Skip to main content
ABC News
There’s More Than One Way To Build A Super Bowl Team

The Kansas City Chiefs and Tampa Bay Buccaneers will have plenty of things in common in Super Bowl LV, starting with a pair of all-time QBs in Patrick Mahomes and Tom Brady. But in many other ways, the teams’ paths to the brink of a championship could not be more different. Because of that, whoever wins will invariably sketch a blueprint for the rest of the league to follow — as difficult as each build might be to emulate.

Kansas City has been playing championship-caliber football for a while now. Starting from an existing talent base that was good enough for a 10-6 record and a close playoff loss with former QB Alex Smith in 2017, the Chiefs turned to Mahomes under center in 2018 and got to another level. Narrowly losing to the eventual champion New England Patriots in the AFC title game, K.C. — which had the league’s fifth-youngest roster1 — had nonetheless made a statement that it would be a force to reckon with in the years to come.

The 2019 Chiefs were even better after addressing some of their big weaknesses from 2018. They added key defensive contributors such as safety Tyrann Mathieu and defensive end Frank Clark, and they subsequently improved from 30th to 13th in schedule-adjusted EPA on that side of the ball. And the offensive core of Mahomes, tight end Travis Kelce and receiver Tyreek Hill remained as dangerous as ever en route to the title. As a result, the Chiefs were one of the better champions in modern NFL history, with a final Elo rating of 1745 — 24th-best among all-time Super Bowl winners.

Going into 2020 as favorites to repeat, the Chiefs clearly took an “if it ain’t broke, don’t fix it” approach to their roster. Only 17 percent of the team’s Approximate Value (AV) came from players who were not with the club a year ago, which ranked 27th out of 32 teams this season. Mahomes was the team’s top passer — and best overall player by AV — for the third consecutive season, while Kelce was K.C.’s leading receiver for the second-straight year. Mathieu, Clark and interior lineman Chris Jones were the team’s three most valuable defenders by AV, just as they were in 2019. Among Chiefs with 5 or more AV, only two — running back Clyde Edwards-Helaire and tackle Mike Remmers — were new to the team in 2020.

The Buccaneers, meanwhile, provide a nice contrast to K.C.’s team-building model. They finished last year with a 7-9 record in coach Bruce Arians’ first season, with a defense that ranked fifth in EPA but an offense that ranked 22nd.2 Going into 2020, Tampa Bay hadn’t made the playoffs since 2007 and had finished with an Elo below 1500 (league average) 13 times in the previous 16 seasons. If the Chiefs had been perennially good in recent years — 2020 was the eighth straight season they finished with an Elo above 1500 — the Bucs had seldom been able to keep their head above water for more than a decade.

What makes Patrick Mahomes so great? | FiveThirtyEight

The 2019 Bucs’ roster clearly had potential. Weighted by AV, it was actually the best collection of pure talent in the league in terms of how highly its players were drafted. But to help convert that promise to performance, Tampa Bay said goodbye to starting QB Jameis Winston in the middle of his prime, replacing him with the 43-year-old legend and 20-year veteran of the New England Patriots. (Brady’s presence alone played a large role in making the Bucs the league’s second-oldest team in 2020, as weighted by AV.)

And despite no other QB ever being remotely successful at that age, Brady put up his best passing efficiency numbers (indexed to the league) and his most AV in a season since 2017. With Brady taking Tampa Bay to the Super Bowl, this is just the 12th time in history that a Super Bowl team will start a QB who before the season had never started for the team. (Brady is on the list twice, which itself is pretty incredible.)3

New starting QBs are rare for Super Bowl teams

For Super Bowl starting quarterbacks, fewest previous career starts for the team going into the season, 1966-2020

Starts for team
Season Team Starting QB In Season* Before Season
1967 OAK Daryle Lamonica 16 0
1968 BAL Earl Morrall 17 0
1977 DEN Craig Morton 17 0
1979 LAR Vince Ferragamo 8 0
1980 OAK Jim Plunkett 15 0
1987 WSH Doug Williams 5 0
1999 STL Kurt Warner 19 0
2000 BAL Trent Dilfer 12 0
2001 NE Tom Brady 17 0
2003 CAR Jake Delhomme 19 0
2012 SF Colin Kaepernick 10 0
2020 TB Tom Brady 20 0
1990 NYG Jeff Hostetler 7 2
1971 DAL Roger Staubach 17 4
2000 NYG Kerry Collins 26 7

*Includes Super Bowl.


