Wednesday, October 26, 2016

The NFL Doesn't Suck



We are through the first seven weeks of the NFL season, and it is reasonable to start looking at some of the trends emerging. An emergence of young talent has pushed the NFC East to the top of the league, while the AFC South continues to occupy the basin. Matt Ryan and Matthew Stafford are playing as well as any quarterbacks in the league, after many had given up on them developing into true stars. And every good team seems incredibly lopsided, with barren defensive teams like Pittsburgh, Oakland, and Atlanta battling with inept offenses like Minnesota, Seattle, and Philadelphia.

But the one trend that has gotten the most attention has nothing to do with any specific team. Seven weeks into the season, the most common refrain is that play across the league is at an incredibly low level. The NFL is putting out a thoroughly mediocre product on a weekly basis, and it is dragging down the ratings and possibly doing serious harm to the future of the sport.

There are a lot of opinions on why quality is down. The most recent CBA limited practice time, leading to sloppy play. The league has cracked down on any sort of celebration, leading to too many penalties. The rules are difficult to understand, and the pace of the game is unbearable, burdened by countless stoppages to study in slow motion ambiguous moments of the game.

There is a lot of debate about what the cause of the problem is, but few people seem to be arguing that there is a problem. Every major game comes with new complaints of miserable play, reaching its apex with the 6-6 tie this past Sunday between Arizona and Seattle. A marquee matchup of two of the league’s top teams, this game saw no touchdowns and two missed chip shot field goals in the final minutes of overtime.

The conclusion of this game led to even more shouting about the declining quality. But the truth is, most of it is off base. The missed field goals were bizarre and unexplainable, but for the first 70 minutes of play, this was actually one hell of a football game.

There are plenty of examples of bad football games out there. I’ve been to bad football games myself. And at times this year, I’ve even agreed that a game has been unpleasant to watch. The second Monday night game of opening week that saw the 49ers shut out the Rams was a miserable game. The prime time matchup between Carolina and Tampa Bay was sloppy and unpleasant. There is bad football out there, but this past Sunday’s game was not an example of this.

Low scoring games can be awful when the offense is failing due to its own ineptitude. And for half this game, that was largely the case. Arizona’s defense played fantastic, but Seattle’s offense is a mess right now, and it was difficult to watch that side of the ball. On the other side however, we witnessed some incredible football from start to finish.

Arizona’s offense actually moved the ball throughout the game, and it was only incredible work by Seattle’s defense that held them to just a field goal. They made a crucial fourth down stop deep in the red zone. Bobby Wagner made a sensational play to hurdle the long snapper to block a field goal. This was truly the case of an efficient offense facing a supreme defense, and it was a hell of a lot of fun to watch.

Of course, this is only one game, and while we can disagree about how enjoyable it was, this doesn’t prove anything about the overall quality of the games. To do this, we need to take a broader view, looking at some of the common trademarks of sloppy, incompetent football.

Below I’ve put together a table of some of these statistics. In the first column I list the averages of each category from 2006 through 2015. Beside these I’ve placed the numbers so far from 2016 to give us some idea of how this year has looked in comparison to the recent past.


Average
2016
Penalties/Game
12.3
14.0
First Downs/Game
38.5
40.6
3rd Down %
39%
40%
Turnovers/Game
3.2
2.8
Punts/Game
9.5
9.0
FG%
83%
84%
Yards/Game
678
711
Points/Game
44.3
45.5
ANY/A
5.8
6.4
Completion %
61%
64%

A few things are immediately obvious from this table. Offense is clearly at a much higher level than it’s been over the past ten years. Quarterbacks are playing better, with higher completion percentage and ANY/A. Offenses are gaining more yards, more first downs, and scoring more points. They are punting less frequently, converting more third downs, and turning the ball over fewer times each game.

Does any of that sound like a lower level of play?

Admittedly, offense has been trending up for the past decade, so it isn’t surprising that the averages would be lower than the current numbers. So it might be better to chart these numbers to see if we can spot a significant deviation from this trend.

(For the ease of charting, I’ve adjusted every stat so that 100 corresponds with the average in the table above. I also inverted the turnovers and punts series so an increase would show up as a decline in the graph.)


The positive trend in offensive production is very clear from this chart. What’s also clear is that 2016 hasn’t reversed that trend. Quarterback efficiency is higher than it has ever been. We’re seeing the fewest punts since 2008, and the fewest turnovers of any year. The only example of clear decline from a year ago is field goal percentage, and even that is still higher than at almost any other point in history.

The one example where it appears things have gotten worse is penalties. There is more than a full penalty extra per game over the average, and it certainly has seemed like the flags have been coming out a little too aggressively this year. But when we plot the series of penalties per game over time, we get a slightly different picture.


