On paper, the Buccaneers added a lot of big names in the offseason. Tom Brady and Rob Gronkowski from the Patriots, a former top-5 pick at running back in Leonard Fournette, and an All-Pro wide receiver in Antonio Brown. Tampa Bay also hit home runs with its first two draft picks, offensive tackle Tristan Wirfs and cornerback Antoine Winfield Jr..

But the Buccaneers did not have a particularly large amount of turnover compared to recent Super Bowl champions. In fact, by one measure, each of the five previous Super Bowl champions had bigger turnover than Tampa Bay. Other than Brady, most of the big names didn’t have a big impact for Tampa Bay, at least until the Super Bowl. Leonard Fournette had just 367 rushing yards during the regular season; Brown had just 483 receiving yards in 8 games; Gronk averaged 39 yards per game, the lowest mark in his career other than during his rookie season (in ’19, O.J. Howard averaged 33 yards per game for Tampa Bay). Of course, in Super Bowl LV it was a different story: our lasting image of the 2020 Bucs will be Gronkowski scoring 2 touchdowns, Brown catching a touchdown, and Fournette picking up 135 total yards and a touchdown.

How do we compare one team to its squad from the prior year? Great question! It’s tricky: we want to compare not just roster turnover, but the value of those players. The Chiefs added both Brady and LeSean McCoy, but that should not count as equal moves. Similarly, you have to look at how the player was used in the prior season. For example, Kurt Warner was on both the ’98 Rams and the ’99 Rams, but it would be poor analysis to say that the ’99 Rams had the same quarterback as they did the prior year.

Really we want to look at two things: how much was the player a part of the Super Bowl team (to avoid overemphasizing the McCoys of the world) and for players on the team, how much was that player a part of the team the prior year (to avoid the Warner problem). To measure how valuable a player was, I am going to use PFR’s AV, which does as good a job as any other metric to compare the value of players across positions and eras. For each player on each Super Bowl team, I will measure how much AV they produced as a percentage of the team’s total AV. I will also do that the year before, and then take the lower of those two values.

Let’s use a few examples. The 2020 Buccaneers had 243 points of AV, and Brady had 15 of them; therefore, he was responsible for about 6% of the team’s value. He was not on the team in 2019, of course, so Brady gets a 0 when we calculate how much of the 2019 Bucs were on the 2020 Bucs. That’s a big deal, because there is only 94% of the AV left to allocate. McCoy had just 1 point of AV, or 0.4% of the team’s AV; he also gets a zero since he wasn’t on Tampa Bay in 2019, but that has a much smaller impact. Donovan Smith had about 4% of the AV on the 2019 Bucs and 4% of the AV on the 2020 Bucs; therefore, taking the minimum doesn’t change things; he gets assigned 4%.

If we do this for every player on the 2020 roster, and take the minimum value between the percentage of AV they had for the ’20 Bucs and the percentage of AV they had for the ’19 Bucs, and then sum the results, we get a value of 60%. This may not mean a ton to you in the abstract — it wouldn’t be quite right to say that 40% of the roster turned over — but it is very helpful if we are comparing teams. [continue reading…]

Clyde Edwards-Helaire has had a really good 13 months. He’s gone 13-1 as a starter for the 2020 Chiefs, including a win in the AFC Championship Game. This is a nice continuation of his time at LSU, where Edwards-Helaire went 15-0 as the leading rusher for the Tigers. Since January 1, 2019, Edwards-Helaire has gone 29-1 in his 30 games.

CEH is on the verge of winning the national championship and the Super Bowl in back to back seasons, a pretty rare feat. There have been 14 players to win a national championship and an NFL championship in back to back seasons. ^[1]The NFL qualifier excludes two other seasons. In 1947, when Frank Kosikowski was on Notre Dame’s back-to-back champions in 1946 and 1947; as a rookie, he played in the AAFC with the Cleveland … Continue reading [continue reading…]

The 2014 Patriots were not very reliant on the ground team. Jonas Gray led the team in rushing yards with just 412 in 8 games, while Shane Vereen was second with 391 over the full season. Stevan Ridley had 340 in 6 games, while LeGarrette Blount had 281 in five appearances. Even as a team, New England only ranked 18th in rushing yards. And while Blount ran all over the Colts (30/148/3) in the AFC Championship Game, he was held to just 40 yards on 14 carries in the Super Bowl, which still managed to lead the team.

