Happy Independence Day, folks. July 4th, 1776 was the day our forefathers declared independence in a remarkable document that’s worth your full read.  What’s known in America as the Revolutionary War began in earnest in 1775, reached ***official*** status as a revolution on this day in 1776: that was when the Declaration of Independence was adopted by Congress.  Seven years later, the British surrendered, and the war is still known there as the American War of Independence.  But let’s not give the British too much grief: after all, Britain was the runner-up in the Revolutionary War, which earns them a silver medal in that competition.

So today, let’s look at the best runners-up in NFL history. Two teams have lost in the playoffs to the eventual champion in four straight years.

1990-1993 Bills; 1967-1970 Raiders (4)

The early ’90s Bills famously lost in four straight Super Bowls, to the Giants, Redskins, and then twice in a row against the Cowboys.  But Buffalo isn’t the only team to lose to the Super Bowl champion four years in a row: the Daryle Lamonica/Willie Brown/Gene Upshaw/Jim Otto Raiders pulled off that in the final three years of the AFL and the first year of the post-merger NFL.

In 1967, Oakland made it to the Super Bowl, but lost to the Packers. In 1968, Oakland staged a classic game against the Jets for the AFL title before New York upset the Colts in Super Bowl III. In 1969, the Raiders went 12-1-1 and led the AFL in points, yards, passing yards, passing touchdowns, and yards per attempt. In the playoffs, Oakland beat Houston 56-7, but fell to the Chiefs, 17-7, in the AFL Championship Game. Then in 1970, the Raiders again were on the doorstep of the Super Bowl, but lost 27-17 to the Colts in the AFC Championship Game. [continue reading…]

The Tennessee Titans were weird last year. On paper, the toughest game of the season would have been a week 15 contest in Arrowhead Stadium. And, on paper, the easiest game of the year would be a week 6 home game against the Browns. And yet against both Kansas City and Cleveland, the Titans won by 2 points.

The Titans won one game by more than 14 points last year: would you have guessed it was a 22-point win over the NFC North Champion Packers? Perhaps even more surprising: Tennessee lost one game by double digits in 2016, a 38-17 thumping put on them by… the Jaguars?

It should go without saying that teams tend to play better against bad teams and worse against good teams. But the Titans were a pretty big outlier last year. The table below shows each of the Titans games last year. The table is sorted by the “SOS+HFA” column, which shows the home field adjusted team rating of each opponent. The Chiefs had an SRS rating of +5.6, so playing at Kansas City goes down as a +8.6. The Colts were at +0.2, so playing in Indianapolis is a +3.2, while hosting the Colts is at -2.8. [continue reading…]

Yesterday, I looked at the least-conforming games of the season. I used the SRS to derive opponent-adjusted team ratings, and then came up with a predicted point spread (based on those team ratings and the location of the game) for each game in 2016. By definition, the amount by which each team will exceed its expected points in “overachieving games” will equal the amount by which it fell short of its expected points in “underachieving games.” Since we are just manipulating the 16-game sample, a point by which a team overachieves in one game has to come from another game.

But what we can do is take the absolute value of the difference between the expected margin of victory and actual points differential to get a sense of how consistent or inconsistent each team was last year. And by that measure, the most consistent team was the New York Giants. In 13 of 16 games last year, the final margin came within a touchdown of expectation, and in 3 of 16 games the final margin came within one point of expectation.

The table below shows how to calculate these ratings. Let’s use week 1, which happens to have been one of the rare Giants games that went off script. [continue reading…]

MetLife Stadium opened in 2010. Over that time period, the Jets are 30-26 at home (a hair below-average), while the Giants are 32-24 (right at league average). And, over that same time period, the Jets are 22-34 on the road (slightly below-average), while the Giants are 26-30 in road games (slightly above-average). ^[1]Note that this includes the two games where the Jets and Giants played at MetLife Stadium; the “road” team won in both games.

So the Jets are +8 in games at MetLife, while the Giants are +6. In home games, the Giants have outscored opponents by 142 points (16th-best), while the Jets have only outscored opponents by 51 points (23rd-best). In road games, the Giants are at -121 (also 16th-best), while the Jets are at -344 points (28th-best).

The Giants had a great year at home in 2016 while the Jets did not; the numbers were much different a year ago, when the Jets appeared to be gaining a much better home field advantage at MetLife than the Giants. That was driven, in large part, by performance in one score games. From 2010 to 2015, the Giants were 8-13 in home games decided by 7 or fewer points, while the Jets are 13-8; that made the Giants one of the worst teams at home in close games, and the Jets were one of the best. But last year, the Giants went 6-1 in home games decided by a touchdown or less, and the Jets went 0-3.