Beyond Brady and the fourth-ranked EPA passing attack in the NFL, the Bucs finally got their breakthrough because of a defense that ranked third in EPA overall, including fourth against the pass and eighth against the run. That’s slightly better than it was a year ago, despite the fact that only one of their nine best newcomers by AV (rookie safety Antoine Winfield Jr.) was on the defensive side of the ball. Tampa Bay’s journey to the Super Bowl is a tribute to all of those teams who have felt like they were “a quarterback away” from contender status.

The teams’ different philosophies are even reflected in how they spent their salaries this season. Although they’re known for their explosive offense, the Chiefs actually spent the second-most salary cap dollars on defense of any team, according to Spotrac, trailing only the Buffalo Bills. (K.C. spent the ninth-most of any team on offense.) And the Bucs spent the third-most salary cap dollars of any team on their offense, while ranking sixth in spending on defense.

Over the course of the season, the two teams’ competing paths took them through different places in the league’s hierarchy. While the Chiefs’ Elo improved by a modest 76 points from preseason through the conference championship — 22nd-fewest of any Super Bowl team ever — the Bucs improved by a whopping 205 points, 10th-most of any Super Bowl team. And they did the bulk of that improvement — 138 points’ worth4 — since their bye in Week 13.

The most (and least) surprising Super Bowl teams

Largest and smallest changes in Elo rating for Super Bowl teams from preseason to the start of the Super Bowl, 1966-2020

Largest Changes Smallest Changes
Year Team Pre-SB Elo Change* Year Team Pre-SB Elo Change*
2019 SF 1674 +251 1979 LAR 1566 -6
1999 LAR 1656 +248 1967 GB 1701 +15
1998 ATL 1705 +242 1974 MIN 1664 +33
1981 SF 1655 +238 2018 NE 1686 +35
1985 CHI 1768 +226 2004 PHI 1641 +40
2007 NE 1849 +216 1992 BUF 1663 +41
1969 MIN 1736 +213 1986 DEN 1630 +45
1967 OAK 1742 +209 1988 SF 1678 +47
2000 BAL 1700 +206 1979 PIT 1727 +50
2020 TB 1703 +205 1995 PIT 1643 +52
2016 ATL 1688 +202 2017 NE 1751 +63
2003 NE 1739 +200 1970 IND 1655 +68
1981 CIN 1666 +199 1987 DEN 1645 +68
1992 DAL 1759 +199 1991 BUF 1703 +69
2001 LAR 1747 +196 2012 BAL 1675 +71
2005 SEA 1677 +196 1973 MIA 1757 +72
2015 CAR 1730 +195 1995 DAL 1710 +72
1985 NE 1698 +193 2006 IND 1709 +72
1977 DEN 1727 +191 1994 LAC 1614 +73
2001 NE 1643 +189 2009 IND 1689 +74
1986 NYG 1769 +188 2015 DEN 1671 +75
1988 CIN 1626 +182 2020 KC 1741 +76
2017 PHI 1691 +180 1993 DAL 1768 +76
1984 SF 1759 +179 1987 WSH 1700 +79
1972 MIA 1760 +177 1993 BUF 1674 +80

*From preseason.


In other words, this Super Bowl between Kansas City and Tampa Bay will be a battle between a team we’ve expected to be here for a long time and a team whose rise came much more recently. It will also feature a team that had the right quarterback but improved its defense to reach a championship level versus the team that had the defense but found its QB later. And it will see a team that stuck with its core between seasons face a team that saw a major shakeup at the game’s most important position. We don’t know which formula will prevail — our forecast gives K.C. a slight edge, at 53 percent to win — but we do know that each team-building model has allowed the Chiefs and Bucs to both come within a game of the sport’s ultimate prize.

Check out our latest NFL predictions.


  1. Weighting the age of every player by their Approximate Value.

  2. That might seem backward compared with the team’s rankings in points scored (No. 3) and allowed (No. 29) last year, but the offense’s ridiculous rate of turnovers (20.2 percent) gave opponents some of the shortest fields any team had to defend. After adjusting for this, EPA implies that Tampa Bay scored fewer points than an average team would have and allowed fewer points than we would expect given the situations its defense was placed in.

  3. Brady is also one of only four QBs to start a Super Bowl for two different teams, joining Craig Morton, Kurt Warner and Peyton Manning.

  4. Which by itself would rank 49th.

Neil Paine was the acting sports editor at FiveThirtyEight.