Among the now 110 teams to make the Super Bowl, the 2014 Patriots rank last when it comes to rushing yards gained by their leading rusher. The 1998 Broncos, of course, rank first: Terrell Davis rushed for 2,008 yards that season. The graph below shows the rushing yards (pro-rated to 16 game seasons for years with shorter schedules) for the leading rusher on each Super Bowl team: the winners are shown in black, the losers in green, and the Bucs and Chiefs are shown in team colors. [continue reading…]

Since the start of the 2010 season, no player has more rushing yards than LeSean McCoy. It is easy to forget, but McCoy spent some time as the starting running back for the 2019 Chiefs, even though he did not get a carry in the playoffs. McCoy rushed for 465 yards and 4 touchdowns in the regular season and did earn a ring with Kansas City and his former head coach.

It might be even easier to forget that McCoy is a member of the 2020 Buccaneers. He was signed in early August, before Leonard Fournette would make his way to Tampa Bay.  McCoy has been the team’s fourth string running back, behind Ronald Jones, Fournette, and Ke’Shawn Vaughn. But no matter: McCoy is, technically, back in the Super Bowl, for the second year in a row.

If Tampa Bay wins, McCoy — who was arguably the best running back of the 2010s — will close out his career with a pair of rings with two teams that nobody will ever associate him with. That said, he would join a pretty exclusive list. Here are the eight players who have won back to back rings with different teams.

• The two famous members of the club pulled off this trick in 1994, as members of the San Francisco 49ers, and in a surrounding year with the Dallas Cowboys. One was Deion Sanders, who left Atlanta for San Francisco in 1994, and then San Francisco for Dallas in ’95. The other was Ken Norton, Jr., who spent the first six years of his career with Dallas, and won the Super Bowl in 1992 and 1993. Norton then joined the 49ers in ’94, where he spent the final seven seasons of his career. Hochstein and Norton are the only two players to win Super Bowls in three consecutive seasons.
• Guard Russ Hochstein was the next player to do so. He was drafted by Tampa Bay in 2001 and played in one game in 2002; he was waived in October and signed by the Patriots a week later. He stayed in New England through 2008, so Hochstein picked up a Super Bowl ring for his cup of coffee with the Bucs and then earned two more the next two seasons in New England. Hochstein was also a freshman with Nebraska in 1997, when the Cornhuskers were named national champions by USA Today and ESPN. He’s got a lot of hardware!
• Defensive back Derrick Martin played in five games for the Packers in 2010 and 14 for the Giants in 2011. Sure, Martin had just 1 interception and 14 tackles during those two years, but it counts!
• Brandon Browner was the next to do it, and he had a big role in both title teams.  He was a star on the ’13 Seahawks, although he was suspended towards the end of the season and did not play in the team’s postseason run. He joined New England in 2014, and then made the play of the game — or, perhaps, the assist of the game — to help the Patriots defeat his former team. ^[1]Caylin Hauptmann was also a member of the 2013 Seahawks, and he was signed by the 2014 Patriots.  I don’t believe he should be included, but if you want to, feel free!
• LeGarrette Blount led the NFL in rushing touchdowns in 2016 with New England, picking up a ring with the Patriots that season.  The next year he joined the Eagles, and scored a touchdown in Philadephia’s Super Bowl win over his former team. Chris Long was also a rotational defensive linemen on both the ’16 Patriots and ’17 Eagles.
• Surely you don’t remember that Kenjon Barner was the kickoff returner for the ’17 Eagles, do you?  Barner, a star of Oregon under Chip Kelly, joined Philadelphia under Kelly’s watch and survived with the team through 2017.  The next season, he was signed by the Patriots and rushed for 71 yards before being cut in midseason.  No matter: when the ’18 Patriots went on to win the Super Bowl, Barner received a ring.  Literally.  Oh, and guess what? Barner is also a member of the ’20 Buccaneers, although he is unlikely to play in Super Bowl LV.  But I assume he is definitely ready for a third ring in four years.

How do Kansas City and Tampa Bay compare to previous Super Bowl participants when it comes to scoring and preventing points? Without any adjustments for era, these two teams look like awful defensive teams and very good (but not historically great) offensive teams.

