Obstinacy and Analytics in the NFL

The NFL is a league of 32 separate entities that are all working towards a common goal (profit) while competing against each other. There is massive conduit that connects all 32 teams that allows staggering amounts of information to flow throughout the league. Ideas and concepts can change hands in the league with very little impedance, especially with how often coaches, scouts, and personnel change teams and take that information with them.

Yet, with such a terrific pipeline through which information can flow, football organizations are a stubborn bunch.

Change doesn’t come easily in the NFL, and when it does, no one likes to admit that something is being changed. They have their way of doing things and until a better way is proven five times over, teams just simply refuse to adapt.

Darwinism states that all species of organisms arise and develop through the natural selection of small, inherited variations that increase the individual’s ability to compete, survive, and reproduce. Without this ability, organisms will eventually die out and become extinct. The speed at which an organism adapts can greater enhance its ability to dominate over other organisms in its surrounding environment. NFL coaches and organizations are essentially anti-Darwin because they’re so remarkably resistant to change.

The problem doesn’t lie in the people themselves. It’s the machine of the NFL that has created this environment and front offices and coaching staffs have fallen victim to the demands.

Hiring for football positions in the NFL is an anti-meritocracy. Ninety-five percent of hires are made via the “buddy system.” It’s not what you know; it’s whom you know. Coaches and staff members are raised in this environment. They use the buddy system to make it to becoming a head coach or a general manager by making promises to hire friends or former colleagues who helped them to get to where they are now. Thus, the vicious cycle continues.

This is an example of a much larger issue of organizations’ inability to adapt, grow, and evolve in the NFL. The problem is exacerbated by the influence of the now- generation. Because of the advancement of technology, humans are so used to everything in our lives being on-demand (information, television, food, etc.). Team owners have developed the same thinking and expect change to come much faster than it actually happens.

Win Now or Get Out

There was a clear message that Cleveland Browns owner Jimmy Haslam sent when he fired his entire staff after only a year on the job. Win now or get out. It’s a strong example of the ideology that a lot of owners are taking throughout the league.

It’s tough for owners to watch Jim Harbaugh and Andy Reid turnaround the 49ers and Chiefs so quickly and so spectacularly, only to watch their own team waddle through mediocrity year after year. Owners are typically men or women who have had life answer every beck and call at their whim. It isn’t a stretch to think that they would expect their team to do the same.

The Chiefs after Romeo Crennel and the 49ers after Mike Singletary are two examples of teams that were ripe for a revolution. In both instances, the rosters were talented and experienced but had been poorly coached and schemed, which caused the team to falter rather than flourish. By all accounts both teams had strong locker rooms even through the tough times. The owners hired intelligent coaches who adapted their schemes to the players already on the roster instead of stubbornly demanding that the players adapt to them.

These turnarounds both happened within a few years of each other but it’s important to understand that they are the exception to the rule. It takes time, diligence, and continuity within the organization to turn a bad team into a good one. The Jaguars right now are a good example of this. They were a bad team but are headed in the right direction because they’ve gotten the right people in the right positions and have committed to making the team better without rushing the results. But in “Generation Now” of the NFL, most owners are too impatient to trust their own hires and allow good people to do their jobs.

The late Steve Jobs, one of the greatest modern day CEOs, said, “You’ve got to be a really good talent scout. No matter how smart you are, you need a team of great people… and be able to help build an organization that can eventually build itself because you need great people around you.”

Jobs hired people the best people available for a specific position and then allowed them to manage that position without interfering too much. He made decisions when necessary but he didn’t micromanage the people below him. He trusted his own skills of talent scouting and he built Apple into one of the world’s most recognized consumer brands. The level of continuity within Apple has allowed the company to continue to progress even after Jobs’ passing.

League owners have fallen victim to the “win now” attitude and are making rash decisions because they’re not progressing fast enough. If the team hasn’t made a turnaround in two to three seasons, owners clean house and start all over. It’s the same as planting a tree, letting it grow for a year, and then hacking it down and planting another one. If it isn’t given ample time to grow it can never blossom.

