The Game Behind the Game: Data analytics, the NBA, MLB, and NFL—is football behind the times?

By John Divine

On February 3, Super Bowl LIII will be played in Atlanta, Ga. Routinely the most-watched television program of the year, over 100 million Americans will tune in to see some of the finest athletes in the world square off against each other in a battle of brawn, speed, wit, and—hopefully—analytics.

To be sure, athletes tend to get bigger, faster, and stronger over time, but as any longtime sports fan knows, even a team that’s skilled on paper is bound to disappoint more often than you might expect.

All things being equal, it certainly helps to have skilled players on your squad. Next-level talent can go a long way. But no team can expect that talent to be there for the long term.

It’s not squat reps or glasses of milk that separate the wheat from the chaff over a decade or more.

It’s a philosophy. It’s coaching. It’s strategy. But even these concepts are derivative. At the top of the pyramid, senior to everything else, is data.

Football—a Sport in Denial

Sports are increasingly data-driven. But of the three major U.S. sports leagues—Major League Baseball (MLB), the National Basketball Association (NBA), and the National Football League (NFL)—the NFL has been the one most reluctant to adapt.

When the two best NFL teams face off in Mercedes-Benz Stadium in early February, we can rightly marvel at their athletic feats on the field. The kick returns, the tiptoeing around defenders, the swim move that leads to a sack, or the jumped route that leads to a game-changing interception.

The NFL boasts some of the best athletes on the planet, putting on a show for the world to see. But sadly, their talents are being misused. The game, as far as the data is concerned, is being played the wrong way.

Two hundred million eyeballs—and that’s just in the U.S. Roughly four dozen elite athletes whose physical bravado is the product of millions of years of evolution and natural selection. Ironically, the evolution of the game they play has been painfully slow.

It doesn’t have to be this way.

Part 1.

Data: A Rich Resource Giving Sports the Option to Evolve.

Assume you’re the general manager of a major U.S. pro sports franchise. MLB, NBA, NFL—take your pick, the sport itself is irrelevant.

Imagine you’ve been thrust into a league with the exact same level of talent on every team. Total parity: The year before, every team was .500. Playoff outcomes were determined by coin flips. A total nightmare.

Further imagine that nothing is expected to change in the coming year, and that the ownership group, in its boredom and desperation, has offered you enormous rewards if you can figure out how to win with relative consistency over the coming years.

Now, imagine you learn of a magical, growing, rich, and plentiful resource that taught you lessons from history, statistics, probability theory, mathematics. Imagine it could be tapped. Furthermore, it happens to be free. Would you avail yourself of it?

Perhaps a better question: Would you ever stop worshipping at its altar?

Thankfully, this magical resource is real. But surprisingly, the three major American sports don’t appreciate this resource equally.


Pro baseball managers have long appreciated data and the fountain of useful knowledge it provides. But just because you have a library card and like to read doesn’t mean you’re checking out the best books.

In recent decades, presumably after getting familiar with data’s equivalent of the Dewey Decimal system, baseball managers have been utilizing the numbers with increasing efficiency.

The story may ring a bell.

The intersection of data analytics and professional sports went mainstream in 2011 with Moneyball, the Oscar-nominated film[i] starring Brad Pitt and Jonah Hill chronicling the creative tactics employed by Oakland A’s general manager Billy Beane in 2002 to field a competitive team despite having a budget of $39.7 million—less than a third of the Yankees’ $125.9 million payroll, and third-lowest in the league.

An archetypal David vs. Goliath tale, Moneyball shows that in order to compete with giants, the little guys have to think differently in order to be successful. Adopting a data-based approach to the game known as sabermetrics, Beane shifted the organization’s focus to fielding players with statistics undervalued by the market but highly valuable in a tradition-agnostic strategy guided by a priori reasoning.

Having lost three star players to free agency entering the year, Beane fundamentally changed the A’s’ approach to signing and fielding players. Despite a rough start, Oakland eventually went on a 20-game win streak and finished with the second-best record in the MLB (next to the star-studded New York Yankees).

Though they lost in the American League Division Series, the A’s arguably stretched each dollar of payroll further than any other team in the league that year—and other organizations took notice. Two years later, the 2004 Boston Red Sox, adopting the sabermetrics approach, won the World Series, and today Beane is considered a pioneer in a sport that is arguably the most progressive of the three major sports in using data—not emotion, perception, tradition, or previously prized metrics—to inform team philosophy and strategy.

