end to ice the game.

Bucs defensive coordinator Todd Bowles likes to send pressure at the opposing quarterback, and he sends extra pass-rushers on third downs. Overall, Tampa Bay sent five or more pass rushers on 37% of pass plays this year, which ranked fourth in the league. Bowles doesn't blitz more on third downs, but he sends more guys. On first and second down, Tampa sent five guys on 33% of passes and six or more on just 5% of passes. On third down, Tampa sent five guys on 19% of passes and six or more on 18% of passes.

Overall, Tampa's pass defense this year was the same whether the Bucs blitzed or not, but the numbers were very different for the big blitzes vs. the five-man blitzes. Tampa allowed 7.1 yards per pass with 7.4% DVOA with five pass-rushers, but just 4.4 yards per pass with -51.9% DVOA when sending six or more.

So will they blitz Patrick Mahomes? They shouldn't, a lesson they seemed to learn during the Week 12 game. Mahomes is the best quarterback in the league against blitzes. He had 78.2% DVOA against the blitz this season. In the Week 12 game, the Bucs blitzed Mahomes 12 times (22% of passes). All but one were five-man pressures, and all but one came in the first half of the game. Mahomes gained 9.3 yards per play with 87.6% DVOA and the only blitz in the second half turned into a 20-yard Tyreek Hill touchdown.

Besides, the Bucs should be able to pressure Mahomes with only their front four. Tampa Bay overall ranked sixth in pressure rate this year, including fourth when sending only four pass-rushers. Kansas City's offense was 23rd in pressure rate allowed, and that was before losing left tackle Eric Fisher to an Achilles injury. He'll be replaced by veteran journeyman Mike Remmers. Unfortunately, pressuring Mahomes doesn't help as much as it helps against other quarterbacks. Obviously, Mahomes is better without pressure, because every quarterback is. But the Chiefs had 0.6% DVOA when Mahomes was pressured, ranking him second in the league behind only Ryan Fitzpatrick. Mahomes also can get rid of the ball quickly. In Week 12, he had his third-quickest average time to throw of the season (2.52 seconds) according to NFL Next Gen Stats. One way to deal with pressure is quick throws out of empty formations, but I don't expect much of that: Kansas City ranked 29th in the league in frequency of empty backfields this year. But the Chiefs are sure to use rollouts, bootlegs, screens, and other strategies to get Mahomes away from the Tampa Bay pass rush.

Another way to slow down the pass rush is to get them to jump offsides, and Tampa Bay did that four times in Week 12. During the regular season, Kansas City led the league in getting opponents flagged for offsides, encroachment, or neutral zone infraction, with 26 flags including declined. And Tampa Bay was second in earning such penalties, with 24 of them on defense.

Pressure will at least limit some of those big deep plays to Tyreek Hill; Mahomes' longest pass completion with pressure in Week 12 was a 15-yarder to Mecole Hardman. Speaking of those deep plays to Tyreek Hill, let's talk a little bit about what Hill did in the Week 12 game: 13 catches for 269 yards. Tampa Bay trusted Carlton Davis to cover the opponent's No. 1 receiver for most of the season. Looking at plays where SIS has Davis listed in coverage, 56% of them are against the opponent's No. 1 receiver. This went well for much of the year. Through Week 11, Davis had allowed 6.9 yards per pass with a 56% success rate when covering the opposition's No. 1 option. Those numbers would have ranked 26th and 16th among cornerbacks this year, pretty darn good for covering the likes of Michael Thomas and Allen Robinson. Then came the Week 12 game against the Chiefs, when Davis got burned for five catches of at least 20 yards, three of which were touchdowns. Since then, the Bucs have moved Davis around a lot more. Last week, he was primarily on the outside, but switched from side to side and didn't follow Davante Adams into the slot. In fact, he got burned on a 50-yard go-route touchdown by Marquez Valdes-Scantling when Sean Murphy-Bunting was on Davante Adams in the slot. For the full season against all receivers, Davis was middle of the pack with 7.6 yards per pass (40th) and a 51% success rate (38th).

The other outside cornerback, Jamel Dean, missed the Week 12 game and he's been statistically the best of the Tampa Bay cornerbacks this season: 5.8 yards per pass ranks sixth and a 56% success rate ranks 16th. Dean is faster than Davis and might be a better option to cover Hill. The third cornerback, Murphy-Bunting, has been a huge playmaker in the playoffs, but his regular season was lousy: 9.8 yards per pass (76th) and 37% success rate (75th). It's my guess that Murphy-Bunting's season-long numbers are more telling than his playoff heroics, and the Chiefs should target him.

One way to prevent a big game from Hill is to play a lot of zone coverage. It's not perfect -- Hill's 75-yard touchdown came against Cover-3 with Davis responsible for that deep third -- but the Bucs were primarily a zone team this year, playing zone 66% of the time and man just 25% of the time according to Sports Info Solutions. In Week 12, the Bucs played man on eight plays and got burned with Mahomes going 6-of-8 for 113 yards and two touchdowns.

The problem with playing zone against the Chiefs is that tight end Travis Kelce will destroy you. Kelce is coming off the second-best tight end season in history according to Football Outsiders' DYAR metric. In Week 12, he caught all eight of his targets for 82 yards. All eight targets came against zone coverage. Tampa Bay ranked only 25th in DVOA this year covering tight ends.

You might be surprised to see that Tampa Bay is poor covering tight ends because of the accolades given to linebackers Lavonte David and Devin White. Well, White has a lot of strengths but pass coverage is not one of them. By SIS charting, he allowed 7.1 yards per pass with a 40% success rate in coverage. David was much better, with 4.3 yards per pass and a 59% success rate. The other issue here, and with all the short passes against Tampa Bay, is broken tackles. The Chiefs ranked fifth in broken tackle rate (12.0% of plays) and the Bucs were 26th on defense (11.3%). Kelce led all tight ends in broken tackles and his 24.8% broken tackle rate was a major outlier for tight ends, with Dalton Schultz as the only other tight end above 15%. And based on the regular-season SIS numbers, both David (17.6%) and White (16.0%) had worse-than-average broken tackle rates for linebackers with at least 50 tackles.

Counterintuitively, despite the broken tackles, the Buccaneers did well defending against yards after the catch. The Kansas City offense had more yards after the catch than the NFL average from every distance except for passes of 20 or more air yards, but the Bucs' defense allowed fewer yards after the catch than the NFL average for every distance except... 20 or more air yards.

With all due respect to North Attleboro's finest, Anthony Sherman, the Chiefs essentially use just two personnel groups except in short yardage: 11 and 12. They're absolutely sick when throwing from 12 personnel, with 9.4 yards per play and 72.8% DVOA. And Tampa Bay had trouble with passes from multiple-tight end sets this year. They allowed 8.1 yards per play and 16.5% DVOA on passes from 12 personnel. Contrary to what you might expect, Kansas City's biggest plays with two tight ends on the field come from their wide receivers, not from Kelce and certainly not from backup tight ends Nick Keizer and Deon Yelder. In Week 12, both the 75-yard touchdown to Hill and the 8-yard gain by Hill that ended the game in the fourth quarter came out of 12 personnel. None of Kelce's receptions in the Week 12 game came from 12 personnel.

The Kansas City running game is probably not going to play a major role in this game. That's a combination of Kansas City's pass-first philosophy and Tampa Bay's exceptional run defense. Regarding the former, Kansas City called a pass on 72% of plays in the first half this year, the highest rate in the NFL. Regarding the latter, Tampa Bay has led the league in defensive DVOA against the run for two straight seasons. The Bucs were also No. 1 in adjusted line yards against the run this season, but they didn't allow a lot of longer runs either. Tampa Bay led the league in the fewest second-level yards per carry (5-10 yards past the line of scrimmage) and ranked fifth in open-field yards per carry (11+ yards past the line of scrimmage). The Kansas City running game wasn't particularly strong at breaking those long highlight runs, either. The Chiefs did well in second-level yards per carry (seventh) but only ranked 30th in open-field yards per carry. And Tampa Bay's defense ranked in the top six in ALY in every direction except right end where they ranked 15th -- but Kansas City's offense ranked 21st.

The Chiefs may not even want to audible to handoffs when they face light boxes. Tampa Bay was excellent against the run no matter whether the Bucs had six, seven, or eight men in the box. Kansas City was better against six-man boxes, but not with a significantly higher success rate. They had a higher DVOA because whatever long runs they did have tended to come against those six-man boxes.

Tampa Bay Run Defense by Men in Box, 2020
Men in Box Carries Yd/Car Suc% DVOA
6 104 4.36 37% -30.7%
7 134 2.80 28% -48.2%
8 67 3.30 48% -21.7%
Kansas City Run Offense by Men in Box, 2020
Men in Box Carries Yd/Car Suc% DVOA
6 118 5.42 48% 11.7%
7 128 3.98 45% -11.7%
8 67 3.39 45% -8.9%

As Vince Verhei pointed out earlier this week, Kansas City actually ran the ball better, not worse, in their worst games of the year. When they're good, it's because their passing game is on fire. The main problem with avoiding the running game is the question of whether the Chiefs will be able to run out the clock if they take a late lead. In the fourth quarter of games this year, Kansas City's run DVOA dropped to -31.5% with a lead of 1-8 points and -42.8% with a lead of 9 points or more. (Remember, that's after adjustments in DVOA that account for teams gaining fewer yards when running out the clock with leads late.) If they're trying to ice the game with a late lead, it may behoove the Chiefs to just keep passing to try to gain enough yardage to keep the sticks moving, with short passes that don't have a high likelihood of falling incomplete.