So the conclusion in today’s post? The Jets and Giants are getting about the same home field advantage from MetLife Stadium. That in and of itself isn’t necessarily an important conclusion ^[2]Although past research showed the Giants may have been better at Giants Stadium. But there are two worthwhile takeaways from this post. [continue reading…]

Even the most hardcore of NFL fans probably haven’t heard of former Giants running back Eddie Price. His career totals — 3,292 rushing yards, 24 touchdowns, 672 receiving yards — are unimpressive; his football career was anything but.

In 1942, Price was considered perhaps the best high school football prospect in the country, but World War II prevented him from being part of the next great super team at Notre Dame. Instead of going to play for Frank Leahy and the Fighting Irish, Price reported to New Orleans for his service. After the war, he stayed in New Orleans and attended Tulane, and helped power one of the best teams in school history. Price played at Tulane in the late ’40s and was a superstar:

He became the first Green Wave player to rush for more than 200 yards in a game, the first to top 1,000 yards in a season and the first in NCAA history to surpass 3,000 yards for his career.

He was named an All-American in 1949 and was twice named All-SEC. He also set the SEC rushing record with 1,178 in 1949. Price nearly beat his record in 1950 with 1,137 yards and his 1949 mark stood unbroken for 27 years.

Price’s career at Tulane was legendary. In his first collegiate game, he had a 103-yard kickoff return against Alabama for a touchdown that led to a 21-20 upset. He later helped Tulane beat the Crimson Tide in ’48 and ’49, too. In fact, Price was so dominant in his three years with the Green Wave that he retired as the NCAA’s all-time leading rusher with 3,095 yards.  He helped Tulane beat LSU by the score of 46-0 in 1948 by rushing for 116 yards and two scores; the Louisiana schools used to face off every year, but the Green Wave didn’t beat the Tigers again until 1973.  As a senior, Price led the NCAA in yards per carry.

In 1950, Price was the 20th player selected, going in the second round to the Giants. And that’s when he really took off.  As a rookie, he led the NFL in rushing yards per game at 70.3. Price missed two games, but otherwise had a magnificent rookie season: he finished the season with 145 yards, 156 yards, and 103 yards rushing in three of the Giants final four games. [continue reading…]

Five years ago, in one of the first posts at Football Perspective, I looked at league-wide passing distribution in terms of what percentage of receiving yards were gained by the WR1, WR2, WR3, TE1, and RB1 for each team. Today I want to examine passing distribution in a different way: how much are teams spreading it around than ever before?

In the comments to Wednesday’s post, Quinton White described one way economists measure how concentrated industries are, using a relevant football example:

If you wanted to incorporate more than just the #1 guy, then you could sum up the squared shares for all a QBs receivers. For example, say a QB threw to 7 guys, and the first guy caught 30% of the yards and the second 20% and the remaining 5 guys each caught 10%, then he would have a concentration index of .3^2 + .2^2 + .1^2 + .1^2 + .1^2 + .1^2 + .1^2 = .18. The higher the number, the more concentrated the passer is. The max is 1 (Brees threw all his passes to Cooks then 1^2 = 1). If he threw 10% to ten guys each, then the index would be .1.

Let’s say we did that for the 2016 Falcons, who had the best passing game in the NFL last season. Atlanta’s skill position players gained 4,960 receiving yards last year. In the table below, column 2 shows the number of receiving yards gained by each player, column 3 displays their number of receiving yards divided by 4,960, and column 4 shows the squared result of what is in column 3. The bottom right cell in the table is the sum of all the numbers in column 4, or 14.14%. [continue reading…]

On June 15, 2012, I launched Football Perspective. Since that day, Football Perspective has posted a new article every single day. Remarkably, this is the 1,998th post published at this site.  You can fact check that claim here, and at the top of every page is a link to the Historical Archive, a page that is updated after each post is published.

There’s no way this site could still be up and running — much less producing content daily — without this community.  Getting to know you, getting help from you, and just learning and enjoying football with you is an awesome experience. Your contributions to Football Perspective is what makes this a website and not a diary. A special thanks to all the guest writers, who help keep this site fresh and interesting.

Every day, I consider myself lucky to be able to participate in a community where people willingly take time out of their busy lives to check this little site.  But today, I consider myself just that much luckier.  Thank you to the many people who have helped me get this site to where it is today. I hope you forgive me if the site’s 1,998th post is a little shorter than most, but hey: we have a birthday to celebrate.