The MLB in 2019 is far different from the MLB of 1999. Gone is the obsession with batting average for hitters and ERA for pitchers. On-base percentage, slugging percentage, on-base plus slugging (OPS), defense-independent pitching statistics (DIPS), value over replacement player (VORP), and wins above replacement (WAR) are some of the prized stats today.

Of course, baseball is ideal ground for stat nerds and has been since its inception. Unlike basketball or football, every play of the game, at its core, begins with an isolated one-on-one scenario between pitcher and batter.

Still, credit must be given where it’s due. Despite inefficiencies left in the game as recently as the turn of the millennium, baseball gets more efficient every year, and recent innovations like outfield shifts in accordance with batter tendencies continue to squeeze out the risk of strategic errors.


Pro basketball has changed dramatically in recent years as an emphasis on data analytics produced not only some new ways of measuring the offensive and defensive capabilities of players and teams, but essentially became prescriptive.

The improved understanding of NBA data today gives front offices far less leeway about what kind of team they should assemble—or what kind of skills players need to have—if they want to win.

It’s a lesson the NFL would be wise to heed.

Using the Trifecta

The most obvious example of what the data prescribed? Shoot more 3-pointers. A lot more 3-pointers.

In 1979–80, the season the 3-point line was implemented, NBA teams shot 2.8 threes a game. By 1989–90, the number of threes attempted more than doubled to 6.6. By 1999–00? Another doubling to 13.7. And if current trends continue, 2018–19 will be the eighth straight year of increased 3-point attempts, with teams jacking up over 30 a game.

The shot that, in its debut season, accounted for just 3 percent of total shots taken, 40 years later, accounts for over one-third.

The reason for this increase is entirely intuitive: A shot worth 50 percent more than another, as long as it goes in at least two-thirds as frequently as its lesser cousin, is economical.

In other words, even a player shooting 32 percent from behind the arc—which is considered awful in the NBA—is shooting just as well, all things considered, as someone who shoots a respectable 48 percent from the field and never attempts a three.

The useful metric of effective field goal percentage (EFG%), which only became a widely used statistic for stat-heads in the late 2000s, adjusts for this. A player shooting 1-for-3 from three in an average game has an EFG% of 50 percent, since it would take 1.5 out of 3 two-point shots to score the same number of points.

And if he was a proportionately better 3-point shooter? All you had to do was convince him that taking—and therefore missing—more shots, all things considered, was better for the team.

The threeball revolution has brought EFG% from a leaguewide average of 47.1 percent in 2003-04 to an all-time high of 52.1 percent in 2017-18, the last full season played.

At first, the reason this eureka moment didn’t occur in 1979, when the NBA first instituted the 3-point line, seems baffling and inexplicable. While tradition may at first appear a reasonable surface-level culprit that explains the slow adoption of the 3-pointer, the truth is that pros 40 years ago simply didn’t have the range today’s players have.

In fact, NBA players were so unsophisticated from three when the new rule was unveiled that the shot wasn’t worth the effort for the average team. In 1979-80, 3-point shooting percentage of the entire NBA was a miserable 28 percent. For the first seven seasons, up through 1985-86, the league’s average never eclipsed 28.2 percent. In a league shooting 48 to 49 percent from the field overall, that didn’t cut the mustard. It wasn’t until the 1988-89 season, when teams shot 32.3 percent from deep and 47.7 percent overall, that the shot became statistically viable for the league as a whole.

Of course, it’s important to understand that the numbers given are averages, and even if it took the NBA as a whole eight or nine years to make threes worthwhile in aggregate, that doesn’t preclude the skillset of any individual team from being ahead of the time or having dangerous sharpshooters. A properly constituted team can show meaningful deviation from the mean, as the Boston Celtics did in the 1979-80 season, shooting 38.4 percent from deep.

But that aggregate skillset problem the 3-pointer presented for the NBA in the 80s didn’t last long, and by the end of the decade a new generation of players, just beginning to understand the potential the 3-pointer had to disrupt the game, would penetrate the league and never look back. The modern NBA league average rarely dips below 35 percent in a season, making the arc not just viable, but a deadly weapon that every team must utilize if it wishes to compete.