In the graph below, the X-Axis shows points scored per game, while the Y-Axis shows points allowed per game. The best teams would be in the lower right, at least according to this metric. The Bucs and Chiefs are color-coded, with Super Bowl winners in black circles and losers in white circles. The ’99 Rams are in the lower right corner at 33 points per game and 15 points allowed per game; the ’13 Broncos are the team at the far upper right, while the ’11 Giants, of course, are in the upper left: they are the only Super Bowl team to allow more points than they scored. [continue reading…]

Yesterday, I looked at how both the NFL and the upstart AFL dealt with the issue of integration in pro football. The common reframe is that the young, nontraditional AFL was more innovative and enlightened than the NFL, particularly on the issue of black players in pro football and the scouting of talent from small historically black colleges. The numbers don’t bear that out with black players generally, but today I want to focus on the game that is most responsible for creating that narrative: Super Bowl IV. [continue reading…]

The last running back to win the Super Bowl MVP was Terrell Davis, which happened over 20 Super Bowls ago. One reason for that is a decline in big rushing games in the Super Bowl, particularly with respect to the winning team. Just once in the last 16 Super Bowls — Dominic Rhodes back in SuperBowl XLI — has the winning team had a 100-yard rusher. Perhaps more interesting is that in the last 9 Super Bowls, the losing team had the game’s leading rusher more than half the time.

The graph below shows the leading rusher for both the winning and losing teams in the Super Bowl. The winning team’s leading rusher is in a full black circle, while the losing team’s leading rusher is in a white circle with a black outline. In addition, in the 13 of 53 Super Bowls where the game’s leading rusher was on the losing team, I’ve put that in a white circle with a red outline. [continue reading…]

Last year, the Kansas City Chiefs were the #1 seed in the AFC, while the 49ers had the second-worst record in the NFL. This year, the two teams are in the Super Bowl. Today, let’s look at how each of the 108 Super Bowl teams fared in Year N-1 — that is, the year before making the Super Bowl.

For each team, I calculated their Pythagenpat record in Year N-1. The best Super Bowl team by Year N-1 Pythagenpat record was, unsurprisingly, the 1973 Dolphins. As we all know, the year before Miami went 14-0 and won the Super Bowl.

The 2019 Chiefs don’t fare all that well in this regard, mostly because the 2018 team ranked 24th in points allowed. Kansas City was the #1 seed, but it was not a dominant team by these standards. The 49ers, however, do stand out as particularly bad (although I’ll note that San Francisco, despite finishing with the 2nd pick in the 2019 Draft, “only” had the 6th worst Pythagenpat rating of 2018). The 2018 49ers are one of the worst 5 teams to make the Super Bowl the following season. [continue reading…]

Can you name the winner and loser from each Super Bowl?

How about the Super Bowl MVP? The final score? The points spread? The stadium?

Test your knowledge below, by filling in as much as you can. Answers appear after the jump.

[continue reading…]

[Editor’s note: this is a repost from last year, with a few updates.]

The first four Super Bowls were all played on grass stadiums: the first Super Bowl was at Memorial Coliseum in Los Angeles, the Orange Bowl in Miami hosted Super Bowls II and III, and Tulane Stadium in New Orleans was the venue for Super Bowl IV.

But the dreaded AstroTurf revolution came to football in the ’70s. For Super Bowl V, the Orange Bowl in Miami was again the host site, but by now, the natural grass surface had been replaced with Poly-Turf, making it the first Super Bowl played on artificial turf.

[continue reading…]

In Super Bowl II, Green Bay quarterback Bart Starr completed 13 of 24 passes for 202 yards, with 1 TD and no interceptions. He also took three sacks for 30 yards, and his backup was sacked once for 10 yards.

That means on 28 dropbacks, the Green Bay passing attack picked up 162 yards and produced 1 TD and 0 INTs, which translates to 142 Adjusted Net Yards. That’s an Adjusted Net Yards per Attempt average of 6.50 in Super Bowl II for Green Bay.

So was that a good performance by Starr and the Packers passing attack? No.

It was a great one. Passing efficiency was a lot different in 1967 than it is today, and the Raiders had a great pass defense. In 1966, Oakland allowed 3.14 ANY/A to opposing passers, the 3rd-best in the AFL; in 1967, that number dropped to just 2.23, the best in the AFL and third in all of pro football. And in 1968, opposing QBs averaged just 2.81 ANY/A against the Raiders. (And in 1969, it was just 2.64.)