Unfortunately, the stubbornness that runs rampant throughout the league extends all the way up the ladder to the owners, as well. Their inability to adapt and change their ways creates extensive problems for the rest of the organization.

What does this have to do with analytics?

Obstinacy is a disease that reaches into every facet of an organization. From fourth down decisions, clock management, to analytics. It is a disease that has plagued the league for at least 45 years.

In 1968, Sports Illustrated published an article on the use of computers and analytics in the scouting process of college prospects. Amazingly, a lot of the ideas, concepts, and calculations are still being used today, 45 years later!

Pretty much everything about the NFL has changed since 1968 – schemes, players, and rules. Technology has advanced exponentially allowing us to do more with greater amounts of information in fractions of a second. Yet the NFL has barely taken a half step forward in how they’ve developed the use of analytics since that article was published 45 years ago.

Only over the last few years have teams begun to install analytics departments into their front offices. Some teams now use analytics to parse out what can inhibit their quarterback’s play, while others are using player-tracking systems to help glean information on how practice habits can help or hinder game performance.

These are huge steps that teams have just recently begun to adapt. The analytics push is finally taking hold but it’s 10 years after the original publishing of “Moneyball” by Michael Lewis – a story of how the Oakland Athletics used advanced statistics to find under-valued assets to help them compete with teams who were spending five-times as much on their player payroll.

Baseball Analytics vs. NFL

Baseball is a unique game. There is no clock. The game is only advanced when an out is recorded. Baseball is also unique in that it is nine players reacting to one. In the NFL, it’s eleven players reacting to eleven players.

Baseball is a very stationary sport. All of the players basically start at the same place on every play. The NFL, by comparison, is a very fluid sport where 11 guys can move in any direction after the snap and pre-snap formation movements have made even that phase difficult to quantify. Anyone who has charted the 49ers over the last few years can attest to this.

Even with these dynamics, it took over one hundred years for a major league team to finally discover, accept, and adapt to the stats and analytics that helped determine the quality of a player and his impact on winning games. The first published article on the topic was from Bill James in his abstract in 1977. It took twenty-five years for the idea to take hold and come to fruition in Major League Baseball.

One of the largest differences between the NFL and MLB throughout history is that baseball has always been very numbers driven. Box scores can tell ninety percent of the story of a baseball game because of the depth and accuracy of the stats. Earned run average is a solid metric to determine a pitcher’s worth. Batting average over the years wasn’t perfect but it was a good indicator of a player’s capabilities at the plate. On-base percentage was a widely kept statistic but didn’t garner the attention that batting average did.

The depth of baseball statistics provided a lot of information that could be manipulated to find more accurate ways to evaluate players. There was value in the use of the stats, but perhaps the most important aspect of those statistics is that they existed.

The NFL doesn’t have the same type of statistics currently available to manipulate – at least publicly anyway. Yards, touchdowns, interceptions, and completions are standard references when analyzing or discussing a quarterback’s play. But the flaw in these statistics is that they don’t translate unilaterally across the NFL because of scheme, surrounding talent, and game situations. They’re indicators of other aspects of the game, but not necessarily a quarterback’s true abilities as a passer.

These types of statistics can be considered compilation stats. Simply having the opportunity for a player to compile these stats is as important as the ability to compile them. Matthew Stafford may throw for 4000 yards every season, but it doesn’t mean he’s a better quarterback than Russell Wilson who throws for 3,300. It just means Stafford simply had more opportunities because of scheme, having the best wide receiver in the NFL to which to throw, and a team that often found themselves trailing in high-scoring games and therefore had a necessity to throw forty times per game.

Sacks are a compilation statistic that is useful but it’s a metric that doesn’t quite take the impact of a pass rusher far enough. A pass rusher doesn’t have to bring down a quarterback with the ball in his hand to have an impact on the play. In fact, he doesn’t even have to touch the quarterback to make a difference in the play. A rusher who gets pressure on a quarterback can force him to make a quick decision and throw the ball into coverage, resulting in an incomplete pass or an interception. That’s a massive impact on a play and on the game. But until Pro Football Focus brought “pressures” to the mainstream statistical vernacular, these stats weren’t tracked. It’s still not an official stat even though tracking pressures is more indicative of a good pass rusher than sacks.