What Today’s 3-Point Revolution Is Doing for—and to—the Game

With 3-pointers constituting over a third of the shots NBA teams launch on average each game, the end result isn’t just more effective offense.

It spreads out the floor. If Steph Curry can pull up and reliably drain 30-footers (the 3-point arc is 23.75 feet away from the basket, excepting corner threes, which are 22 feet away), defenses have to play farther away from the rim. This makes it much easier to beat defenders and pass to open teammates (as opponents abandon their assignments and rush to defend the man driving), which in turn benefits the 3-point shot-making ability of everyone else on a sharpshooters’ team.

“Positionless basketball” is emerging. For decades, it was simply an unspoken, unquestioned, and taken-for-granted rule in pro basketball: There was a point guard (PG, or the “1” position), shooting guard (SG, 2), small forward (SF, 3), power forward (PF, 4) and center (C, 5), with the last two playing post-up ball close to the basket. That was your starting five.

The emergence of the 3-pointer has led to stacking teams with shooters, who are usually guards, and thus, shorter. Even the archetypal big men of the NBA are disappearing, in favor of leaner, more nimble “stretch 4s” who can shoot the 3, and centers who are great passers and decent shooters themselves.

In general, distinctions between the 1 and 5 spots are becoming far less clear than they used to be.

Today’s centers also must be able to do one more crucial thing in today’s game—a rare skill that compounds the accelerating trend toward positionless basketball…

Bigs must be able to guard opponents far away from the rim, by the 3-point line. Gone are the Anthony Masons and “Tractor” Traylors of the game—the 6’ 7”, 265-pounders who can’t run the court well and are only good for some physical, back-to-the-basket post plays and some above-average rebounding.

In today’s NBA, teams need to be able to defend the 3-pointer, not just shoot it well. If you’re a good post-up player but can’t defend the stretch-4? You’re seen as a net liability to your team.

Consider 7-foot-3, 290-pound L.A. Clippers center Boban Marjanovic. While his name may not ring a bell, he is a historically elite scorer. Period. One of the top five all-time. Going into December, Marjanovic’s career effective field goal percentage was 57.4 percent, the fourth-best clip for all players since 1973 who have played at least 100 games and scored at least 30 points per 100 possessions.[ii]

His true shooting percentage, which takes 2-pointers, 3-pointers and free throws into account, is 63.7 percent, the best all-time by a respectable margin. Steph Curry is second at 62.3 percent.

By the way, Marjanovic also hauls in over 20 rebounds per 100 possessions. That too is unparalleled, although this time second place is even further back. Shaquille O’Neal, the best rebounder of any retired player meeting the above criteria, got 16.1 boards per 100 possessions.

So why isn’t Boban a household name, the way Shaq is?

It’s impossible to know what would happen if the NBA weren’t a three-happy league, but it certainly wouldn’t hurt Marjanovic’s chances at a sneaker deal. As of Nov. 30, Marjanovic was playing just 11.3 minutes per game (a career high) in a league worried about his ability to defend deep-shooting bigs and the pick-and-roll game.

Perhaps the pendulum is starting to swing too far in the NBA. Just looking at the data, it’s not at all clear that limiting Marjanovic’s minutes is a good move. Even defensively, his defensive box plus-minus is +1.7. Anything above zero means you’re benefiting your team purely with your defense.

Back in the NBA of the’90s, most teams would find a way to give a guy averaging 25 points and 15 rebounds every three quarters a little bit more shine.

Pro basketball has come a long way since then, with the 3-pointer at the center of it all. We may very well be witnessing the NBA establish a new tradition, a common knowledge from which it is dangerous to stray.

That means no more 5’3” point guards like Muggsy Bogues. It means post-up play is dying. It means one of the most efficient scorers in NBA history sometimes doesn’t see the floor (for better or worse). And it means the old guard is cranky and complaining about the new status quo.

Here’s what Greg Popovich, who coached the San Antonio Spurs to five NBA championships, said in November about the game today:

Now you look at a stat sheet after a game and the first thing you look at is the threes. If you made threes and the other team didn’t, you win. You don’t even look at the rebounds or the turnovers or how much transition [defense] was involved. You don’t even care. That’s how much an impact the three-point shot has and it’s evidenced by how everybody plays.

I hate it, but I always have. I’ve hated the three for 20 years. … If we’re going to make it a different game, let’s have a four-point play.