The graph below shows every game for the Raiders defense from 1966 to 1968, in chronological order. Joe Namath consistently gave the Raiders problems, including in the famous Heidi Game in 1968. But otherwise, the Raiders pass defense tended to overwhelm most quarterbacks. On the X-Axis, we have games played in order. The Y-Axis shows the ANY/A allowed by the Raiders, shown from -6 to 14 (since the league average ANY/A was around 4.0). [continue reading…]

On January 3, 2019, the St. Louis Blues had the worst record in the NHL. But by the end of the year, the Blues had captured their first championship. There is no similar success story in NFL history, but I was curious: which Super Bowl champion had the worst record after X games, for all X?

There have been 9 Super Bowl champions that began the year 0-1. That is not a very big hole from which to escape. But only three teams — the 1993 Cowboys, the 2001 Patriots, and the 2007 Giants — have won the Lombardi Trophy after an 0-2 start. And in the first two cases, it’s noteworthy that Emmitt Smith and Tom Brady were not part of that 0-2 start, but entered the starting lineup in week 3.

What about the worst team after 3 games to win a Super Bowl? 5 teams have begun the season with a 1-2 mark; the three teams mentioned above, along with the 1981 49ers and the 2018 Patriots.

After 4 games? Just one team has started 1-3 and won the Super Bowl: the 2001 Patriots.

After 5 games? Two teams have started 2-3 and won the Super Bowl: the 2001 Patriots and the 1980 Raiders.

After 6 games? No team has been 2-4, with the ’01 Patriots, ’80 Raiders, and the 2010 Packers.

After 7 games? Only team, the 2001 Patriots, had a losing record at 3-4 after 7 games.

After 8 games? Only the 2001 Patriots were 4-4 after 8 games; the other 52 Super Bowl champions had a winning record at this point in the season.

After 9 games? No team had a losing record and won the Super Bowl; in addition to the ’01 Patriots, the ’00 Ravens were the only other team at 5-4 after 9 games that won it all in the Super Bowl era (the ’57 Lions also pulled off this feat).

After 10 games? Yes, once again, we have a field of one: the 2001 Patriots are the only team without a winning record after 10 games, as New England began this season 0-2, 1-3, 3-4, and 5-5. That was the team’s last loss of the season.

After 11 games? Three teams started 6-5: the ’88 49ers, ’01 Patriots, and a new entry: the ’11 Giants.

After 12 games? Just one Super Bowl champion, the 2011 Giants, failed to have a winning record at this point in the season.

After 13 games? The 2011 Giants, at 7-6, are the lone entry here.

After 14 games? The 2011 Giants were 7-7! Only one other team, the 2010 Packers at 8-6, had even 6 losses after 14 games.

The 2011 Giants were the only team to win the Super Bowl with 7 losses, so they are the Super Bowl champion with the worst record after 15 and 16 games, too.

So the Patriots had the single worst record, either alone or tied with other teams, of any Super Bowl champion through X games, for all values of X from 1 to 11. At 6-5, the 2011 Giants were tied with the Patriots (and ’88 49ers) for worst record, and the 2011 Giants hold the mark for worst record of any Super Bowl champion through X games, for all values of X from 11 to 16. The graph below shows the winning percentages of both teams after each game of the regular season.

What do you think?

Yes, this was the lowest-scoring Super Bowl ever. Just as the game ended, I tweeted this graph:

But this game was even more anemic than you think, once you consider the two teams in the game. Consider that in the regular season, Los Angeles averaged 32.9 points per game, just 1/16th of a point per game more than the famed ’99 Rams. In fact, the ’18 Rams were the 6th-highest scoring offense to ever make the Super Bowl, behind five other Super Bowl losers: the ’13 Broncos, ’07 Patriots, ’83 Redskins, ’16 Falcons, and ’67 Raiders. But by being held to just 3 points, Los Angeles came in at a whopping 29.9 points below their regular season average. That narrowly (by, in fact, 1/16th of a point per game) bests the ’13 Broncos as the biggest underachieving result in the Super Bowl. There have been 106 teams to play in the Super Bowl, and despite being the 6th-highest scoring team to make it, LA finished as one of just two teams to score only 3 points in the Super Bowl (no team has ever been shut out).