Penalties are up from a year ago, but when compared to the trend this doesn’t look that bad. There are fewer penalties now than there were in 2005, and only 0.15 more penalties than a year ago. This difference is trivial, and anyone who tells you that two extra penalties for every week of play has severely impacted their ability to enjoy football is a liar.

So why is this the overwhelming narrative of 2016? Surely we aren’t all just imagining that the games have been worse. There has to be some explanation for this almost universal belief.

One possibility is that the games that have received the most attention have been the worst. And it certainly has seemed like the prime time games have been bad. Even if you discount last Sunday’s game, there have been some pretty miserable matchups. We’ve seen the 49ers twice, the Bears twice, and the Texans three times. I’m sure if we plotted those numbers for just the prime time games we’d see a notable difference.


Average
2016
Prime Time
Penalties/Game
12.3
14.0
13.4
First Downs/Game
38.5
40.6
39.5
3rd Down %
39%
40%
39%
Turnovers/Game
3.2
2.8
2.7
Punts/Game
9.5
9.0
10.0
FG%
83%
84%
84%
Yards/Game
678
711
661
Points/Game
44.3
45.5
41.1
ANY/A
5.8
6.4
5.7
Completion %
61%
64%
64%

Is there something there? Maybe a little. Quarterback play has definitely been worse, and offense on the whole seems diminished from the rest of 2016. But even so, the numbers aren’t that far out of whack from the averages over the past ten seasons. The prime time games have been lower quality than the rest of the season, but to claim they’ve been the worst football the league has ever seen is just ignorant.

The truth is, the quality of football has been just fine this year. There have been some bad games, and these isolated incidents have driven the narrative. Because once we start talking about bad football, that’s all we are able to see. There are bad plays in every game, just as there are great plays in every game, and unfortunately this year the stories we’ve been telling ourselves have been more about the former than the latter.

We talk about how terrible Case Keenum was in the London game between the Giants and the Rams, forgetting the incredible individual effort by Landon Collins on his pick-six. We talk about the shanked field goals in overtime of the Seahawks-Cardinals game, rather than the spectacular block by Wagner. Games become an exercise in confirmation bias, and with more than a hundred plays in every game there are plenty of opportunities for us to pick them apart and complain.

There are ten weeks left in the NFL season, followed by four weeks of playoffs. There is a lot more football ahead, and it’s up to us to decide how we’ll enjoy it. And in the end I can only speak for myself, but I personally find more pleasure in appreciating the great moments of the game than complaining about the bad.

Saturday, October 22, 2016

The Future of the Quarterback Position



The past decade has been a golden age of quarterback play. Records have been set left and right, and standards have been raised to the point that it’s almost impossible to compare a modern quarterback’s statistics to ones that have come before.

The quarterback has always been the most important position in sports, but that is now even more the case, as it has become almost impossible to even compete for a championship without a star under center. And yet, so far this year it appears that things might be changing in the NFL.

I wrote about this a couple weeks ago, touching on how the five then undefeated teams had succeeded despite merely average quarterback play. Things have swung back the other way since, as teams like the Ravens, Broncos, and Eagles have dropped multiple games, while quarterback strong teams like Seattle and Atlanta have pushed to the forefront of the league. But it’s still difficult not to look at the league right now and wonder if we are seeing the beginning of a new trend.

Anecdotally, it isn’t hard to find explanations for this. Five years ago the league was dominated by an almost unprecedented group of quarterbacks. Tom Brady and Peyton Manning are all time greats. Drew Brees, Aaron Rodgers Ben Roethlisberger, Philip Rivers, and Eli Manning will likely end up in the Hall of Fame as well. And between 2003 and 2011, all seven of these players were in the prime of their careers, combining to win every Super Bowl in this period.

But time remains undefeated, and this era is starting to come to an end. Peyton Manning retired this past offseason. Brees and Rivers are stuck on teams devoid of talent, and Rodgers and Eli Manning have regressed sharply. Brady and Roethlisberger are still going strong, but their time is winding down as well.

Every great career reaches its end, and it isn’t stunning that these greats are getting ready to move on. It wouldn’t be a problem either, if there was another generation in place to step forward and claim the mantle. But while plenty of good quarterbacks have entered the league over the past ten years, they have yet to show the promise of measuring up to the generation that came before.

Matt Ryan is the early favorite for MVP, but he is coming off of a couple of disappointing years. Joe Flacco has been a mess ever since his Super Bowl run. Cam Newton and Andrew Luck are great players in bad situations, and as good as Russell Wilson is, he still has a ways to go before he enters the same conversation as the stars who are on their way out.

This is a story we’ve heard a lot about over the past few seasons, and it’s worth questioning just how much truth there is to it. Are the young quarterbacks of this era really a step back from what we’ve had, or is this just another common trick of nostalgia, trying to convince ourselves that the future can never live up to the past?