Because if everybody likes the three, they’ll really like the four. People will jump out of their seats if you have a five-point play. It will be great. There’s no basketball anymore, there’s no beauty in it. It’s pretty boring. But it is what it is and you need to work with it.[iii]


Agree or disagree with Popovich, the startling thing about this comment is the passive view Pop takes on the three-pointer, referencing it as a phenomenon that happened to the NBA—and to him—against his will.

In reality, the 3-pointer has been around since the late’70s. Four decades. And it took that long for an entire league of billionaire owners, brilliant coaches, top-notch scouts, and world-class talent to start utilizing a fundamental part of the game to its full potential.

Granted, in the beginning, the skillset on a league-wide basis hadn’t been cultivated, and the 3-point shot was in its infancy, considered more a spectacle than a legitimate tool waiting to be exploited. But in less than a decade it had become economical, and still only eight percent of attempts were from beyond the arc. Its usage remained lower than it should’ve been long after it became an obvious asset to the average NBA team.

Let’s just put it frankly: The 3-pointer, for decades, was arguably the lowest-hanging fruit in professional sports.


Part 2: The Contrarian.

“If we see something that might benefit us, and there’s some sound reasoning behind it, we’re gonna do it,” says Kevin Kelley, head coach for the Pulaski Academy, a 5A high school football team in Little Rock, Ark.[iv]

“Everybody wants to win, but are you wanting to win bad enough to do what you have to do to win? To try something different to risk being the fool?” Kelley asks.

Kelley’s own answer to that question came pretty quickly after taking up the reins at Pulaski Academy in 2003. When he got the job, he asked himself, “Wait a minute, how are we going to be any better than we have been?”

“I think the school had been—in the history of the school—to the semifinals two times, and never past that. So how are, all of a sudden, we going to make that jump? I mean I’m no better than anybody else. I better start thinking.”

The new coach broke down the game of football from first principles, shedding the rule-of-thumb approach to play-calling and the traditional wisdom that’s ruled the game of football since time immemorial.

Like Billy Beane, Kevin Kelley was forced to face a brutal fact: A cookie-cutter approach to the game will get you cookie-cutter results. All coaches at any competitive level want to woo the best talent and practice hard and often and yada, yada, yada. To set your organization apart long-term, you have to zig when others zag.

So zig he did:

There’s not a 4th-and-anything I wouldn’t go for. You could have 4th-and-10 on your own 5 yard line.

Kelley thinks about these situations conditionally, like a walking, talking if-then Excel function.

You go for that, you throw an incomplete pass, they’re gonna get the ball on the 5. An offense will score from the zero to 10 yard line about 92 percent of the time.

That sounds pretty bad. Let’s consider the alternative: punting. What’s that if-then look like?

You’re not gonna punt it as far because your guy’s taking one step and kicking it off. You’re gonna get a net of 32 to 33 yards in that situation. That means they’re gonna get the ball inside your 40.

Inside the 40, teams were scoring a touchdown 77 percent of the time. So you were giving up 15 percent—they’re gonna score at 77 percent as opposed to 92.

Over the history of us going for it, I’ve got over a 50 percent chance of making that. Put all that together and I should go for it.

So, Kelley did go for it. The team’s unofficial motto became “Never Punt.” Through 2016, they’d punted eight times in as many years. Onside kicking? That’s fair game, too—in fact, they do it every time (“unless we’re up 21 points,” Kelley says). Attempting 2-point conversions (“If you can get it over 50 percent of the time, you go for 2”) doesn’t bother him in the slightest.

How does this unusual strategy work out?

Between 2003 and 2017, Kelley’s teams won seven state championships. Playing up two divisions against schools with bigger enrollments (and thus tougher competition). In late 2018, they were going for their fifth consecutive state title.

That incredible success sprouted because of—not in spite of—the Arkansas coach’s willingness to depart with convention. There’s nothing wrong with the old school at face value, but if the numbers aren’t there, one has to ask, why is the crowd?

“I’m not the gambler. If you’re doing it against the numbers, that’s when you’re the gambler.”

Sure, Kelley zigs. But he zigs rationally, if you will. And he openly prescribes this analytically driven, first-principles approach to coaches at higher levels willing to make the plunge.

“I think the people that need to implement this system the most are the bottom-level teams that are consistently bad in the NFL and in [college football]. Analytics, this could be the future of football.