Simply put, there are a lot of important NFL stats that don’t exist yet. And because of the stubbornness of the NFL, they likely won’t find their way into box scores, and therefore mainstream football vernacular, until years down the road. Advanced NFL Stats has done a remarkable job of exploring this uncharted territory.

Both Pro Football Focus and Football Outsiders have people who chart every play of every NFL grade and have done so for a few years now. Because of these projects, they have both begun to track statistics that were never previously recorded, like quarterback pressures. Football Outsiders also has taken the information to the next step and created formulas to help decipher team efficiency in different areas. Defense-adjusted Value Over Average, or DVOA, has become a go-to metric for determining the true level of play of a team.

Human Error

As you can see, a lot of the advancements in NFL analytics have come from outside of the league. In other words, the NFL has let everyone else lay a lot of the groundwork and then began to infuse it into the front office. The actual leaps that analytics has taken in the NFL has very little to do with what NFL teams are doing on the inside.

This can likely be attributed to human error in the environment of the NFL. In other words, stubbornness strikes again.

With a larger investment into analytics before the rest of the league caught on, a team could have clearly set themselves apart from the pack and given themselves a major competitive advantage. Instead, the “old way” was accepted as the best way and here teams are, 45 years after that Sports Illustrated article, still using a lot of the same analytical tactics.

To give credit, there are some teams that are now pushing forward into unexplored areas. There is more information available to teams now than there ever has been. However, as important as access to that information is, it’s all irrelevant if the analyzing and application of that information isn’t accurate.

It’s like a carpenter with power tools. First, he must have access to those tools. Second, he has to know how to use them. If used correctly, he can create a beautiful piece of furniture. If used incorrectly, he could saw off his own hand.

Teams will have to sift through large quantities of data, determine what is important and what isn’t, and then take that information and then apply it to the correct areas of the organization and game. Whether it’s learning how to better manage the clock at the end of the game, investigating how to help their quarterback perform better, determining the most effective scheme, negotiating contracts, or evaluating college prospects, teams must use the information correctly to catapult themselves ahead of the competition. If used incorrectly, they could cripple the organization for years to come. The Raiders could have used analytics to better manage their contracts over the last ten years. It’s taken Reggie McKenzie three years to dig Oakland out of salary cap hell.

No matter how much information there is available, it will always be up to humans to translate that data and apply it within the organization. Analytics can help to mitigate biases in most situations but it is always something that is present in nearly every decision and evaluation. It is the responsibility of scouts, coaches, and front office executives to try to make logical, objective decisions based on the information at hand.

Changing The Game

It’s in this vain, that Nate Silver references the “search for intelligent life,” when discussing the adoption of analytics into the NFL. While Silver seems content with blaming the decision makers themselves, I’m more apt to push the onus onto the culture of the NFL.

Coaches have very short tenures and often don’t get a second chance if something goes awry on the first one. It’s the win-now culture that has owners so quick on the firing trigger and coaches are less likely to deviate from the norm because it leaves them open to criticism. Most are content to continue doing things the way they’ve always done so that they can say, “Any coach would have done the same things I did.”

It’s guys like Bill Belichick who have incredible job security that are more apt to take risks and stray from the norm. Remember 4th and two against the Colts? The ability to experiment is something that gives the Patriots a competitive advantage because when they hit on something new, the rest of the league spends money, time, and resources to catch up while the Patriots are reaping the benefits.

The league still doesn’t seem to be able to consistently stop or mimic the two tight- end offense that the Patriots have used recently. While the other teams in the AFC East were trying to find safeties or cornerbacks that could hang with Rob Gronkowski and Aaron Hernandez, the Patriots were building an offensive line and backfield that allowed them to run with power or zone based on how teams decided to match up. Teams spent draft resources to cover those two specific players, which only created a larger gap when the Patriots built an offensive line and strong backfield while teams were focused elsewhere.