Those fourth-quarter troubles might be exacerbated by the fact that the Tampa Bay defense gets much better at the end of games. Tampa Bay's performance this year suggests a situation where, like in Week 12, Kansas City gets out to a lead and Tampa Bay catches up. Tampa Bay's defense ranked only 26th in DVOA in the first quarter of games. It then improved to 16th in the second quarter, sixth in the third quarter, and best in the league in the fourth quarter.


TB Graph KC graph

We know that offense is more consistent than defense, and more important for predicting game results. That fact might lead you to favor Kansas City in this game, except for one problem: since Week 10, Tampa Bay's offense has actually been better than Kansas City's.

Tampa Bay's offense ranked only 11th in DVOA for Weeks 1-9. In Weeks 10-20, including the playoffs, the Buccaneers rank second trailing only Green Bay. Tampa Bay has a 30.5% offensive DVOA since Week 10. Kansas City, even if we remove Week 17, has a 28.5% DVOA over the same time period.

At the same time that the Tampa Bay offense has been improved, the Kansas City defense -- at least against the pass -- has been worse. Well, at least it was during the regular season. This is where that "flip the switch" theory comes in. The Kansas City defense has flipped the switch more than the offense has. After declining over the second half of the season, the Kansas City pass defense has reversed and been absolutely outstanding over the last two games. The run defense, which had improved, has been the league's worst during the playoffs. I think the Chiefs will happily accept that trade-off.

Kansas City Defensive DVOA by Week, 2020
Weeks All Defense Rk vs. Pass Rk vs. Run Rk
Weeks 1-9 0.6% 14 -5.4% 6 7.7% 32
Weeks 10-16 5.2% 25 14.6% 23 -11.1% 18
Weeks 18-20 -26.9% 2 -44.5% 1 13.0% 14

Tampa Bay has not only been better on offense, but also more aggressive since these teams first met in Week 12. Tampa Bay's use of motion has increased from 46% in Weeks 1-11 to 56% in Weeks 12-20. The latter number would have ranked fifth this season. (For those curious, the Chiefs motion on 60% of plays, which ranked third.) Week 12 was also when Tom Brady became dramatically more aggressive on third-down throws. When it is third or fourth down with less than 5 yards to go, the percentage of Brady throws that went at least 5 yards past the sticks increased from 16% before Week 12 to 66% since Week 12. Since that point of the season, Brady and Mahomes rank first and second in ALEX on third/fourth downs, 4.8 yards past the sticks for Brady and 4.5 for Mahomes.

In the previous section, I detailed Mahomes' outstanding long-term record on third downs. Well, Brady has been better than that for the last two months. Since Week 12 and through the playoffs, Tampa Bay's DVOA on third and fourth downs has been narrowly higher than Kansas City's DVOA on third and fourth downs for the entire season including the playoffs (50.3% to 48.1%). And while the Chiefs' defense ranked 13th against the pass on third downs this season, they were dead last against the run.

Kansas City had an average defense against tight ends this year, but ranked 31st against running backs in the passing game. Their defense funneled targets to the tight ends and running backs, giving up more yardage to those positions than the NFL average but less yardage to wide receivers. In the Week 12 game, Rob Gronkowski gained 106 yards on six catches (seven targets). That was 28 more yards than his most productive game this season. The 140 yards given up to tight ends (Gronkowski and Cameron Brate) was the most given up by Kansas City in any game this season. Particularly bad was a five-man blitz where defensive end Alex Okafor had to drop into coverage on Gronk while two second-level defenders came on the other side. The Chiefs should not run that in the Super Bowl.

Watch out specifically for Gronk on second downs. This year, Gronk had 41.3% DVOA on second down and 36 targets, almost as many as he had on first and third downs combined (41).

Like Kansas City, Tampa Bay excels passing with two tight ends on the field. The Bucs had 61.4% DVOA and gained 8.2 yards per play when passing from 12 or 02 personnel. Kansas City allowed 8.4 yards per play and 21.9% DVOA on passes from 12 personnel this season. Brady was 6-of-8 from these two personnel groups in the first game, including big 29- and 48-yard strikes to Gronk.

Tampa Bay is not the best team to be taking advantage of Kansas City's weakness against running backs in the passing game. Passing to running backs was a huge part of Tom Brady's offense in New England. Things haven't gone as smoothly in Tampa Bay. Ronald Jones was second-to-last with -88 receiving DYAR this season, and Leonard Fournette was at -9 receiving DYAR. Sportradar lists Jones with five pass drops and Fournette with seven. However, Jones did have maybe his best pass reception of the season against Kansas City in Week 12: a 37-yard touchdown where Jones was wide open in the flat after play-action and then broke a tackle from Daniel Sorensen and went all the way to the end zone. Fournette had catches of 8, 6, and a loss of 4 yards in that game.

As far as covering the wide receivers, the Chiefs played 40% man coverage this year, one of the higher rates in the league. For the season, the Chiefs were better in man than zone. They allowed 6.8 yards per pass and 47% success rate in man, compared to 7.6 yards per pass and 54% success rate in zone. In Week 12, they split about 50-50 and the Bucs were much better against the zones: 5.0 yards per pass, -12.6% DVOA vs. man compared to 12.1 yards per pass and 66.2% DVOA vs zone. This wasn't a first half/second half split, as the Chiefs actually played more man than zone in the second half as the Bucs were coming back. But the biggest passes of the second half came against zone, including Gronk's two biggest gains and the 44-yard strike down the middle to Chris Godwin.

Rookie L'Jarius Sneed had the best year of the Kansas City cornerbacks, when he was healthy enough to play. He doesn't have enough targets to be ranked among cornerbacks, but he allowed just 5.1 yards per pass with a 56% success rate. He's also a big part of their blitz packages. Charvarius Ward was the weaker of Kansas City's other two cornerbacks. Ward allowed 7.4 yards per pass with a 50% success rate, both ranking in the 30s. Bashaud Breeland allowed 6.8 yards per pass with a 54% success rate, both ranking in the 20s. The Kansas City cornerbacks will have to worry in this game about penalties, in particular pass interference. Defenses were flagged 23 times for pass interference on Brady throws this year, a record total in our database that goes back 35 years. Brady and the Bucs got 395 free yards out of DPI penalties. Kansas City was fifth in DPI penalties during the season, 15 total including four each for Breeland and Ward.

Another issue with the Kansas City defensive backs is broken tackles. The Chiefs defense ranked 30th in the percentage of plays with broken tackles and the problem is particularly big in the secondary, as SIS tracked Daniel Sorensen with 16 broken tackles, Tyrann Mathieu with 12, and Breeland with 11. However, Tampa Bay may not be the best team to take advantage of this weakness, as it was a weakness for them as well. Tampa Bay's offense ranked dead last in the percentage of plays with broken tackles. Kansas City also allowed a higher average of yards after the catch than the league average at all distances -- but Tampa Bay was lower than the league average in YAC at all distances except for intermediate passes of 10-19 air yards.

The Chiefs spent much of the game with two high safeties, which fits their m.o. from most of the season. According to ESPN Stats & Info, the Chiefs ranked sixth in the frequency of playing Cover-2 (20%) and third in the frequency of playing 2-Man coverage (13%). For the year, the Chiefs ranked third in the NFL against deep passes (16+ air yards) but 22nd against shorter passes. However, the Chiefs are also willing to put their cornerbacks on an island. No team played more Cover-0 this season according to ESPN Stats & Info, including five plays in Week 12 which resulted in four incomplete passes and a short gain by Cameron Brate.

This tied into a specific trend and perhaps the biggest difference between these two offenses in 2020. While Patrick Mahomes destroys the blitz, Tom Brady has struggled against blitzes for two and a half years now. Tampa Bay's team pass DVOA dropped from 41.8% to -4.0% when opponents sent more than four pass-rushers at Brady. And Kansas City likes to blitz. They ranked fifth in blitz frequency at 35%, and No. 1 sending big blitzes of six or more at 15% -- although they allowed more yards per play and a higher DVOA with big blitzes than with regular five-man blitzes. In the Week 12 game, Brady did well against the five-man blitzes (despite an interception) but got destroyed by the six-man blitzes:

Tom Brady by No. of Pass-Rushers in Week 12
Pass-Rushers Plays Yd/Play aDOT DVOA Pressure Rate INT
3-4  23 8.7 7.3 52.9% 4% 0
5 10 10.9 11.1 43.3% 20% 1
6-7 9 3.7 13.3 -94.8% 33% 1

For the season, Brady's pressure numbers are essentially the opposite of Mahomes. Mahomes faced a lot of pressure but also was one of the league's best quarterbacks against pressure. Tampa Bay ranked second in pressure rate on offense, in part because of Brady's own style and ability to get rid of the ball quickly. But when Brady was marked with pressure, he was dismal, ranking 30th out of 36 qualifying quarterbacks in DVOA under pressure.