“What do they have to lose? They’ve been doing the same thing over and over, and getting the same result. Why would you not try something different?”

That question, like Kelley’s team, is damn good.

Part 3: The Big Leagues.

It’s not often that you see a new NFL coach touch down in a city and shake things up with onside kicks, an abolition of the punt unit, trick plays and 2-point conversions. Change is gradual, and tradition is sacred.

But to dramatically improve the efficiency of an NFL offense, you don’t have to be radical. To begin with, you could just change one simple thing: pass more. A lot more. And because games aren’t getting any longer, that means running less.

We hear a lot about how teams are passing more than they used to. That’s true. What’s even more remarkable is that they’re passing more effectively than they used to as well.

You can see this anecdotally just by looking at some of the well-publicized milestones. In 1984, Dan Marino became the first player in the history of the NFL to pass for 5,000 yards in a single season. It wouldn’t be done again until Drew Brees accomplished the feat in 2008, 24 years later.

But after that 2008 season, and through the end of the 2017 season, that feat was accomplished seven more times. In 2011, three players passed for 5,000 yards in a single year.

Here’s how passing has evolved on a league-wide level: Between 1980 and 2006, average passing yards per game remained stubbornly fixed between 190 and 210 yards per game, with few exceptions. Teams attempted 30.6 passes per game in 1980, averaging 7 yards per attempt. Twenty-six years later, teams were passing 32 times a game and earning 6.9 yards per attempt.

Steady Eddie.

Are you ready for the stats on this “passing revolution” you may have heard of in the media? Between 2006 and 2017, attempts per game grew from 32 to 34.2 and passing yards per game went from 204.8 to 224.4. Yards per attempt ticked up slightly from 6.9 back to 7.

In other words, modest progress. Very modest. We aren’t seeing the consistent doubling in 3-point attempts that the NBA seems to experience every decade, for instance. And it’s not quite clear—yet—why passing like a maniac should be the goal.

Let’s have a look at the run game.

Unfortunately, there’s not much to see. A superstitious person might find genuine reason to believe in the football gods by looking at the data, which seem to mandate roughly 27 carries a game at a 4.1-yard average for 110 yards a game or so from 1990 to present.

When the game is broken down and viewed in this binary way, as a game of opportunity cost, decisions become much simpler to make. You either pass or you run. Which is expected to get you further down the field? The pass. All day long. It’s a no-brainer, and it’s so unfair to the defense that it’s not even allowed in the game of rugby.

Take a moment to seriously consider the art of the pass as a philosophical exercise.

The advantages it offers are litany:

  • Receivers know the route. Defenders do not.
  • No touching receivers within 5 yards of the line of scrimmage.
  • A football can travel far faster than players can run, so as the QB, you can throw to where the receiver’s going to be (which again, you know to the footstep in a called play).
  • Increased emphasis by officials on defensive pass interference—“be gentle or forfeit field position.”
  • Increased enforcement of the roughing the passer penalty.
  • Increased enforcement of defenseless receiver rules.

That’s a lot of built-in advantage. So it makes sense passing would result in more yards than an average run. But in 2017, with the average yards per pass attempt at 7 and yards per rush attempt at 4.1, with an expected gain that’s 70 percent higher with a pass—why ever run?

The logical answer, other than an occasional play to take advantage of a defensive scheme or perhaps a hard-to-fail 4th-and-centimeters quarterback sneak, is frankly unclear.

That answer doesn’t always jibe with longtime football viewers, but here’s why the Le’Veon Bell of today—the running back who’s sitting out this season with the Pittsburgh Steelers—could be tomorrow’s Boban Marjanovic (minus the efficiency): The data do not align with the traditional narrative.

Football gods of yore apparently handed Moses a commandment that teams must run about 27 times a game in order to “open up the pass game.” Once they respect your running threat, then you can murder them with the pass game.

What we’ve seen however, is that as pass attempts slowly increased from 30 in 1990 to above 34 in 2017, the effectiveness of the pass trended way up with it. Completion percentage (56 percent to 62.1 percent) and passer rating (77.3 percent to 86.9 percent) both soared while interception percentage (3.6 percent to 2.5 percent) and sack percentage (7.3 percent to 6.4 percent) both declined.