Belichick, Chip Kelly, and a few others are willing to draw outside the lines and have created problems that other coaches have never even dreamed about. Analytics is something that has fallen outside the lines for organizations for years. Now that franchises are expanding their horizons, the evolution of analytics could see a major jump in the next few years.

Don’t expect any breakthrough statistics or metrics to come to light anytime soon as teams will protect that sensitive information. The NFL and its teams are much more careful about their internal information than to allow an author to pen a book about their secret formula. We may not see specific metrics developed by teams in the public forum until those metrics have spread throughout much of the league.

Whether public or private, analytics will change the game for the better, even if the obstinacy of the NFL tries to keep it out.

facebooktwittergoogle_plusredditpinterestlinkedinmailby feather
Posted in Coaching, Philosophy | 9 Comments

Tampa Bay Buccaneers 2013 Roster Evaluation

In 2013, the Tampa Bay Buccaneers fell way short of the preseason projections. Because of this, the coaching staff and general manager were unceremoniously released. But what went wrong? Was it coaching or did the roster lack talent?

Like any new general manager and head coach, Jason Licht and Lovie Smith’s first duties after being hired were to evaluate the current roster and find out what needs improvement and where the strengths of the roster lie. I decided to do the same.

Below is the link to my personal evaluation of the Buccaneers’ roster from last season. It’s not a summary, it’s strictly a scouting report on each player that played enough snaps to get a solid idea of his performance. I’ll let you decide where the strengths and weaknesses lie.

2013 TB Roster Eval

facebooktwittergoogle_plusredditpinterestlinkedinmailby feather
Posted in Evaluations | Tagged , , | Leave a comment

Super Bowl XLVIII Preview

The unstoppable force meets the immovable object is a paradox that has been studied for centuries and can be traced back to a 3rd century BC philosophical book, Han Feizi.

Between the Denver offense and the Seattle defense, we have our own modern day version of this paradox.

The Denver offense, led by wobbly-dart throwing Peyton Manning, hung an unimaginable 606 points on defenses this year – the first to ever crack the 600 point mark. They were quite the unstoppable force.

The Seattle defense, led more by its pressure-generating defensive line than its headline-generating All-Pro cornerback Richard Sherman, didn’t record any historical numbers but proved to be the best in the NFL this season by allowing a mere 14.6 points per game – nearly a point lower than the next best team. The were quite the immovable object.

What happens when the two opposing forces meet? The difference in Super Bowl 48 is likely to be found in the chess match played on the field by these teams and coaching staffs rather than the violent collisions of the forces.

“We don’t vary or disguise coverage on anybody. We play a pretty simple defense; for the most part you know what we are going to do every play. And you got to line up and play it and I think that is how we have been all season. This is the last game of the season — there is no time to change it now. No new wrinkles for us.” – Richard Sherman

Sherman is correct. The Seahawks are a very basic team when it comes to how they play their coverages. That doesn’t mean that they’re any less effective than other teams who use multiplicity in their defensive secondary to confuse and bait quarterbacks into throwing into trouble. In fact, the composition and scheme of the Seattle secondary is a terrific fit to stop Peyton Manning and his plethora of talented receivers.

Every team in the league knows that blitzing Peyton Manning is only giving him a larger window to throw into. He is too good at deciphering a defense in the pre-snap phase and then executing in the post-snap phase with precision and concision. Peyton is a model of efficiency with the football and defenses know it. What stands out to me on film when watching Peyton is his pace through his progressions. He can get through 4 or 5 reads faster than most quarterbacks can take their eyes off their primary target. This allows him to make quick decisions with the football and exploit favorable matchups with minimal risk.

The Jaguars blitzed twice against Manning in week 6, and even though they were thumped 35-19, that game was close late in the 3rd quarter until Denver’s overwhelming talent advantage took over. Rex Ryan talked about using nothing but “seven-man spacing” on Mike & Mike this week, meaning he was able to play 2-deep safeties and not have to blitz excessively – something Ryan is known for.