The pressure in Week 12 forced Brady into his quickest time to throw of any game this year: 2.28 seconds on average according to NFL Next Gen Stats. Note that this interesting trend of a longer depth of target when Brady was faced with more pass-rushers wasn't found in the rest of Brady's season.

The Chiefs really needed to bring the blitz to get consistent pressure during the 2020 regular season. Kansas City ranked only 18th in ESPN's Pass Rush Win Rate. With four pass-rushers, Kansas City had a 25% pressure rate which was slightly below the league average. With five or more pass-rushers, Kansas City had a 48% pressure rate, which was second in the league behind only Philadelphia.

Tampa Bay's offensive line is nowhere near as wrecked as Kansas City's right now but they do have their own weakness, which is backup right guard Aaron Stinnie playing in place of Alex Cappa. The Chiefs bring a lot of pressure up the middle -- not just Chris Jones tied for third in the league with 39 hurries but also rookie Tershawn Wharton with 15 hurries. Given how many accolades they receive from offensive line experts, it's a bit shocking to see that Tampa Bay ranked only 17th in ESPN's Pass Block Win Rate this season. Right tackle Tristan Wirfs, who was lauded for a fantastic rookie season, somehow came out only 30th among tackles in PBWR, and left tackle Donovan Smith ranked only 51st.

Tampa Bay is a strong run-blocking line, however. The Buccaneers ranked ninth in adjusted line yards on offense, while Kansas City was just 25th on defense. Tampa Bay was stronger running to the right side, which is where Kansas City was weaker defending the run. Tampa Bay likes to run the ball a lot more than Kansas City, and the Chiefs' relative struggle against the run will entice the Buccaneers to run more. The Buccaneers have also been running better in recent weeks. Leonard Fournette's yards per attempt have not changed since Week 12, but his success rate has increased from 47% to 55%. Ronald Jones finished eighth among qualifying running backs this year with a 57% success rate.

It's still generally better to pass than to run -- on first downs, for example, Kansas City gave up 2.0 more yards per play on passes than on runs -- but the ground-game matchup has a lot more promise for Tampa Bay than for Kansas City. But I don't know if Tampa Bay will run as much as you might expect. Watching Tampa Bay games this year, you might have gotten the feeling that every first down was handed off to Ronald Jones for four yards, but it wasn't really that way. Including the playoffs, Tampa Bay ran on only 33% of plays in the first half of games. Remember how Kansas City ranked 32nd in this stat in 2020? Tampa Bay ranked all the way down at 28th, running not much more than the Chiefs.

There is one time that Tampa Bay absolutely should pound the ball on the ground, however, and that is in short-yardage situations. The Tampa Bay offense was No. 1 in the league converting short-yardage runs, with an 88% conversion rate. You know about the Tom Brady sneak and its historical effectiveness but most of these conversions were by the running backs. (Brady did go 6-for-6 on sneaks during the regular season.) Meanwhile, Kansas City's defense was dead last in preventing these conversions, with a 78% conversion rate. This ties into one of the big differences in this game, the aggressiveness of the head coaches on fourth downs. Andy Reid, who was very conservative for much of his career, has changed in the last few seasons and ranked sixth in EdjSports' critical call index this year. Bruce Arians has always been conservative and ranked 30th in the critical call index. But both coaches have the players who should be encouraging them to go for it on fourth-and-short. Reid has Mahomes. Arians has this huge advantage in the short ground game. One typical complaint about analytics people is that we constantly suggest teams go for fourth-and-short without considering the actual players involved and the team strengths and weaknesses. This is one time where the team strengths and weaknesses clearly suggest to Tampa Bay: pound that rock in short yardage instead of punting.


The gap between these teams is larger than it looks at first glance, because the Chiefs' rating for this year is dragged down by one horrible special teams game in Week 3 against Baltimore. They were second in 2019 and currently rank eighth in weighted DVOA for 2020. The "true quality" of their special teams is probably closer to those rankings than to this year's No. 17 regular-season finish. Tampa Bay has had two positive special teams games in their three postseason contests, but they were generally below average for the season.

Mecole Hardman was average returning kickoffs and punts and had a 67-yard punt return touchdown against Miami. Punt returns were the biggest negative for the Kansas City special teams and that was fluky, as most of the negative value came from the occasional return attempt by Tyreek Hill (he muffed his one try) or Demarcus Robinson (remember the safety against New Orleans)? Last week's muffed punt by Hardman was his second of the season, but those are a pretty random event. The other big negative was kickoff coverage, which was subpar (and allowed a touchdown in the Baltimore debacle mentioned above). Kicker Harrison Butker has only missed three field goals (including the playoffs) but he's missed seven extra points (one blocked).

Tampa Bay wasn't horrible in any particular area of special teams this year, but they were below average in all five phases that we measure. Bradley Pinion is fine on punts and kickoffs but the coverage team gave up some big returns. Jaydon Mickens returns both kickoffs and punts; his 43-yard kickoff return against Green Bay two weeks ago was his longest of the year. Ryan Succop is very average and doesn't have as strong a leg as Butker if these teams get bogged down around the 35 and face a big decision. Butker is 5-for-5 from 50 yards or more, including the playoffs, and was 3-for-6 a year ago. Succop's long for the year is 50 yards, and he hasn't hit one over that distance since 2018.

One difference in special teams that goes in Tampa Bay's favor: Kansas City had 18 penalties on special teams this season, while Tampa Bay had only seven.


On one hand, we have our stats which suggest that Tampa Bay was a better team this year than their 11-5 record showed. Including the playoffs, they were the top team in the league according to DVOA. On the other hand, we have a Kansas City team which the underlying stats suggest was not as good as its 14-2 record, a team that dramatically underplayed its Pythagorean projection based on points scored and allowed (by 3.3 wins). Normally, that would clearly suggest an upset pick, with Kansas City wrongly favored in this game.

But then there's the "flip the switch" theory. I don't think we can ignore the fact that the Chiefs did, in fact, ease up on opponents when they had a large fourth-quarter lead. The Buccaneers built their rating in the same situations -- in fact, they had the league's second-best defense with a two-score lead in the fourth quarter. I also feel that the matchups in this game favor the Chiefs. Nobody really has the right defense to cover both Tyreek Hill and Travis Kelce, and the Buccaneers are no exception. Tampa Bay's ability to get pressure with just their front four will be mitigated by Patrick Mahomes' ability to make plays under pressure. The Buccaneers should gain good yardage on the Chiefs on the ground, and their passing game has been on fire for the last two months. But at some point, the Chiefs will pressure Tom Brady with blitzes. That will lead to a mistake or two, and those turnovers will likely be the difference in the ballgame. This matchup is closer than many fans may expect, but I think we'll see a close Kansas City victory and our first repeat champion since the New England Patriots in 2004.


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).

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).


69 comments, Last at 13 Feb 2021, 8:43am

#1 by Aaron Brooks G… // Feb 04, 2021 - 11:25am

There are some similarities between the Chiefs and the 2011 Packers or 2009 Colts, but those teams didn't win the Super Bowl.

1998 Broncos? Defending SB champ played a straightforward 12 weeks, took 4 weeks off, and then the defense re-appeared for a playoff run where no one was within a score in the 4th quarter in any game.

That run involved trouncing the 16-2 Falcons and handing Testaverde his second loss as a Jets starter in 1998 (13-2). 

Season-long DVOA still liked Denver, but weighted had them as substantial dogs to the Jets and a small dog to the Falcons.

Points: 0

#22 by Jetspete // Feb 05, 2021 - 11:41am

The 98 Broncos, like the 04 patriots and a few other teams that tried to go back to back, had an intrinsic quality to put their throats on teams.  It'd be the second or third quarter and youd be in a 4 point game or even ahead, then you'd come back from the kitchen and denver would be up 17 and it was over.  Happened to the Jets in the title game, they were up 10-0 and almost in a blink of an eye it was 20-10.  The 2020 chiefs most definitely do NOT have that quality.  

I'm sure DVOA has been tweaked over the years to account for this, but my guess is it has to do with KC's inability to run the ball consistently on offense while also being unable to stop the run on defense.  The opponent is always playing from behind, but pretty much every game between the Jets game and the AFC title felt in reach for the opponent until the closing minutes.  

Points: 0

#58 by MC2 // Feb 06, 2021 - 8:31pm

This year actually reminds me more of the 1997 Super Bowl (Broncos vs. Packers).  That year, just like this year, you had the defending champs, led by a young, cannon-armed gunslinger with a propensity for jaw-dropping, highlight-reel plays (in fact, Favre probably made more of those jaw-dropping throws than Mahomes, although he also made more mistakes) vs. a wild card team, led by an aging, lock HOFer, who many people thought still had something to prove (for Elway, that he could "win the big one" and for Brady, that he could succeed without Belichick).

Points: 0

#2 by ChrisLong // Feb 04, 2021 - 11:36am

This is the way the flip switch theory should’ve been handled all along. Not by fudging the numbers.

That said, a reader on the other thread proposed that the reason including performance with a late lead improved DVOA performance was that it represented regression to the mean.

As another hypothesis, would this also be affected by the shifts in run/pass ratio by offenses when they have a late lead? We know that passing is more efficient than running, and we know teams with a late  lead tend to run more and be more conservative in general. I think this could be a good explanation for why DVOA goes down, and especially so for Kansas City which has a hyper-efficient and aggressive downfield passing offense.