In 2016, Odell Beckham gained 34% of all Giants receiving yards, the highest share in the NFL. For 31 of 32 teams, at least one player gained 20% of their team’s receiving yards, but for the Bills, Robert Woods led the team in receiving despite being responsible for only 19% of Buffalo’s receiving yards.

But since Drew Brees came to the Saints in 2006, no team has spread it around more than New Orleans. On average, Brees’ leading receiving has been responsible for only 22% of the Saints receiving yards each year. The table below shows the average percentage of team receiving yards gained by the top receiver (RB, WR, or TE) for each team in each season over the last 11 years. The Falcons, buoyed by long runs of success by Roddy White and then Julio Jones, have been the most WR1-heavy passing game, while the Saints have been the most diverse: [continue reading…]

Five years ago, I asked two questions: how often does the first receiver selected in the Draft turn out to be the best rookie receiver?  And how often does the best rookie receiver turn out to be the best receiver from his draft?  In the 2017 NFL Draft, the Titans selected Corey Davis, the excellent wide receiver from Western Michigan with the fifth overall pick.

At the time of my original post, the protagonist was Justin Blackmon, the highest selected receiver in the 2012 Draft.  And at the time, the odds looked ugly: from 1970 to 2010, only 4 out of 31 times did the first receiver drafted lead his rookie class in receiving yards: Ahmad Rashad in 1972, Isaac Curtis in 1973, Jerry Butler in ’79, and then Willie Gault in 1983.  When A.J. Green did it in 2011, it ended a streak of 27 straight years where the top receiver didn’t lead the league in receiving yards.

So what’s happened since then? Well, Blackmon did in fact lead all rookies in receiving yards, although the margin over T.Y. Hilton was just four yards. In 2013, Tavon Austin was the first wideout drafted, but he ranked 9th among that group in receiving yards as a rookie with 418. Instead, Keenan Allen (1,046) took top honors that year.

In 2014, Sammy Watkins was the first wideout selected in perhaps the best wide receiver class ever.  Watkins had a very good year with 982 yards (ranking 4th among wide receivers drafted that season), but that was a far cry behind Odell Beckham and his 1305 yards (in just 12 games).  But then two years ago, Amari Cooper joined Green and Blackmon by being the top rookie wide receiver in both the draft and the regular season. Cooper was the 4th overall pick and had 1,070 yards, beating undrafted Willie Snead (984).  Finally, last season, Corey Coleman was the first wide receiver drafted, but he had only 413 yards in 10 games.  In 2016, there was just one great rookie wideout: Michael Thomas had 1,137 yards, and no other rookie receiver had even 700 yards. [continue reading…]

Yesterday, I wrote Colin Kaepernick was an extreme outlier in 2016 in terms of TD/INT ratio relative to his Net Yards per Attempt average.  Kaepernick ranked tied for 6th in TD/INT ratio, but was 2nd-to-last in NY/A.  At a high level, we have a good clue that the sparkling TD/INT ratio wasn’t as valuable as it seems: that’s because Kaepernick went 1-10 as a starter last year, and the one win came in a game where Kaepernick threw an interception! Now we all know that win-loss record isn’t a good way to judge quarterbacks, especially considering that Kaepernick played for a team that ranked last in points allowed and yards allowed. But the 49ers ranked 27th in points scored and 31st in yards gained, so it’s not as though the defense deserves all of the blame. Because while Kaepernick had a great TD/INT ratio, that disguises how ineffective the passing attack really was. [continue reading…]

This website has been pretty light on coverage of Colin Kaepernick, despite his name turning into a traffic boom for the rest of the football world. The last time Kaepernick’s name appeared in a headline was over a year ago, when I wrote about him declining for three straight years (that ended last season). In fact, Kaepernick’s name has appeared in the text of just three articles at FP in 2017, where his name was used in passing in each case.

And I’m not interested in getting into the usual Kaepernick debate. But there is something that Football Perspective is well-equipped to address: the citing of Kaepernick’s 16/4 TD/INT ratio as evidence of his productive play. Regular readers know that I’m not a fan of TD/INT ratio, and Kaepernick is a pretty good case study in why TD/INT ratio is a poor way to judge a quarterback. A 4.00 TD/INT ratio is very good, no doubt: but in the abstract, it doesn’t mean much. And what do I mean by the abstract?