Edelman didn’t score and the Patriots went 3/12 on 3rd downs…. and he was still the best player on the field.

So there you have it: the Rams were the lowest scoring team (relative to their average) of any team to lose in the Super Bowl, and the Patriots were the lowest scoring team (relative to their average) of any team to win the Super Bowl. With just 16 points, this was the lowest scoring Super Bowl, of course, but it was expected to be the third highest scoring Super Bowl ever! The Over/Under was 55.5 points, which trails only the ’16 Super Bowl (New England/Atlanta, with an O/U of 57.0 points and 62 points actually scored) and the ’09 Super Bowl (New Orleans/Indianapolis, 56.5, 48). [continue reading…]

The first four Super Bowls were all played on grass stadiums: Memorial Coliseum in Los Angeles, the Orange Bowl in Miami for Super Bowls II and III, and Tulane Stadium in New Orleans.

But the dreaded AstroTurf revolution came to football in the ’70s. For Super Bowl V, the Orange Bowl in Miami was again the host site, but by now, the natural grass surface had been replaced with Poly-Turf, making it the first Super Bowl played on artificial turf.

The same thing happened the next year, at Super Bowl VI, as Tulane Stadium followed suit and resurfaced its field with Poly-Turf. And after a return to Memorial Coliseum for Super Bowl VII, the Dolphins and Vikings met in the first Super Bowl played on nylon texturized AstroTurf in Super Bowl VIII:

[continue reading…]

There have been 52 Super Bowl games played so far, but they have featured just 17 upsets. But a Super Bowl upset was once extremely rare. The Jets and Chiefs produced the first two Super Bowl upsets, in Super Bowls III and IV, respectively. But then there wasn’t a single Super Bowl upset during the decade of the 1970s! (Excluding, of course, Super Bowl IV, which concluded the 1969 season but took place on January 11th, 1970).

In the 1980s, there were 4 Super Bowl upsets, all featuring the Raiders or the Redskins (or both). Oakland beat Philadelphia in 1980, Washington beat Miami in 1982, the Raiders beat the Redskins in ’83, and the Redskins upset the Broncos in ’87. All featured point spreads of 3.5 or fewer points, in contrast to the two double digit upsets from the AFL teams.

In the ’90s, there were just TWO upsets in the Super Bowl: the Giants over the Bills and the Broncos over the Packers. When Tom Brady and the Patriots took the field as underdogs in Super Bowl XXXVI, but 8 of the first 35 Super Bowls had featured upsets. But beginning with that first Patriots/Rams Super Bowls, there have been a whopping 9 upsets in 17 games (and the denominator is really 16 here, since one Super Bowl had an even point spread and therefore couldn’t feature an upset)! In other words, we’ve gone from an upset happening less than 25% of the time to an upset happening over 50% of the time.

Beginning with the first Patriots Super Bowl, and excluding the New England/Seattle Super Bowl (which featured the even points spread), there have been 16 Super Bowls with a favorite during the Brady/Belichick era. Of those 16 Super Bowls, the favorites have lost 9 times, and failed to cover in 3 of the 7 times that they actually won! There have been just 6 times in Super Bowl history that the favorite won but failed to cover the point spread, so that means half have occurred in relatively recent history.

From 1966 to 2000, favorites went 22-11-2 against the spread during the first 35 Super Bowls. In the 16 Super Bowls since 2001 to feature a favorite, those favorites have gone just 4-12 against the spread. The graph below shows the point spread (in red) and margin of victory (in blue) for the Super Bowl winner in each game. The point spread is shown as an expectation of how much the winner was expected to win by (so the Jets in Super Bowl 3, rather than shown as +18, are shown as -18). All Super Bowl upsets feature red bars that are negative.

What stands out to you?

Today’s guest post/contest comes from Thomas McDermott, a frequent guest contributor. As always, we thank him for his hard work.

Author’s note regarding the numbers: The Win Probability (WP) numbers shown below were, for the most part, generated using a formula presented by Wayne Winston in his book Matheletics, and subsequently improved upon by Pro Football Reference for their WP model. ^[1]While in the process of messing around with PFR’s WP Calculator, I have noticed differences between the results from the Calculator and the results from using the formula as shown on their web … Continue reading