Similarly, their defense would likely do worse in the eyes of DVOA because opponents are passing more and more aggressively, and running less.

Even if DVOA accounts for how passing or running change in efficiency depending on situation, passing would still be more efficient and total DVOA doesn’t really document shifts in run/pass ratio well. This explanation, if it is true, accounts for the observed changes in efficiency without resorting to a vague theory that somehow Kansas City is lazy or disinterested in some games.

Points: 0

#3 by theslothook // Feb 04, 2021 - 12:00pm

Much like garbage time and to a lesser degree clutch, no one can ever define "let foot off the gas".  "Its become a know it when you see it" ex post facto explanation that can cover pretty much anything. "The browns clearly didn't care about the Steelers, how else do you explain get throttled in the regular season and reversing course in the playoffs!"

I don't think KC's defense took its foot of the gas anymore than other defenses do when they have a big lead. I think the KC defense, much like the Colts 09 defense, isn't good but its equipped to play with a lead. Get behind and have the offense struggle and its going to suck. 

The KC offense is the bigger question but honestly its not that interesting either. Last year's KC offense similarly underwhelmed relative to the playoffs. This year's has to and I think playing possum is the wrong for it. I think its simply a matter of behavior shifts. How threatened does this offense feel. If it feels threatened, it starts to fire guns blazing. If it doesn't, it won't. That helps explain its latter season slumps as its competitors fell further behind in the standings. 

Points: 0

#5 by coltsandrew // Feb 04, 2021 - 1:27pm

Pretty much all analysis is an ex post facto explanation. The point Aaron's getting at is that DVOA doesn't consider KC a 14-2 (really 16-2) team, and yet they are. When a model can't explain reality, you adjust the model because you can't adjust reality. All adjustments are by definition subjective, but the relevant question is whether the adjustments make the predictive model more accurate.

Points: 0

#7 by Aaron Brooks G… // Feb 04, 2021 - 1:49pm

Much like garbage time and to a lesser degree clutch, no one can ever define "let foot off the gas".  "Its become a know it when you see it" ex post facto explanation that can cover pretty much anything...I think its simply a matter of behavior shifts. How threatened does this offense feel. If it feels threatened, it starts to fire guns blazing. If it doesn't, it won't. That helps explain its latter season slumps as its competitors fell further behind in the standings. 

I like how you argued letting your foot off the gas couldn't be defined, and then defined it. =)

It seems possible to define garbage time/letting off the gas, using a combination of game length remaining and ExpWin%. Something like a DVOA in 4th quarter with and without an EW%>0.95 or something like that.

Points: 0

#28 by nsheahon // Feb 05, 2021 - 3:01pm

"I don't think KC's defense took its foot of the gas anymore than other defenses do when they have a big lead. I think the KC defense, much like the Colts 09 defense, isn't good but its equipped to play with a lead. Get behind and have the offense struggle and its going to suck. "

If this were the case than the KC defense would have a BETTER DVOA when playing with 2 score leads not WORSE which is what they have. I don't think they let off the gas, but if you watch them play, KC is very aggressive on defense when the game is close. When they have a 2 score lead, they play much softer in coverage and try to keep things in front. The problem is, they play too soft and teams march up field on them. They need to keep the aggressiveness up, which they have done in the playoffs.

Points: 0

#4 by Will Allen // Feb 04, 2021 - 12:45pm

Only have watched 3 games this year, so I only have the most ill-informed of wild-assed guesses to offer (which describes most opining, on most topics, this year) but my inclination is that Chiefs win because Mahomes will make plays that very few can. My sentimental wish is to see Tom Moore on the coaching staff of another champion, because I know people who were coached by him in college 50 years ago, and they sang his praises before he collected his first Super Bowl ring on Noll's staff in Pittsburgh. He and Wade really do deserve ugly yellow HoF blazers as assistants; ugh!, now I've broken my pledge to never comment on the HoF again.

If Moore, Bradshaw, Manning,& Brady ever play 18 together, I'll volunteer to drive one of the carts, and I'll pick up the tab for the beer.


Points: 0

#19 by anotheroldguy // Feb 04, 2021 - 10:46pm

I noticed you missing for a while from some threads with high Vikings content, and 2020 being the hot mess it was, assumed the worst. No, you don't know me, but I've appreciated your commentary here on FO for awhile. 2021 really may be better!

Points: 0

#43 by Will Allen // Feb 06, 2021 - 9:21am

Thanks. Just been a horrid year on multiple fronts, precluding spending much time watching football, which was the least of concerns. Been more than 40 tests in my family since March, and not single positive, so we've nothing to complain about. We're also roughly half vaccinated, so that's good. Looking forward to summer.

Points: 0

#6 by murftastic // Feb 04, 2021 - 1:34pm

When you take your foot off the gas in a car, it's to conserve gas. In order to offer a complete explanation for why KC declined in garbage time, you need to at least hazard a guess at what resource they were conserving.

On offense, it was almost certainly red zone trickeration. Their red zone was astonishingly good in September, then regressed at the end of the season, then got amazing again in the playoffs. It seems pretty obvious at this point that they were conserving some of their best red zone ideas for when they really needed them.

I'm not a sophisticated enough fan to evaluate pass pressure, but the stat guys say the defense got demonstrably worse when ahead this season. I suspect that means Spagnuolo saved some of his more exotic blitz packages for the playoffs, too. That seems to jibe with what the film room guys are saying on YouTube. Again, outside my area. 

If that's all that happened - the Chiefs knew by October 1 they were a mortal lock to make the playoffs and planned accordingly - there's no way they were the first team in the Super Bowl era to do so. It's just that our models aren't sensitive enough yet to detect that. 

Points: 0

#10 by jds // Feb 04, 2021 - 3:27pm

I like this explanation to flip the switch.  That is, they are not flipping the switch to try harder, they are flipping the switch to try different things.  Much like the pirouette play last year; they've got stuff at the back of the playbook, that they never run, until they really have to.  On D, that might mean they play a lot of prevent, on O, that means they do their normal good stuff - and that is enough to win.  When they get a tough opponent or tough game situation, they aren't trying harder, they are just trying different stuff.  

Points: 0

#26 by medelste // Feb 05, 2021 - 2:02pm

Good points. When you see how effective their red zone offense and their third and fourth down offense is, it tells me the Chiefs offense is incredibly effective when the NEEDs to be. In the Super Bowl, they will need to be effective, and so they will get it done.

Points: 0

#8 by jheidelberg // Feb 04, 2021 - 1:51pm

I just can not buy the Chief's flip the switch theory, especially when looking back and finding that the 2011 Packers are virtually the only comparable team.  Why would anyone do this?  If the Chiefs lost a few more games, they would not have had the number 1 seed, and the whole playoff scenario would have been different.

Lets remember that the Chiefs offense won a game against the Chargers in which Anthony Lynn punted on 4th and 1 from their own 34 after receiving the OT kickoff.  The Chiefs needed Butker to kick a 58 yard FG to win the game.  That is not stellar offense or "flipping a switch"

When the Chiefs beat the Raiders, had the Raiders not scored so quickly, the Chiefs offense would not have had over a minute to score on the flip side against the Raiders poor defense.

When the Chiefs beat Denver the second time, Nick Fangio called for a punt leading 19-16 at his own 49 with 6:18 left.  The Chiefs took over 5 minutes to kick a 48 yard FG to tie the game and give Denver a shot to win in regulation.  That is not stellar offense or "flipping a switch"

In the coin toss games, it always came up heads for the Chiefs, except for the first Raiders game.

I think that the narrative is being used to fit the results.  This is like creating a hypothesis, after collecting the data as opposed to having a hypothesis and using a controlled experiment to prove/disprove it.  This is in part why we have multiple hypothesis regarding the Chiefs performance this year.

If you flip a coin 6 times it will come up heads 1/64 times and tails 1/64 times.  The Chiefs simply won almost all of their coin flip games due in part to poor coaching decisions by their opponents and in part by making long game winning/tying field goals.  Do the Chiefs really want a game coming down to a long FG with a kicker that missed 6 extra points?

Why not have the "unable to flip the switch back theory" to describe Atlanta? The theory would say that Atlanta took it easy with a big lead on multiple occasions but unlike the Chiefs could not flip the switch back to on from off and thus did not recover to win.  Just like they did not recover an onside kick, so they lost to Dallas. Does anyone believe that Atlanta took it easy?  Starting in the Super Bowl a few years back?

I am shocked that on this website we are not sticking with DVOA and instead have created a scenario in which Tampa is the better team at overall DVOA, weighted DVOA, the home team (very small advantage this year), and yet given only a 47% chance of winning by changing the input to the computer model that runs the iterations.


Points: 0

#11 by coltsandrew // Feb 04, 2021 - 3:31pm

If you're shocked, it might be because you don't get the point of DVOA. The point of the metric is not to predict the results of super bowl, but to analyze how good all the teams in the league really are. The acronym stands for Defens-adjusted Value Over Average. The key word is average. One of the main things this analysis provides is an understanding of how much of a team's record is due to ease of schedule and how much is due to coaching/talent. 