For starters, it only tells us what happened on 5% of all dropbacks Kaepernick had last year. The much more predictive measure of passing performance is Net Yards per Attempt, and there, Kaepernick ranked 29th out of 30 qualifying passers. ^[1]Neither the Bears nor Browns had a single passer finish with 224 attempts. And, for what it’s worth, he has the worst NY/A average over the past two seasons among the 35 passers with at least 400 attempts since 2015.

So we have a pretty significant disconnect, with Kaepernick ranking 2nd from the bottom, ahead of only Brock Osweiler, in passing efficiency, but tied for 6th with Sam Bradford but in TD/INT ratio. The best thing to do, of course, is to combine the two metrics as we do in ANY/A. There, Kaepernick ranks 23rd out of the 30 qualifying passers. That’s bad, but not horrible, for a starting quarterback. [continue reading…]

Yesterday, I looked Josh McCown’s weird, winding career. McCown started 33 games through age 33, but has since started 27 more games. He’s had one of the weirdest and back-loaded careers in NFL history.

Which made me wonder: how can we measure which quarterbacks had the most front-loaded or back-loaded careers? Here’s one clean way to do it. For every quarterback, identify his exact age for every start of his career, and then calculate the average age in all games he started. For McCown, with 60 starts, his average age (summing his age in every start, and dividing by 60) so far is 30.7. If Josh McCown somehow starts 16 games for the Jets this year, the first will come when he is 38.2 years old, and the last will come when he is 38.5 years old. That will bring his average age of start up to 32.3. That would be pretty old, but not remarkably old the way his median age would be (more on this in a minute).

By this method, the quarterback with the oldest average age is Doug Flutie. The CFL superstar and Bills fan favorite started 66 games in his NFL career, but on average, he was 35.5 years old during his average start. If you read yesterday’s post, you won’t be surprised to learn that after Flutie, Warren Moon, Roger Staubach, and George Blanda have the next oldest average age. Babe Parilli, who is 5th in median age, is down at #10, thanks to 9 starts coming at the age of 22 or 23.

The table below shows the average age and median age of start for all 179 quarterbacks with at least 50 starts. Some fine print: this only covers starts beginning in 1950, so this list may overstate the average age for quarterbacks who played pre-1950; similarly, for current quarterbacks like Luck, this obviously is biased in the other direction. The table below is fully sortable and searchable; by default, it lists the 15 oldest players based on average age of start. [continue reading…]

The Cleveland Cavaliers and Golden State Warriors are facing off in the NBA Finals for the third straight season. That’s never happened before in NBA history, and it only happened once in pro football history… and it also involved Cleveland.

In 1952, the Browns won the American with an 8-4 record, while the Detroit Lions won the National division with a 9-3 record (after defeating the defending-champion Rams in the National tiebreaker game). Otto Graham and Bobby Layne were the two top quarterbacks in the NFL that year according to both the AP and the NY Daily News. Detroit traveled to Cleveland on December 28th and defeated the Browns 17-7, with Doak Walker’s 67-yard touchdown providing the biggest blow.

The next season, Graham had a season for the ages by any measure.  You’d be hard-pressed to argue for a better regular season by any quarterback from World War II to 1983, when a Graham-led Browns passing game finished with a Relative ANY/A of +5.00.  The Browns began the 12-game season with 11 straight wins, while Detroit finished 10-2 with both losses coming against the 8-3-1 Rams.  Cleveland lost the season finale in Philadelphia, and then traveled to Detroit for an NFL Championship rematch.

The Browns and Lions were tied 10-10 after three quarters, and Cleveland was up 16-10 late in the game.  But in the final minutes, Layne found an unlikely hero in Jim Doran for a 33-yard game-winning touchdown (video here), with Walker’s extra point providing the margin of victory. The bigger story? Graham having one of the chokiest games in football history, finishing with 2 of 15 for 20 yards with 2 interceptions. [continue reading…]

In 2007, the Jets drafted David Harris in the second round, and the linebacker has turned into one of the best players in the league who has never made a Pro Bowl. Since then? That round has failed to yield a single productive player. Here are New York’s second round picks in each of the last ten years.  Note that this is not a pretty table, and that’s before realizing that the Jets traded up from 47 to 43 in the ’12 Draft with Seattle to draft Hill, while the Seahawks settled for Bobby Wagner.

[continue reading…]

The graph below shows the percentage of NFL (or AFL or AAFC) games that have ended in a tie since 1940. Note that the Y-Axis goes from only 0% to 20%: that may be misleading, but showing the graph from 0% to 100% would make for a much less useful visual in my opinion.