The idea of not blitzing Manning comes from multiple advantages: The ability to drop seven men into coverage making the windows tighter for Manning and it invites the run which takes the ball out of Manning’s hands – always a good thing for a defense. Blitzes are also often used in the NFL to confuse quarterbacks, something that Manning is unorthodoxly immune to. Manning will decode where the blitz is coming from and his chemistry with his receivers allows them to “go hot” or break off their routes to find the open space and generate big gains.

This idea is the main reason that I’ve championed a “static” pre-snap look on defense against Manning. This naturally creates other issues for a defense like an inability to get to their man or spot in coverage, but it eliminates Peyton being able to win in the pre-snap phase. If he has nothing to read, he has no advantage to gain. Aligning in a basic cover 2, or for the Seahawks a cover 3, shell, means that all of Manning’s keys come in the post-snap phase. Even that half-second of hesitation on his part can allow a defender to get in place or for the pass rush to get to him and cause a sack or a tough throw.

Dropping seven men into coverage means that a defense is only rushing four. This is something the Seahawks are comfortable with because of the pressure generated by their defensive line. The addition of Michael Bennett was likely the biggest move of the offseason as he generated the 3rd highest pass rush productivity of 4-3 defensive ends this season, per Pro Football Focus, trailing behind Robert Quinn and Cameron Wake.

The Seattle defense may be unvarying on the back end, but they love to mix up their front four with good frequency. Bennett will rush from both the outside and inside at times and they will use a rotation of Cliff Avril, Chris Clemons, Brandon Mebane, Red Bryant, and Tony McDaniel to keep offenses guessing and adjusting protection schemes. Seattle will find a matchup advantage they like along the line of scrimmage and will consistently try to exploit it in multiple ways to help proliferate pressure.

Denver’s offensive line will be tasked with maintaining inside leverage and keeping the pressure out of Manning’s face. Manning is nearly impervious to edge pressure because his manipulation of the pocket and ability to climb the ladder are extraordinary. If the guards and center can anchor inside then Manning will stand tall and be able to deliver the ball. Everyone knows that Manning isn’t a threat to run the ball as scouts could time his 40 with a desk calendar, so the Seahawks will give little thought of trying to play contain on the edges when rushing.

Manning will see a lot of cover 1 and cover 3 zone from the Seahawks. Earl Thomas is the rangiest and most instinctive free safety I have ever seen play and he’s really the lynchpin that allows this defense to cover in the manner that it does. Thomas plays centerfield as the single-high safety while Kam Chancellor often plays in the box and helps defend the run. The spin that the Seahawks put on their coverages is that they’ll play cover 3 much like they play cover 1 in how the corners will align as if they’re in man coverage. Many teams will play press-bail, where the corners align in press but drop into off-coverage just before the snap, when in cover 3. The Seahawks will maintain their press alignment and be physical at the line of scrimmage, knowing that Thomas is likely to be able to get to ball if it’s thrown deep down the sidelines. It also helps that the corners for Seattle are so tall and lengthy and are able to swat down balls that other corners couldn’t touch.

In terms of personnel packages, the Broncos play predominantly out of their 11 personnel (1 running back, 1 tight end, 3 wide receivers). Seattle will want to match up in their base 4-3 personnel, but with Wes Welker and Julius Thomas such receiving threats, they may have to go to their nickel packages more than they’d prefer. Sherman may move around and shadow Demaryius Thomas at times, but I don’t think he’ll align that way exclusively. Seattle will blitz selectively – especially when Bruce Irvin is in the game – and will do so in certain areas of the field or down and distances.

On the opposite side of the ball, no starting quarterback threw fewer times per game this season than Russell Wilson. This offense is built on a physical run game and the passing game plays off of that. Marshawn Lynch will finally be able to stop not-talking and be all about that action, boss. Lynch will see a heavy workload while Robert Turbin may garner a few carries, depending on how the game goes.

It wouldn’t surprise me to see the Seahawks come out with a very conservative gameplan earlier on. Turnovers are killers in any game, but against Peyton Manning they’re death sentences. Plus, with every second that Manning is on the sideline, that’s another second he’s unable to put points on the board. The offense can’t give points to the Broncos or else they’ll never be able to keep up.