That said, DVOA is still relevant to the super bowl because the super bowl is a verbatim rematch of the week 12, when the Chiefs went to Tampa and beat the Bucs. If you've read the quick reads preview articles, you know that Tampa's single game DVOA was less than 5% and KC's was more than 45%. As has been repeated ad nauseum on football outsiders at this point, while TB has been the better team thus season, KC is a matchup nightmare for them, as evidenced by the single-game DVOA earlier this season. It really is a bit dense to rely more on proxy analysis than direct analysis, even if the direct analysis is a sample size of one. In a previous head-to-head earlier this season, KC totally outclassed the Bucs, according to DVOA.

So, the question that remains is, does anyone really think that TB's performance against non-KC teams is really that predictive of how they'll perform in a rematch?


Points: 0

#16 by jheidelberg // Feb 04, 2021 - 7:01pm

The point of the playoff odds report is to run 30,000 iterations and based on DVOA/weighted DVOA give a percentage of each team making each playoff slot.  DVOA/weighted DVOA is clearly used to predict future wins.

I would like to know that if the iterations were done without changing the inputs, the percentage of times that Tampa or KC will win.  The only DVOA adjustments ever made during the season are due to injury.  You have challenged my comment and upon further review FO has stepped out of bounds, using its own rules, when running iterations regarding the Super Bowl.  Therefore my comment stands as called.

Points: 0

#23 by Joseph // Feb 05, 2021 - 12:07pm

Um, you realize that Aaron, et al. DO NOT run the playoff odds report? They do the DVOA. Playoff odds are compiled by Mike Harris--click the statistics link at the top of the page, then click playoff odds.

Mike DOES manually adjust playoff odds for things like an injured starting QB. Aaron does NOT. Do they use each other's data? Yes--but playoff odds come from DVOA, not the other way around. Do they both use their computers to run thousands of iterations? Yes. Has Aaron cited his reasons in the commentary for using the numbers that Mike produced? Yes. 

Bottom line--DVOA says Tampa is the better team by a little bit. Playoff odds says the Chiefs are. Betting lines say the Chiefs are. I don't know if Aaron gambles--but by his commentary, he would bet the Chiefs straight up, in spite of DVOA saying the opposite.

So, your comment that "DVOA  adjustment ever made during the season are due to injury" is WRONG--those are playoff odds. Your argument that Aaron manually adjusted DVOA for the SB is wrong also. Admit you misunderstood, confused the two systems (that are very similar) and their authors, and be done with it.

Points: 0

#52 by jheidelberg // Feb 06, 2021 - 3:01pm

Yes, the DVOA ratings are not adjusted by Aaron, they are used for the playoff odds report and adjusted for injury.  I had made a comment earlier in the season that it does not seem correct that the ratings were reduced an equal amount for Jimmy Garoppolo, Michael Thomas and Christian McCaffery being out.  Aaron expressed his doubts about the adjustment for McCaffery as well.

<team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>

Aaron had commented earlier in the year that subjectively he felt that the Chiefs were the best team.  This goes against the DVOA concept and is certainly poor form.

Good form may be:  The Chiefs are clearly ranked to low because me and so many others subjectively feel that they are better, just look at the Chiefs record and the way they can put the petal to the meddle when they need to.

I added poor grammar to the multiple misspells.  I think that poor grammar used to be part of the DVOA preferred complaint format.

Points: 0

#67 by Joseph // Feb 08, 2021 - 4:19pm

Just noticed your comment in the DVOA thread, after Aaron's in-article apology.

"Well said" to both of you, and sorry if I came off harsh in my earlier reply.

Eric Fisher can use this game footage to get a big raise.

Points: 0

#12 by Led // Feb 04, 2021 - 4:53pm

From the beginning the guys at this site have said that analytics, and DVOA in particular, is very useful for evaluating teams but not the only useful tool.  If you are convinced by DVOA that TB is clearly the better team, there are certainly ways for you to monetize that information!   

Points: 0

#29 by nsheahon // Feb 05, 2021 - 3:29pm

"When the Chiefs beat Denver the second time, Nick Fangio called for a punt leading 19-16 at his own 49 with 6:18 left.  The Chiefs took over 5 minutes to kick a 48 yard FG to tie the game and give Denver a shot to win in regulation.  That is not stellar offense or "flipping a switch"

This is not how the KC/DEN game played out. KC took a 19-16 lead with 1 minute left in the 3Q. Denver punted on 4th and 7 from their own 39 . Mahomes hit Hill for a 48 yard TD that next drive that was called back for an iffy at best hold, and KC punted from their 42.

THEN Denver punted from their 49 on 4th and 3 down 19-16 (not up). Butker hit a 48 yd fg to EXTEND the lead to 22-16 (not tie it up) with 1 minute left in the game and left Denver with 0 timeouts. KC outgained them 450-330 so it wasn't a very close game and KC was terrible in the RZ.

Points: 0

#50 by jheidelberg // Feb 06, 2021 - 2:39pm

You are correct about the situation of this game, my point is that the Chiefs did win a number of games in part due to poor coaching decisions of their opponents.  Of course Andy Reid is the best coach is this division by far, so that is at least 6 games per season in which the Chiefs have a large coaching advantage.

Despite this, Denver still had a shot at winning with a minute left.   

Points: 0

#9 by Will Allen // Feb 04, 2021 - 2:05pm

"This led to a number of games where the Chiefs' win expectancy never really fell below 95% but the score ended up close because of late scores by the opposition."

I really would like to see the specific number on this, and what historical precedents there are for it, in, say, the last 30 years.

Points: 0

#13 by Pat // Feb 04, 2021 - 5:06pm

The "when KC has the ball" image is broken: the link shows no file found at that address.

Points: 0

#14 by Ridiculous15 // Feb 04, 2021 - 5:58pm

For transparency I am a life long Chiefs fan and while I try to avoid it likely subject to motivated reasoning.  With that said it is my distinct impression that when they are not in danger of losing the level of aggression goes way down and risk-aversion goes way up in both play calling and QB play.  More runs into the line, more thrown aways, more punts/FG’s on 4th down, fewer down field shots, less motion and play action, no jet sweeps.  I can’t say if that really shows up early with lesser opponents (they also have a tendency to come out with the yips against all opponents.)

I expect data is available to support or reject this hypothesis, but would take a lot of effort to gather and filter out the normal coin flip/variance noise.

Throughout Reid‘s KC tenure and even before they were really really good this has been a common complaint on Chiefs fan sites and everyone expects them to drop a handful of games that they should dominate because of this tendency.

Points: 0

#17 by PirateFreedom // Feb 04, 2021 - 7:14pm

"when they are not in danger of losing the level of aggression goes way down"

maybe DVOAAL, where the AL is for aggression level, would predict the Chiefs.

deciding the aggression level of plays seems complex, even before adding time and score factors, but some first approximation should be possible. 

Points: 0

#15 by nat // Feb 04, 2021 - 6:12pm

The KC offense chart is not showing.

Is there something broken in DVOA-land? Not the “making stuff up for the playoffs” kind of broken. But the “our data is actually missing” kind of broken.

Points: 0

#18 by Boots Day // Feb 04, 2021 - 8:49pm

It's funny to say, because Brady is a legend and still a very good player, but the biggest advantage the Chiefs have is at QB. Mahomes is arguably the best QB (and best player) in the league, while Brady has descended to the Derek Carr/Matt Stafford level.

Brady has also been unimpressive in his season-ending games the past two years (divisional loss to the Titans, Super Bowl vs. Rams), which I think is to be expected when a 40-plus-year-old player has toughed it through 17 or 18 games. My prediction is that Brady throws a couple interceptions while Mahomes is fairly mistake-free, and the Chiefs win a shootout 37-30.  


Points: 0

#20 by ImNewAroundThe… // Feb 04, 2021 - 11:49pm

Because he's still good. At least better than those guys.

Points: 0

#21 by crw78 // Feb 05, 2021 - 11:21am

Mahomes is clearly the better QB at this point, but I think you're selling Brady's performance this year a little short. He finished 3rd in DYAR (1518), well ahead of Carr (7th, 925) and Stafford (14th, 684). He finished 4th in DVOA (25.0%), ahead of Carr (9th, 14%) and Stafford (14th, 7.7%). QBR is closer (Brady 11th, 72.4, Carr 12th 71.2, Stafford 16th, 68.8).

If we look at traditional stats (yards, TDs, etc), Brady performed better than Carr and Stafford as well, though he did throw a few more interceptions.

Part of this of course is the superior supporting cast Brady had compared to them, but I'd be surprised if any team that needed a QB for one year to win a SB would have chosen Stafford or Carr over Brady given his performance this year. And the fact that he did it at 43 and it's just something we haven't seen before.

You are definitely correct though that his season-ending games have not been good the last couple of years. Even in the game against the Packers, he was great in the first half but terrible in the second.

Points: 0

#24 by Aaron Brooks G… // Feb 05, 2021 - 12:28pm

Part of this of course is the superior supporting cast Brady had compared to them, but I'd be surprised if any team that needed a QB for one year to win a SB would have chosen Stafford or Carr over Brady given his performance this year.

Decisions made in retrospect have a high degree of success.

Still, there are unknowns. We did not have an opportunity to conduct a test where Carr-Stafford-Brady all play a season for each team.

An interesting off-season project may be analyzing QBs by their win/loss splits. Some QBs are terrible in their losses; their losses are because they failed. Some QBs are pretty good in their losses; their teams lost because their defenses sucked.