I’m short on time today, but most observers seem to think that shortening overtime from 15 to 10 minutes is likely to result in more ties. It certainly seems unlikely to result in fewer ties, although it’s possible that teams will just engage in tie-averse behavior earlier in overtime now.

We have spent a long time debating the best way to handle overtime. In general, I’m not too opposed to more ties in the regular season. While unsatisfying at the time, they arguably serve as a better tiebreaker than traditional tiebreakers. When two teams are 9-7, the tiebreaker to determine which one advances to the playoffs is not necessarily better than 50/50 at deciding which is the “better” or more “deserving” team. But if one of those 9-7 teams wound up being 9-6-1 rather than losing in the 74th minute of the game, that would be the team that advances. That seems more likely, in the long run, to identify the “better” or more “deserving” team. I think. Of course, that does not to change the unsatisfying result in the short term.

What do you think?

This week, James “Four Touchdowns” Hanson had a couple of interesting posts on the support four star quarterbacks received.  James provided some very extensive, in-depth analysis, but that doesn’t mean there isn’t still a place for simple, surface-level analysis, either!

I was wondering which quarterbacks received the most and least support from their team’s rushing attacks. Which brings us to Jeff George. There are 179 quarterbacks who have started at least 50 games in the NFL. George started games across five different teams — Indianapolis, Atlanta, Oakland, Minnesota, and Washington.  And in the 124 games he started, his teams averaged just 87.9 rushing yards per game, the fewest of any quarterback in NFL history.

In Oakland (23 of 124 starts) in ’97 and ’98, George had an in-his-prime Napoleon Kaufman, so that wasn’t a bad situation: his Raiders averaged 105.9 rushing yards per game. And in 10 starts with the Vikings in 1999, George had the impressive combination of Robert Smith and Leroy Hoard, and the Vikings averaged 126.4 rushing yards per game.

But his fortunes were much different in his other stops. George began his career, of course, with the Colts from 1990 to 1993. In his 49 starts in Indianapolis, the Colts were absolutely terrible on the ground. There were 112 team seasons from 1990 to 1993 — that’s four years during the 28-team NFL — and the four Colts teams ranked 106th, 108th, 111th, and 112th in rushing yards over that period. Over those four years, Indianapolis rushed for just 4,841 yards, more than 1,000 rushing yards behind the second-worst team (Miami). The Colts averaged an anemic 3.38 yards per carry, also worst in the league. In George’s 49 starts, Indianapolis averaged just 75.6 rushing yards per game.

George then went to Atlanta, and from ’94 to ’96 (35 starts), the Falcons rushed for just 82.3 yards per game in George’s starts. George started 16 games in both ’94 and ’95, and Atlanta averaged the fewest rushing yards per game of any team in the NFL during that period.  As you probably know, those Falcons famously ran the Run-N-Shoot under head coach June Jones, so some of this was a reflection of philosophy rather than lack of talent ^[1]It’s also worth noting, even this should always be implied, that rushing yards is highly correlated to team success, and George’s Colts were terrible, going 14-35. In 1994, the Falcons ranked dead last in both rushing attempts and rushing yards, and 3rd in pass attempts and 5th in passing yards.

The personnel was suited for Jones’ offense: Terance Mathis, Andre Rison, ^[2]Ironically, Rison was traded from the Colts to the Falcons Indianapolis traded up to acquire George when he was the projected first overall selection. Bert Emanuel, and Ricky Sanders were all starters in the Falcons 0-TE/0-FB offense, with Ironhead Craig Heyward and Erric Pegram at running back.  That offense worked pretty well (and would likely work even better today), but a high number of interceptions and a bad pass defense caused the team go to 7-9 in 1994.  In ’95, Rison (who signed an enormous contract to play in Cleveland) and Sanders (just two catches in his final season) were gone, but Eric Metcalf was acquired in Rison’s place and J.J. Birden (from Kansas City) filled Sanders’ role.  Heyward actually made the Pro Bowl and rushed for 1,000 yards, but the Falcons remained a pass-heavy team.  George was benched three games into the 1996 season, ending his time in Atlanta. ^[3]And to complete the story: in the final 7 starts of his career, in Washington, George’s team rushed for only 86.3 yards per game.

The table below shows the rushing yards per game averaged in games started by each quarterback, for the 179 quarterbacks with at least 50 starts. The table is sorted by rushing yards per game, from most to fewest, so George is at the very bottom (the table is fully sortable and searchable). [continue reading…]

You may have heard that Sam Bradford set the completion percentage record in 2016 by completing 71.6% of his passes.