To answer this question, I decided to take a look at some stats. My favorite metric for measuring quarterbacks is Adjusted Net Yards per Attempt (ANY/A). It is calculated by taking total passing yards, adding 20 yards for each passing touchdowns, subtracting 45 yards for each interception, and then subtracting sacks yards, the final total of which is divided by attempts. This metric has the advantage over ESPN’s QBR in that it is straightforward and easy to understand, while having the advantage over traditional quarterback rating in that it is based on statistical studies rather than some 40 year old bullshit equation.

Starting in 2000, I pulled the season by season statistics of every quarterback to throw at least 5 passes in a given year. (Thanks to Pro Football Reference for their fantastic play finder tool.) Using this data, I was able to break down the performance by age of quarterbacks over the past fifteen years.

The first thing I looked at was the average age of quarterbacks in the NFL. Now there are a number of ways to do this, but the best is to simply do a weighted average by number of attempts. That is shown by the blue line in the chart below.

But this isn’t just a question of whether or not young quarterbacks are playing. It’s a question of whether or not they are good. Blake Bortles has plenty of pass attempts, but that doesn’t mean he’s actually an NFL caliber passer. So in addition to the attempts weighted age, I weighted by total ANY as well, in order to get a measure of how old the average value producing quarterback was in each given year.

 
Looking at the chart, there aren’t many obvious conclusions to draw. It looks like there’s been a big swing over the past eight years, but on the whole the trends look pretty level. Older quarterbacks typically produce more value, a gap that looks like it might be widening over recent years.

To zoom in on this gap, I plotted the difference between the two lines.


This time a fairly clear trend emerges. The gap has been widening over the past eight years, as older quarterbacks have produced increasingly more value than younger. The gap shrunk pretty drastically last year—likely due to the struggles of Peyton Manning—but it looks to be rebounding in 2016.

There are other ways we can look beyond just the average age. I separated the data sets into four five year age gaps to see how much value each has contributed. Below I charted what percentage of ANY was produced by each age group since 2000.


Quarterbacks older than 35 years follow a pretty typical pattern, bouncing back and forth around 10 percent depending largely on the production of a few individuals. Quarterbacks between 21-25 haven’t shown much variance either, though the past few years have been a dropoff from their 2012 peak.

Where the difference really shows up is in the two middle groups. Between 2008 and 2014, the 31-35 group contributed a higher percentage of ANY than the previous season every year but one. Their contribution dropped off last year, but it’s on its way to bouncing back in 2016, bolstered by the performance of the 31 year old Ryan.

The 26-30 group is the most troubling. Over the first 12 years of this sample they produced less than 40% of the total ANY only once, briefly dipping below the line in 2002. Since 2011 however, they have eclipsed that mark only once out of four full seasons and are on their way to falling short once again in 2016.

Quite often we find ourselves complaining about things that aren’t nearly as bad as we make them out to be. Children these days are running wild. The quality of football games is at the lowest point in history. And all the good quarterbacks are getting old and retiring, leaving us with a league full of Matthew Staffords and Jay Cutlers. But for once, it looks like this might actually be the case. A historic era for quarterbacks is coming to an end, and the heir apparents have yet to step forward.

This might bode poorly for the NFL, if it wasn’t something we’d seen before. Between 1995 and 2001, the NFL bid farewell to Joe Montana, Jim Kelly, John Elway, Steve Young, Dan Marino, and Troy Aikman. Six Hall of Fame quarterbacks, responsible for eighteen Super Bowl appearances and ten victories between them. It was a staggering loss of talent, and there were real concerns with how they would be replaced.

But of course, we know the answer to that. Between 1995 and 2001, the NFL also welcomed Peyton Manning, Brady, and Brees. And in 2004 they added Eli Manning, Rivers, and Roethlisberger, followed by Rodgers in 2005. The talent came back, and it dominated the league much as the generation before had done.

But in the middle, something interesting happened. Between 1988 and 1998, the Super Bowl was won by a future Hall of Fame quarterback nine times. Just as between 2003 and 2011 every single title was won by a passer who will end up enshrined in Canton.

The four years between these stretches saw a very different sort of quarterback win however. Kurt Warner tore the league up in 1999, and he may very well end up in the Hall of Fame someday. But he was still a castoff who lucked into the best situation a quarterback could find himself in, riding one of the greatest assemblies of offensive talent to a title. In 2000 Trent Dilfer did the same with a historically great defense, as did Brad Johnson in 2002. And while Tom Brady is a sure fire first ballot Hall of Famer, in 2001 he was a game managing second year player forced into the lineup due to an injury to the established starter.

Quarterback groups run in cycles in the NFL. At times the position seems impossibly deep, and teams without a star under center have almost no hope of fighting to a title. But every so often a window will open for a team with a solid core and a mediocre passer to claw their way past the thin lineup of superstars. The quarterbacks will be back, and they will dominate the NFL again. But that may not happen for a few more years, and until then the league is wide open.