\I'm still trying to find a good example of a QB who sucked even in his wins, but was carried along by his defense/running game
\\No, it's not Tebow or Boller, I checked

Points: 0

#25 by theslothook // Feb 05, 2021 - 1:03pm

How are we defining QBs who sucked? Turned the ball over a lot or scored very few points? I think even defining that a QB sucked in the data, unless its super obvious like they turned the ball over a bunch and didn't score any points, is not so easy.


Did Brady have a great game, a good game, a mediocre game, or a bad game last week? What about against the Saints?

Points: 0

#27 by Aaron Brooks G… // Feb 05, 2021 - 2:30pm

I was eyeballing TD/INT ratio and YPG as a proxy. Era-adjustment will be hard, especially with guys who spanned epochs. Hell, I think Blanda shows up in at least three distinct eras across two leagues.

Brady's last two games have been okay or worse, where he was fortunate that Brees played worse and GB kicked two FGs, because Rodgers outplayed him. The game Brady played against GB was consistent with the kinds of games he plays in losses.

His wins are usually something like 2.5-0.5-270. His losses are 1.25-1.25-235. 3 INTs is the 7th-most in his career; he's played 345 games. He's actually 3-1 in 3+ INT games in the playoffs, but 0-9 in the regular season. He lucked into playing SD in two of those games. So yes, the GB game was a bad game. Career-wise, he's 3-10 in 3+ INT games.

Just as an example, Rivers is 2-0.5-250 in his wins, and 1.5-1.25-270 in his losses. He's a better QB in his wins, but he's still pretty good in his losses. His losses are about league-average play during his era. Marino was like this, too.

Points: 0

#33 by Bob Smith // Feb 05, 2021 - 4:58pm

Aaron-wouldn't Favre be a better comparison to Rivers when comparing level of play in PO losses ? But Marino is a perfect comparison to Rivers when comparing level of play in championship game (Conf. and/or S.B.) losses however. Consider this: Rivers had 0 TD's, 2 INT's, 211 yds. passing, Rating of 46.1, and he led his offense to only 12 pts.    Marino had 4 TD's and 6 INT's, avg. of 278 yds., avg. Rating of 59.4, and he led his O to an avg. of 12 p.p.g. also.                                                                                                                                                                                                                    For comparison sake, Favre had 9 TD's and 7 INT's, an avg. of 277 yds., an avg. Rating of 78.9, and he led his O to an avg. of 25 p.p.g. I doubt if many others could beat those stats in their Champ. game losses. 

Points: 0

#31 by DisplacedPackerFan // Feb 05, 2021 - 4:08pm

Even in the game against the Packers, he was great in the first half but terrible in the second.


I'm being nitpicky (about the word great)

I'm not sure "great" is the right word for the first half.
Before the Rodgers INT that allowed the last 27 second drive he was 10 of 17 for 151 yards and a TD with a 5 yard sack. 52 of those yards were on that crazy Godwin catch on a mediocre throw.
Before the last second 39 yard TD on a horrendous defensive play call he was 12 of 21 for 163 yards and a TD. 
That's OK.
Even with that TD his actual stats of 13 of 22 for 202 and 2 TD is good it's not great.
They punted twice in the first half and both were directly related to strings of incompletions and the sack.
In the 2020 NFL punting twice in a half isn't great.

And yeah the the 7 of 14 for 80 with 1 TD and 3 INT in the 2nd half is awful. Note the TD was a one play drive for 14 yards after a fumble recovery.  

Brady was carried to this victory he didn't lead them to it. Which happens for every great QB but as others have pointed out Brady has some very striking performances where the team bailed him out. To his credit he has also played some amazing games that are the primary reason for his team winning. I don't like to admit it but he is the GOAT.


Now I'm being grumpy, though I think fair, about the game.

As I've mentioned several times in threads relating to that game. The Tampa Front 7 (sometimes 6 since they played 5 DB at times but folks generally know what the term means the DL and LB) should get the majority of the credit for that win. That fumble to start the second half, a great play by Whitehead. The Rodgers INT there was some pressure and the uncalled defensive hold had the throw not perfect. Those two plays were HUGE without them as they lead to 14 points within less than a minute and a half of game time. It was a lot more real time thanks to the half being in between them. But a 28 - 10 lead is a lot different than a 14 - 10 lead or even a 17 - 14 deficit. That was almost all because of the TB front 7.  Brady's 3 INT all happened after that, so in the what if world they are still more likely to occur with a changed game flow than in other situations. 

I also credit the TB front 7 for part of the reason the Packer's play calling looked bad in the red zone. Rodger's knew he couldn't run away from them from how things went in the first game when he scrambled and I think that was part of why he didn't run on that one pass at the end. Rodgers was a little quick on the trigger and throwing a littler harder than normal on all those redzone incompletes. I credit the TB front 7 for that. 

The next largest share should go TB coaching staff for prep and in game play calling. The TB offense was mediocre at best. They were given great field position because of the defense and special teams. They executed in those situations and just well enough in the non advantageous situations. Then they were bailed out after the 3 turnovers by the defense. They were bailed out by both coach staffs TB making good calls and GB making bad calls. 

Points: 0

#34 by theslothook // Feb 05, 2021 - 6:02pm

I find this NFC championship game fascinating because it shows the incredibly fickle nature of how legacies are defined.

In a way this game really came down to two things that went against the Packers beyond the horrendous blown coverage at the end of the first half or the Brady lame duck that Godwin rescued from the DB or even the Aaron Jones fumble. All of those things can happen in any game.

You forgive and forget all of that if Rogers manages to score a touchdown on six passes on two goal line possessions or score anything meaningful after two more Brady ints. 

That I think is really the point here. If even one of those events changes, ie the Packers score on one of the ints or on one of the red zone trips, we start debating Aaron Rodgers' place in history Brady gets the never ending speculation about his future.


This post is really aimed at people who it caught up in spinning narratives ex post. When narratives get completely turned on their heads at the whim of several high leverage moments that can easily go the other way, it's worth reconsidering whether your whole world view is flawed.

I don't judge Aaron Rodgers the quarterback based on six plays in this game. Don't judge him for his win-loss record in NFC championship games or his personality or anything like that. It's not hard to imagine an alternate universe where Rogers career ends up with multiple super bowl  championships he gets regarded far better than he will today

Points: 0

#40 by DisplacedPackerFan // Feb 06, 2021 - 12:36am

I don't disagree with that. The Packers lost and deserved to lose because they had about 15 plays that could have turned the game around and failed on all of them. As I said I attribute a lot of that to the TB front 7. Mahomes being more mobile than Rodgers and having the off platform throwing ability of a 2011 Rodgers will make a difference, as will having Kelce and Hill vs just having Adams. I suspect the Chiefs offense is not going to look quite the same in the SB as people expect and it won't be mysterious. It just won't be quite as bad of a match-up as it was for GB.

Anyway, yes if the Packers had won Rodger's legacy would be different, it will be still be hugely different if he somehow manages to win another Super Bowl next year too. It would have impacted Brady, but not as strongly as it would have impacted Rodgers though. 

Part of why I begrudge Brady the GOAT is that he doesn't play for my team, but I don't really question it anymore. What he did with TB this year was enough for me to separate him from Belichick which was my last lingering doubt. Obviously that many years together there are all the effects of learning, etc. But we did get to see him with a completely different system and coach and he looked like Brady. He had some great games he had some poor games. Try to age adjust some of that and yeah he's really damn good.

When I think about GOAT there are a few things.
Longevity: Any QB who is in consideration for GOAT has to play sufficiently long.

Separation for supporting cast: Generally playing long enough will allow you to have multiple supporting casts. You can get an idea of what the player is like independent of the players around them. This is actually way easier in the modern era where the salary cap forces moves. Brady while playing most of his career with BB got to play with a lot of different casts and he exceled with all of them.

Separation from coaching/system: I think every QB that is in a serious conversation about being the GOAT at this point, due to needing a sufficiently long career, will also have played with multiple coaches and likely multiple systems. Brady finally got this, but he also got to play in several systems without going to TB. Though the move to TB might help people realize it. BB was good at designing offenses around the talent he had, and that talent changed a lot. Because of how well Brady played in the multiple BB systems and then in the Arians system that speaks very highly of him. This is actually a bit of a knock on Manning oddly since he was often considered his own offensive coordinator so he always played in the same system. The same knock might apply to Otto Graham since he played so long ago and to Unitas. Of course being your own OC also is a plus, but it does make it harder it see how flexible that player is. Of course if you are as good as Manning does it matter if you can only play in the system you control?

It's not just passing Being the greatest QB is also more than just being the greatest pure passer. This is why I never really had some of the trouble with the Manning/Brady differences. Manning threw a better pass, but there is more to being a QB than that. Both were also insanely good at those skills too, but in different ways.
Just like the GB GOAT discussion with Favre/Rodgers/Starr (none of whom should be in the GOAT convo BTW). You've got era adjustment issues with Starr but with Favre/Rodgers it's clear that Rodgers is the better thrower of the football. Favre was great and had a cannon but was "sloppier". I think Favre was clearly better at broken plays though, as good as Rodgers is Favre was one of the best ever. Though part of why I think Favre was better was because he was more willing to take the risk with the throw. Rodgers is more likely to take the sack, both finally got much better with the throwaway later in their careers. So Manning being the better pure passer and having the better physical tools than Brady, doesn't mean he was the better QB. Besides for the GOAT convo, Brady developed some of those skills later. He may still be more likely to throw a duck but he can also complete 50 air yard passes so it's not like he is severely limited. Also he generally knows what he can't do and makes up for it by making the throws he can. As has been pointed out, more options isn't always better options. I will say one of the joys of watching Rodgers is that he almost never throws a duck. But there are times where he clearly doesn't take the throw he should have. I think the MLF system actually helps him with that.  