What you may not have heard: Bradford also ranked last in the league in passing yards gained per completion, which makes his record-breaking performance a somewhat hollow achievement. Bradford is the fifth quarterback in the Super Bowl era to rank 1st in completion percentage and last in yards per completion, joining David Carr (HOU 2006), Eric Hipple (DET 1986), Joe Montana (SFO 1980) in his first year as a starter, and Sonny Jurgensen (WAS 1969). In general, things didn’t work out well for those quarterbacks:  Carr posted a 6-10 record in 2006, while Hipple went 3-7, and Montana went 2-5.  Bradford went 7-8 last season, meaning only Jurgensen (7-5-2) posted a winning record of that bunch (and Washington had a negative points differential and faced a very easy schedule that year).

Expand the list to finishing 1st or 2nd in completion percentage and last or 2nd-to-last in yards per completion, and you bring in four more quarterbacks: Chad Pennington (NYJ 2007, 2nd in both), Joe Montana (SFO 1981, 1st in comp%, 2nd-to-last in YPC), Fran Tarkenton (MIN 1977, 1st, 2nd) and Len Dawson (KAN 1972, 2nd in comp%, last in YPC).  The results there were mixed: Pennington went 1-7, while Montana went 13-3, Tarkenton went 6-3, and Dawson went 7-5.  It is worth pointing out that Montana and Tarkenton both had above-average Y/A ratios that year: in other words, having a high completion percentage is great, but only if it doesn’t come at the expense of your yards per completion average.

How much of a checkdown artist was Bradford last year? He finished 1.95 standard deviations above average in completion percentage last year among qualifying passers, a metric commonly referred to as a Z-score. He also finished 1.82 standard deviations below average in yards per completion. If you take his Z-Score in completion percentage (+1.95), and subtract his Z-Score in yards per completion (-1.82), you get a result of +3.77.

That may not mean much in the abstract, but it ranks as the 3rd most extreme result in the Super Bowl era, behind only Jurgensen 69 and Carr 06. The table below shows the top 200 most extreme checkdown artists — by this metric — since 1966:

As always, please leave your thoughts in the comments. Tomorrow, we’ll look at the opposite result: any guesses as to the leaders in that category?

After the Jaguars drafted Leonard Fournette with the 4th pick in the 2017 Draft, NFLResearch tweeted the following:

That was, at least for me, a surprise. And it is true: there have been nine running backs drafted in the top 5 since 2000, and those teams have improved by 43 wins. There is some natural regression to the mean built in to any analysis like this, along with two big outliers: the 2016 Cowboys and 2006 Saints used a top five pick on a running back, but also added Dak Prescott and Drew Brees, producing two of the greatest improvements in passing efficiency in NFL history. Those two teams produced 16 of those 43 wins; without those two teams, the average increase drops to a still-impressive 3.9 wins. [continue reading…]

Since Joe Thomas entered the league, the Browns have a record of 48-112. That translates to a 0.300 winning percentage, the second-worst in the NFL over the last decade.

That’s bad, but not as historically bad as I would have thought. From 2000 to 2006, the Detroit Lions had a 0.295 winning percentage, the worst in the NFL. And from 2007 to 2011, the Rams had an anemic 0.188 winning percentage, the worst in the NFL. There was a common denominator for both of those teams: defensive end James Hall.

How did I find Hall? I looked at all players with 100 career games played. Then I calculated the winning percentage for his team in each season of his career, weighted by the number of games he played that season. So when calculating the adjusted career winning percentage for Hall, who played in 165 games and in 16 games for the 2008 Rams, 9.7% of his Adj Car Win% is based off of that team’s 2-14 record. The 2000 Lions went 9-7, but Hall only played in 5 games that year, so the 9-7 mark only counts for 3% of his career record. And the fact that the Lions went 3-2 in the 5 games Hall actually played is irrelevant: for calculating Adj Car Win%, I just used the team’s overall winning percentage multiplied by the number of games he played.

There are over 4,000 players who have played in 100 career games, and Hall has the lowest adjusted career winning percentage. The second-lowest? That honor belongs to John Greco, who was a teammate of Hall’s in St. Louis from 2008 to 2010, and has been a teammate of Thomas in Cleveland ever since.