A little like Seattle, Denver’s strengths lie in their defensive line and cornerbacks. The biggest difference is that the Denver safeties can be exploited fairly easily. Defensive tackle Terrance Knighton has been a penetrating force on the interior in both collapsing the pocket in the pass game and disrupting the run. Robert Ayers flashes the ability to rush the quarterback but he’s inconsistent. How Ayers plays could dictate a lot of how this game goes for the Seahawks on offense.

Russell Wilson has looked frenetic and confused lately. He’s faced a murderer’s row of defenses lately, and Denver’s defense is a slight step down from those, but Wilson needs to find the poise that was his hallmark since he started in this league. He seems hurried recently and it’s led to unreliable reads and inaccurate passes. He’s still one of the best quarterbacks in the league at throwing on the run and will make a “wow” throw or 2 on Sunday.

The Seattle receivers aren’t imposing or dominating, but they’re slippery. It’s hard for defenders to get a good jam at the line of scrimmage because they’re quick and it’s hard to get hands on them. Golden Tate and Doug Baldwin are deceptively quick and sort of oddly get over top of cornerbacks with consistency even though neither are burners. My favorite receiver to watch on film is Jermaine Kearse who is a blast in the run game, both figuratively and literally. Kearse is one of the best blocking receivers in the league and will wall off defenders with ease, or blow them up when he cracks down on them. He also has some big-play capability in his game and if they Seahawks can get him matched up on a safety, he’s likely to hit for a big gain.

There are a couple of keys to the game that I believe will turn this game in favor of one team or another. I’m omitting turnovers from the discussion because they’re always a key to the game. Instead, I believe that physicality and interior line play will determine the outcome of the game on Sunday.

The Seahawks are the most physical team that the Broncos will have faced all year because, well, they’re the most physical team in the league. The Seattle secondary isn’t dubbed the Legion of Boom for nothing. How the Denver receivers handle that physicality is going to drive how effective they can be. Welker has shown that he doesn’t like getting knocked around. Decker and Demaryius Thomas aren’t afraid to get physical but I’m interested to see how they respond after getting hammered by Chancellor or Thomas one good time. If they can take the hit and keep going after the ball in the air with no regard to life or limb, they’ve got a shot.

Interior line play is important because of the effect is has on quarterbacks, especially immobile quarterbacks like Peyton Manning. But the real key here is how well the front seven of Seattle can stop Knowshon Moreno and Montee Ball without help from the secondary. I actually anticipate that the Seahawks will play more 2-deep coverage this game than normal, but will still stick to their base cover 1 or 3 defenses. Cover 2 means there is less help against the run and more space for running backs to roam. If the Seattle interior line can stuff the run without much help, then it turns Denver into a one-dimensional football team and lets the pass rushers tee it up against Manning.

Interior line play for the Broncos is going to be important in keeping Lynch & Co. from getting to the second level and using his Skittles-fueled burst to run through and over smaller players. Knighton has played tremendously well of late and I don’t expect him to have much problem with the guards of Seattle. It will be interesting to see if offensive line coach Tom Cable decides to double team Knighton for most of the game. If the Broncos can force Wilson to throw more than he’s used to, I believe they’ll have shots at turnovers, and that’s the only way I see Denver winning this game.

It’s hard to factor in the “Peyton Manning effect” in terms of how impactful he can really be, but I think the Seahawks are a better team all-around than the Broncos are. The problem is that the Broncos are better at the one spot that truly matters: quarterback. The highly-touted matchup is the Denver offense versus the Seattle defense, and it’s fun to discuss, but the difference in level of play there isn’t vast. Where I think the greater gap in talent/matchup differential is the Seattle offense versus the Denver defense. I think that Seattle has the chance to keep Manning at bay with the physicality and scheme of the defense, while the Denver defense doesn’t have the best matchups against the Seattle offense.

Final verdict: If Russell Wilson takes care of the ball, Seattle should win this game. The immovable object gets moved, but still halts the unstoppable object.

facebooktwittergoogle_plusredditpinterestlinkedinmailby feather
Posted in Game Preview | Tagged , , , | Leave a comment