Wins (ugh) matter and they don't matter. Brady as I just discussed in the other post was not the reason his team won. But he clearly was good enough that they were in the position for him to be that reason.
EVERY good QB will win games despite their performance and it will happen in big games and little games.
EVERY good QB will lose games despite their performance and it will happen in big games and little games. 

Being prepared and skilled enough to take advantage of the random stuff out of your control is a thing, and not just in sports. Distribution of those opportunities is not equal, but there is skill in taking advantage of them. Brady did do that in the NFCCG several times. Heck Arians was quoted as saying Brady is the one who said to get back on the field and go for it at the end of the first half. I hate it, but it counts. If Brady wasn't really good in general then we aren't debating one of his bad games in a high visibility setting.

Who had a career like Brady? The closest I can think of is Joe Montana and not just because of the "rings". Joe was generally not thought to have quite as good of a physical tool set as the best of his contemporaries. Like Brady who that was varied because of playing through a few eras. He also developed a bit from a "system guy" to someone who really could play. While he didn't take KC to a Super Bowl in his final seasons what he did there helped dispel some of the "it was all Walsh/Rice/West Coast system" that swirled around him too. Of course it's not like KC was a bad team before and after he played there either but it did help Montana's legacy with some. Taking multiple teams to conference championships helps anyone's legacy, Super Bowls even more because Legacy is more than just skill it's the story around the skill. It helped Favre, it helped Manning, it got Warner into the HoF. It helps cement the GOAT for Brady. 

Points: 0

#46 by Raiderfan // Feb 06, 2021 - 10:50am

“BB was good at designing offenses around the talent he had, and that talent changed a lot.”

I have never seen anything—and,of course, stuff has been written about NWE ad nauseum—that suggest BB designed any of the offenses.

Points: 0

#35 by theslothook // Feb 05, 2021 - 6:06pm

I find Brady the goat has taken on an interesting discussion.

It reminds me a little bit of Bill Russell and his 11 championships. the number is so absurdly high and distances itself quite a bit from his next competitor that it becomes a kind of concession point to a different conversation.

Why do I say Brady gets the Bill Russell treatment instead of the Jordan treatment? 

Because I can already hear narratives from popular commenters who say yeah Brady is the goat but what Rogers or Mahomes is doing is just better. Even today people respect insane physical ability over Brady's rinse and repeat microwave style of offense.

Unfortunately for Tom Brady, I can already envision a world where Mahomes ends up with three or more super bowls people say he was the better quarterback.

it's ironic in a way but I think what really hurts Tom Brady here is the presence of Peyton Manning who he had to co-share the spotlight with for much of his career.

Points: 0

#36 by young curmudgeon // Feb 05, 2021 - 8:30pm

But had Brady not had Manning with whom to share the spotlight (can one "co-share" or is that redundant?), we would have missed out on the hours of fun reading or even participating in the famous Irrational Brady-Manning thread.  I still stand by my own contribution to that thread, which in a rational world would have been the last word: "Except for the aspects of play in which Manning is the superior quarterback, Brady is better.  And vice versa." 

Points: 0

#37 by theslothook // Feb 05, 2021 - 9:11pm

I probably spent more time thinking about the Brady Manning debate then what was advisable or even healthy.


I think the problem with this debate was it was foolish when it started. Only a truly blind Patriots fan could have watched the two play Early in their careers and concluded Tom Brady was better than Peyton Manning. In fact you had to hand wave a lot of things before making that claim seriously.

I've watched pretty much all of Tom Brady's career. I can't remember another player who had a career trajectory quite like his. He really did begin as game manager and unthinkably morphed into something ridiculous. It's seriously unprecedented.

This is always a fun exercise for me because I think the question of who is better than whom ultimately comes down to if both are available in the draft again who are you taking? maybe this becomes more of a discussion if Manning doesn't hurt his neck and plays just as long as Brady does. We will never know honestly, but Brady's longevity is going to tip the scales and should. 

At their peak, I think it comes down to personal preferences and how you adjust for supporting casts in both directions. Manning on a aggregate had better skill position talent for longer parts of his career than Brady, though offensive line coaching, and defense probably slant in Brady's direction.


The reason I say Tom Brady is going to get the Bill Russell treatment instead of the Jordan treatment because unlike Jordan there was always a debate about who the best quarterback was during that era. I think had Peyton Manning not been in the league, we probably remember Tom Brady more in the jordanesque light. consider that Patrick Mahomes right now is considered unquestionably the best quarterback in football and looks to be that way for years to come. 

Points: 0

#61 by t.d. // Feb 07, 2021 - 3:35pm

Manning never had a Gronk/Randy Moss-type (in the discussion for greatest ever at their position);  Brady's had that for 2/3 of his career (and all of the "his stats are as good as Manning's" part of his career).  Hell, Mike Evans is better than any receiver Peyton ever had.  Having said that, there was an argument when Peyton retired, but Brady's 4 SBs in 5 seasons since pretty much ends the discussion.  I do think that Mahomes passes him, and maybe much sooner than you'd expect, and the 'wow' plays will be a big part of it

Points: 0

#62 by theslothook // Feb 07, 2021 - 4:49pm

"Mike Evans is better than any receiver Peyton ever had"

As of right now, there is no way I see Evans as better than Harrison, who was an all time receiver IMO. 

As for Mahomes, I would like to see his evolution and consistency after Kelce and Hill are gone/aged. This same offense turned Alex Smith into a probowler. Mahomes is awesome but he won't have this embarassement of riches forever.

Points: 0

#63 by Aaron Brooks G… // Feb 07, 2021 - 5:24pm

Manning also had a season with Marshal Faulk.

‘Course, the Chiefs should also have Kareem Hunt. This offense was intended to be even better.

Points: 0

#57 by coltsandrew // Feb 06, 2021 - 7:43pm

As can probably be guessed, I'd argue that Manning is the GOAT ( and, less likely to be guessed, that Brady isn't even in the top ten greatest). That said, the real problems in the debate are that a) the sample size of games is small and b) there are too many compounding variables.

Baseball is almost the ideal sport for statistical analysis because a season is 160+ games long, especially if you make the postseason, the roster is smaller, and you have fewer starters (plus your can isolate the performance of batters). Football has  one-tenth the games, 2.5x the starters and 3.5x the number of rostered players. All that is to say, it's extremely difficult to rank football players simply by using stats because their performance can't be isolated at all, and because the relatively small number of games means that luck on high-leverage plays has an outsized effect on cumulative stats. As such, it's more than fair to rely on tape and scouting when analyzing QBs because stats just aren't robust enough to be useful substitutes for the eyeball test.

I guess the upside is that football provides an endless fodder for debate.

Points: 0

#66 by Scott P. // Feb 08, 2021 - 3:29pm

The confounding variables is a significant problem, but not the number of games, as there are so many more events in a football game than a baseball game. Brady has more passing attempts than Alex Rodriguez had at-bats, for example.

Points: 0

#68 by theslothook // Feb 10, 2021 - 12:15pm

If anyone cares to admit, its an unknowable exercise. If Aaron Rodgers is placed onto New ENgland and then Tampa, does he win as many SBs? Half? More? We don't know.

Its not as if every single victory was derived entirely from offense - that would at least let us narrow it down to one side of the ball. 

At some point, we have no choice but to go with how the results played out in this reality. And in this reality, Brady has 7 rings and has been part of a well functioning offense for over 2 decades. That's pretty unmatched so its compelling enough to view Brady as the goat, even if you could make an argument for others at various peaks. 

Points: 0

#69 by Bob Smith // Feb 13, 2021 - 8:43am

I agree with your conclusion above with one caveat-don't just count the Rings-how did each QB  play in his championship games ? I look at all championship games-Conference and League (S.B.). That is where Brady, Montana, Bradshaw, Starr, Elway, the Mannings, Favre, etc., really shine. They played good to very good a majority of the time in those championship games.                                                                                                                                                                  If the study on Value given in the playoffs was up to date, the top 10 or 20 would really tell us something.                                                                                                                                                                     All we would have to do is compare the Value with the Weighted Value to see who stands out since the Weighted Value reflects championship games.

Points: 0

#32 by Boots Day // Feb 05, 2021 - 4:33pm

I'm not saying that anyone would prefer Carr or Stafford over Brady, but I think he's closer to that tier than he is to the Mahomes-Rodgers tier. I know Brady did very well in DVOA, but the metric I usually look at is adjusted net yards/attempt from PFR, and Brady is indistinguishable from Carr by that measure. (Stafford was a bit lower because he takes so many sacks, although he does beat  Brady on simple yards per attempt.)

And of course, as others have mentioned, neither of those guys had near the weapons that Brady had in Tampa. 


Points: 0

#30 by theslothook // Feb 05, 2021 - 3:55pm

I'll make my prediction.