Below are the 100 players with the worst adjusted career winning percentages: Thomas checks in at #35. [continue reading…]

On Tuesday, I looked at the passing offenses, as measured by ANY/A, of each of the Super Bowl champions. Today, let’s do the same for passing defense. Just over half (26) have ranked in the top 4 in ANY/A, although as we saw with passing offenses, there isn’t a trend towards pass defense mattering more than it used to. If there’s a conclusion to be drawn, it may just be that worse teams are winning it all now than ever before. [continue reading…]

In 1981, the Chargers and Bengals met in the AFC Championship Game. That game isn’t the most memorable game played that day, in part because of the cold weather: nicknamed the Freezer Bowl, San Diego struggled on the road in a game played with a windchill of -32 degrees.

But the bad weather obscured the fact that Dan Fouts and Ken Anderson — by far the top two passers in the NFL that season — were facing off for the right to play in the Super Bowl. Over in the NFC, the 5th and 6th leading passers in ANY/A — Danny White and Joe Montana — were meeting in what would turn out to be one of the most memorable games in NFL history. [continue reading…]

Just 9 of 32 teams had a different player lead the team in pass attempts in 2015 and 2016. Take a look:

As you would suspect, some of the changes were intentional, and some were not. Denver saw their 2015 QB retire, while Minnesota lost their starting quarterback in the preseason…. which led Philadelphia to trade their 2015 starter and anticipated 2016 starter to the Vikings.

Three others were injury related. The Cowboys switched quarterbacks intentionally, of course, but still had their expected 2016 quarterback change due to injury. And Chicago lost Cutler early in the year, and he was limited to five starts.

The other four? The 49ers started 2016 with Gabbert, but benched him for Kaepernick. And the Texans, Rams, and Browns intentionally moved on from their quarterbacks, although McCown stayed with Cleveland and did wind up starting three games. [continue reading…]

Using data from Pro-Football-Reference for the 2016 NFL season, I calculated the average number of snaps per play taken by players at each position across the league. Here are the numbers for offense:

QB: 1.00
RB: 1.10
WR: 2.60
TE: 1.25
OT: 2.05
OG: 2.00
OC: 1.00

Some of these are pretty obvious: you have exactly one quarterback, one center, one left guard, and one right guard on each play. For the most part, you have two tackles, although occasionally teams will have three tackles on the field (apparently, about once every twenty plays).

The more interesting numbers come in the split among the skill position players: running backs (including fullbacks) only get about 1.10 snaps per play; in the ’70s, that number would be much closer to 2.0, although there’s no way of getting more precise than that. Tight ends are at 1.25, while the average play in 2016 featured more than 2.5 wide receivers on the field. Three wide receivers has become the base formation in the NFL. [continue reading…]

In 1973, the 14 AFC teams housed 8 Hall of Fame quarterbacks. The AFC East had Joe Namath and Bob Griese with the Jets and Dolphins, the AFC Central had Pittsburgh’s Terry Bradshaw, and the AFC West had five HOF QBs: Len Dawson was with the Chiefs, while the Chargers had a first-year Dan Fouts and a last-year Johnny Unitas. The Raiders? They had Ken Stabler and George Blanda. And in the NFC, Sonny Jurgensen and Roger Staubach were the signal callers for Washington and Dallas, while Fran Tarkenton was the Vikings quarterback. That means the ’73 NFL (along with the ’70 and ’71 versions, which didn’t have Fouts but did have Bart Starr) housed 11 future Hall of Fame passers. And that excludes Ken Anderson, of course, who entered the league in ’71.

Meanwhile, in ’81 and ’82 — at a time, I’ll note, when Ken Anderson was doing pretty darn well — there were just four active HOF QBs. Stabler, who finally made it as a seniors’ nominee last year, Fouts, Bradshaw, and Joe Montana. On average, there have been about 7-8 active HOF quarterbacks at any one time. [continue reading…]

There were really only three notable games in this year’s playoffs. The Super Bowl, of course, was a classic game, if not necessarily a good one to watch from start to finish. The Patriots completed a historic comeback and won in overtime, 34-28.

And there were two upsets: the Packers went into Dallas and won, 34-31, in what was the best game of the playoffs. And the Steelers went into Kansas City and won in a sloppy game, 18-16, where Pittsburgh kicked six field goals.

The other 8 games? All were won by the favorites, and all were won by at least 13 points. That matched the number of times the favorite won by over 10 points in the three previous years combined.

Since 1990, the favorites have won 7.6 of 11 games, on average, in the postseason. With 9 wins by favorites in 2016, that matches the most times the favorite has won in the playoffs, but it happened six other times, too. So 2016 wasn’t all that notable in that regard.