On paper, the Bucs are probably with the Saints as the most talented overall team in the league. Their receiving core is awesome. Their offensive line is great. Their pass rush is great, linebackers really good and safeties solid. CB is a clear weakness on this team.

KC is KC. 

Ultimately, on paper, I think the Chiefs offense unstoppable in todays NFL. They have broken the league. But it appears Brady has brought the Foxborough magic to Tampa Bay, where multiple ints doesn't kill you and the opposing offense inexplicably melts down. 

Tampa Bay 34 KC 28

Points: 0

#38 by Raiderjoe // Feb 05, 2021 - 10:30pm

Even if chiefs win, big whoop. Raiders still will have more Super Bowls won since 1976

 And thenChiefs dog and pony show ends later in 2021 when Raiders rise to claim AfC Wets divisioj crown. 

Points: 0

#39 by theslothook // Feb 05, 2021 - 11:29pm

Come on Joe...Patty and Reid aren't going anywhere. There's optimism for the Raiders but also confronting reality that is the life in the AFC West. 

Points: 0

#48 by Scott P. // Feb 06, 2021 - 12:51pm

Mahomes could certainly go to more Super Bowls, but I'm surprised how confident everyone is that that will certainly happen. Rodgers went to a SB, never got to another one. Ditto for Brees. Wilson got to two straight, hasn't been to another. Even Manning only made it four times. It's hard to get there!

Points: 0

#55 by coltsandrew // Feb 06, 2021 - 7:28pm

I have the distinct feeling that the Bills are going to four-peat a championship loss, but this time it'll be conference championship game losses to the Chiefs. I kid, but really, barring major management miscues or terrible injuries, does anyone expect any teams other than the Chiefs and Bills to have a shot at dominating the AFC for the next 2-3 years?

Points: 0

#56 by theslothook // Feb 06, 2021 - 7:36pm

The thing is, a lot depends on how the Chiefs restock their talent once this iteration ages. The offense is otherworldly in part because it has two Hall of Fame skill players working in tandem. If age or injury chips away at them, it will have ripple effects. At the very least, Rodgers showed otherworldly QB play isn't enough to dominate an era.


Also someone else will rise that we don't see coming. It always happens. Maybe that's Indy :)

Points: 0

#51 by theslothook // Feb 06, 2021 - 2:46pm

You are right. Just because they look unbeatable now does not mean they are etched in stone as the next dynasty. I will say that the big difference between Manning, Rodgers etc is that this combo pairs elite qb with hall of fame coach. Yes technically Manning had a hall of fame coach in Dungy, but that was a brief period and Dungy was no Bill Belichick. 

Points: 0

#41 by bluemetal // Feb 06, 2021 - 6:31am

I'd like to offer an explanation for some of Brady's interceptions since midpoint of the season, when the offense got in some rhythm at least. I have seen him chuck the ball 30 or 40 yards on 3rd down from midfield, which results in an interception sometimes. However, some of these are probably no worse than a punt in actual yards. An interception caught at the 20 or 25 yard line is really almost like a punt by the QB. 

I have not gone and counted / analyzed these. But think its not a bad strategy, even if it results in bad numbers for him optically. The pass can be caught by the WR, could get DPI, could fall incomplete,or could be intercepted. Lots of upside and an effective punt on the downside  



Points: 0

#44 by nsheahon // Feb 06, 2021 - 9:41am

But it doesn’t work at a high rate and just converting on 3rd is more efficient. I think when he gets pressured he just chucks and prays which is why he is so bad against pressure.

Points: 0

#53 by ChrisS // Feb 06, 2021 - 5:07pm

I think long interception returns are more likely than long punt returns. Too lazy to try to find evidence, so this is completely subjective

Points: 0

#42 by BJR // Feb 06, 2021 - 9:14am

I've already vented on the 'flip the switch' theory on the DVOA thread. At least here it is presented as just that - a theory - rather than a cast iron reason to fudge the numbers. Still I find it extremely vague and speculative. Why  would the Chiefs have jeopardized the #1 seed (at a time when it has become so incrementally valuable) by playing far beneath themselves for long periods of the season?

Given what we know about the volatility of defensive performance, I'm not prepared to take a 2 week sample as evidence of flipping the switch. The write up does actually presents a valid reason why the Chiefs might have 'underperformed' on offense when they had big leads: they are not very good at running the ball. Once the offensive equation shifts from chewing up yards to chewing up clock, they become less effective. The obvious antidote to this, given how great their passing offence is, would be to say 'to hell with burning clock, lets keep slinging'. But whilst Andy Reid has undoubtedly gravitated further and further towards the aggressive end of the play-calling spectrum, he is still a long way from abandoning the run entirely. 

This game on paper looks like one of the clearest invitations to the 'pass on every down' strategies for the Chiefs. I cannot see any reason why they should run the ball at all until actual, actual garbage time, or unless there is some extremely obvious strategic reason to do so. I'll be keenly keeping a mental note of the Chiefs rushing attempts, and their success, throughout this game.

Points: 0

#45 by nsheahon // Feb 06, 2021 - 9:44am

You don’t really risk your #1 seed by giving up garbage time TDs and your win prob never gets under 80%. They had some bad games (like any team) but when they absolutely controlled NO, DVOA said NO was actually better due to the last part of the game, while KC had it in hand.

Points: 0

#47 by young curmudgeon // Feb 06, 2021 - 12:19pm

I'm not certain the "flip the switch" theory really accounts for what's going on.  What follows is impressionistic and anecdotal, rather than statistical or analytic, and I apologize for that: (1) No athlete, or anyone else, can perform at peak efficiency constantly and indefinitely.  Football season is long, there's a lot of opportunity for random variation to show itself.  There are going to be times when a player, a unit, or even an entire team simply aren't 'firing on all cylinders.' 

(2) Football, like many other sports, is very emotional and motivation can stem from emotion.  Sometimes, highly emotional players will be highly motivated, and will perform extremely well (although sometimes, too much emotion or motivation can be counterproductive.)  Conversely, when emotions are low, it can be hard to motivate oneself to reach peak efficiency (and remember, no one can always or automatically reach that peak). 

(3) When the negative effects involved in 1 and 2 coincide, the player or team is going to underperform (as, e.g., the Steelers have demonstrated in at least one game per season for a number of years now.) 

(4) Effort is difficult to sustain.  Two good examples come from basketball: (a) a team is 17 points down in the second half and mounts a furious effort to come back, finally tying the score with two minutes left.  They end up losing by 8.  (b) An underdog team stays just behind its superior opponent all game and, in the last minute, hits a shot that sends the game to overtime.  They also end up losing by 8.  My point is that maximum effort is draining and sometimes the effort is used up before the game reaches its end. 

Perhaps this is all just a way of elaborating on the "flip the switch" theory, but I think it is likely that the variability in Kansas City's performance comes from a combination of random variation, diffusion of motivation, and inability to sustain 100% effort 100% of the time.  In addition, some have suggested that they hold some of their best (trick, gadget, unexpected) plays back for times when they are more needed, and that makes strategic sense.  But if they are doing what looks like "flipping the switch," I suspect it's coming from normal and unavoidable aspects of a demanding team sport. 


Points: 0

#54 by jheidelberg // Feb 06, 2021 - 6:40pm

How many losses does it take to end this theory? Just one?  Flipping the switch works until it gets stuck in the off position.  Now that Brady is not a Patriot, it is no longer egregious for me to root for a team led by Tom Brady.  Thus, I am rooting for Tampa to turn the KC switch to off, at least for the time being.

Points: 0

#59 by MC2 // Feb 06, 2021 - 8:45pm

The big problem with the "Flip the Switch" theory is the dubious assumption that Kansas City is the only team that drastically changes their strategy when they have a big lead. 

In fact, virtually every team switches to vanilla offense and prevent defense when they have a big lead.  They aren't running trick plays or exotic blitzes when they're up by 3 TDs in the 4th quarter.  Even in cases like Belichick's Patriots, who are notorious for leaving their starters in until the very end, they still get more conservative with the play-calling (on both sides of the ball).  In fact, I can't remember a single team that has consistently "kept the pedal to the metal" with big leads. 

The only difference is that most teams run "vanilla stuff" more effectively than the Chiefs.  Or, to put it another way, the Chiefs struggle to ice games.  Is that flaw fatal?  Probably not.  But, fatal or not, it is a flaw, and it makes very little sense to pretend it doesn't exist, or that it's completely irrelevant.

Points: 0

#60 by SportsPhan8 // Feb 07, 2021 - 3:59am

Can you please also include/add regular season + postseason to date DVOAs and weighted DVOAs in the first table?

Points: 0

#64 by ImNewAroundThe… // Feb 07, 2021 - 10:47pm

this time

Points: 0

#65 by trfields // Feb 08, 2021 - 11:02am

Play by play measurement is more accurate.  There is no need to change methods to fit the market predicted winner based on winz (KC 14-2 vs TB 11-5).  KC was burning money all season not meeting market expectations, winning games on the positive side of variance, not performance.  Aaron made a mistake, learn from it, trust the process, trust the methods, trust the numbers. 

Points: 0

Save 10%
& Support Aaron
Support Football Outsiders' independent media and Aaron Schatz. Use promo code SCHATZ to save 10% on any FO+ membership and give half the cost of your membership to tip Aaron.