And since 1990, teams have won by over 10 points in just over half of all playoff games. With 8 such wins, that is the most ever, but it happened four other times, too (although not since 2002). But what makes the 2016 playoffs stand out is the combination of the two factors: 8 times the favorite won and won by over 10 points, compared to just 4.4 times on average. The only other time that happened was in 1996. ^[1]And 8 of the 10 times, the home team won, which is high, but also not particularly unusual (the home team won 6.8 games on average).

The table below shows the average results (from the perspective of the winning team) in every playoff year since 1990: [continue reading…]

Yesterday, I looked at quarterbacks from 2016 who started at least 8 games and threw at least 150 passes. For those passers, I calculated how many standard deviations above average they were in Relative ANY/A (i.e., how much better they were, statistically, than average) and in winning percentage. I sorted the list by the difference between the two, to find the quarterbacks whose stats and winning percentages diverged by the largest amounts.

What about historically? I performed the same study going back to 1970. And the season that stands out the most is Archie Manning’s 1980 season. That year the Saints were the worst team in the league: New Orleans went 1-15, and every other team won at least 4 games. ^[1]The Saints’ troubles continued into the draft; New Orleans selected George Rogers first overall, when two of the top four, and three of the top eight players went on to be Hall of Famers. Manning started every game for the team because he actually had a strong season, at least statistically: he ranked 9th out of 30 qualifying passers in ANY/A, and had a Relative ANY/A of +0.53. That, of course, is pretty unusual given his team’s 1-15 record.

That stands out as the biggest example of a divergence of stats being more impressive than team record. The best 100 seasons (although by default, the table only lists the top 20) are below: [continue reading…]

Since 1970, you would do a pretty good job estimating any team’s record by regressing that team’s record back to the mean by about 60%. More specifically, if you take 40% of the team’s actual winning percentage the prior year, and 60% of the league average winning percentage, you would get a pretty good estimate of their record the next season (though the R^2 is just 0.17).

That’s over a long period, though, and team variability is on the rise. If you look at the last 20 years, it’s more like 70%, with the best fit formula to project winning percentage being something like 35% plus 30% of the team’s winning percentage the prior year.

Last year was a bit of a weird year, with some notable outliers. The 3-13 Browns would have been expected to progress to the mean, but instead went 1-15. The 49ers went from 5-11 to 2-14. The Bears went from 6-10 to 3-13. And on the positive side of .500, the Panthers dropped from 15-1 to 6-10, while the Jets win total dropped in half, from 10 to five.

The Patriots defied regression to the mean for the umpteenth straight year, improving from 12-4 to 14-2. The Raiders zoomed past the mean, going from 7-9 to 12-4. The Giants similarly went from 6-10 to 11-5.

The correlation coefficient between team winning percentage in 2015 and team winning percentage in 2016 was 0.27, which is pretty low, but not abnormally so. Here are the correlation coefficients for each pair of years since 1970: [continue reading…]

It’s been an unusually busy offseason for the Patriots. New England signed Buffalo Pro Bowl cornerback Stephon Gilmore to a big contract, and also added former Bengals running back Rex Burkhead.  The Patriots were also active in the trade market, acquiring WR Brandin Cooks from New Orleans, DE Kony Ealy from Carolina, and TE Dwayne Allen from the Colts.

Departing from New England? TE Martellus Bennett went to Green Bay, Jabaal Sheard went to Indianapolis, CB Logan Ryan went to Tennessee, while CB Malcolm Butler and RB LeGarrette Blount, among others, could still be on the move.

Which made me wonder: do the Patriots, as you might suspect, do better adding players from other teams than other teams do when adding players from the Patriots? The table below shows the most productive players (by AV) in year 1 in New England after playing for a different team the prior season. Note that this excludes Dion Lewis in 2015, who was not on an NFL team in 2014. [continue reading…]

The New England Patriots won Super Bowl LI, but it wasn’t because the team was packed full of high draft picks. Of the five Patriots who had more than 10 points of AV, only one was drafted in the first four rounds. Regular readers know that I created an AV-based draft value chart, which assigns points to each draft pick based on the expected marginal production produced by that pick.

Well, you can calculate a team’s weighted average draft value by doing the following:

• Calculate the draft value spent on each player on the roster who produced at least 1 point of AV that season.
• Calculate the percentage of team AV produced by each player. This is key, otherwise Chris Long would skew the results in the wrong direction.
• Multiply the results in steps 1 and 2, and then sum those values.

Here’s how it would work with the 2016 Patriots, who had an average draft value (as a roster, and weighted by AV) of 6.72.

[continue reading…]