It’s as if Steph Curry were a timid, fringe player on an NBA team. Imagine that team had data on Steph that said, “This guy is an incredible 3-point shooter—we get a lot more points per shot with this guy than the next-best option … and there’s another thing too, not only does he shoot 40 percent from three, but when you make him shoot more, he makes a higher percentage!”

Now imagine seeing that data, thinking about it for a second, and deciding to do this about it: “Steph, instead of shooting 2.2 threes a game, you’ve got the green light on 2.6. Don’t say I never did anything for ya.”

To be fair, there’s an important caveat here that fans should understand: not every team has the same dynamics, and some can have poor quarterbacks and receiving corps while fielding a truly talented running back. NFL averages are just that – averages – and a given team’s strengths and weaknesses do matter when determining strategy.

But still, barring the extreme outlier here and there, the Steph Curry analogy holds up. With the secular improvement in passing efficiency and the longtime outperformance of pass attempts over rushing attempts, passing slightly more per game simply isn’t the step towards optimal performance the NFL needs.

No. If the average team has a proverbial Steph on their bench, you don’t take his usage rate up marginally. You have him shoot until his arm gets tired.

NFL Coaches Also Grossly Bungle Fourth-Down Situations

These situations—a team down to its final play of the drive, and choosing to punt—are not rare. Unfortunately, there are no Kevin Kelleys in the NFL.

Whether it’s higher stakes, a bigger stage, tradition, pride, fear, or ignorance, teams prefer to go for the low-risk, zero-reward option. To Kelley, the football is sacred. He thinks of punts and kickoffs as turnovers to be avoided at all costs. To NFL and college coaches, the ball is cheap.

Between 1994 and 2017, professional football teams punted 72 percent of the time on 4th down with 5 yards or less to go.[v] The teams that went for it ended up getting a first down 56.4 percent of the time. On 4th-and-3 or less, the numbers go to 59.6 percent. On 4th-and-1, 65.6 percent.

Again, keep in mind that these are aggregate numbers. They can’t be extrapolated to every individual team, and certain teams have been far better or far worse than average over time at converting fourth downs. Team makeup matters.

With that said, aggregate numbers still tell a compelling story. And the story, it turns out, is getting better. Through mid-December, coaches were showing a far higher risk tolerance in the 2018-19 NFL season than they had in the previous quarter century – and they’re being rewarded for it.

Through December 16, punting on 4th-and-5 or less was down materially, with teams punting just 63 percent of the time instead of 72 percent. On top of that, the so-called “risk-takers” (some might call them “stat-observers”) were being rewarded for their more liberal playcalling. Compared to the 1994-2017 period, teams going for it on 4th-and-5 or less got a first down 62.2 percent of the time (vs. 56.4 percent); 4th-and-3 or less conversions reached 66.4 percent (vs. 59.6 percent); and 4th-and-1 conversions grew to 71.4 percent (from 65.6 percent).

Time to Redefine Risk

By viewing the NFL alongside the MLB and NBA, we can gain a simple but mighty insight: Athletes aren’t the only thing with the power to usher in a new era in how sports are played. Even mainstream sports with hundreds of millions of fans can be turned on their head by bold coaches and managers willing to take risks, players willing to execute, and organizations willing to put their chips on the table.

Although, as Coach Kelley says, is it really those working with the numbers to back them up that take the risk? For a long time, the NBA had a misconception of what risk looked like on a shot chart, and the MLB undervalued many hidden gems in the draft and free agency.

The glib tendency to assume that there’s nothing new under the sun—that if better strategies existed they’d have been conceived, attempted, and systematically exploited—gives too much credit to the coaches, managers, and institutions that have shaped sports throughout history.

The 3-point line existed for decades, full of potential energy, waiting to be put to proper use by anyone with the chutzpah to look temporarily foolish in order to increase offensive production. Tradition mandated that there was a way an NBA team was supposed to be organized, and the basketball world thought in terms of five positions and what an archetypal player at each position should do.

As it turns out, that was objectively not the best way to get the most out of each shot. Legendary coaches, players, general managers and owners in the ’80s, ’90s and aughts largely missed this, with the simple but effective 3-point line hiding beneath everyone’s noses, remaining underutilized until the last decade.

This change has dramatically impacted how we view the game: who is considered an ideal player; how teams draft, coach, train, and play; and even what players will make it to the professional level. Perhaps most important, it establishes a new popular philosophy and default way of thinking about the game, and influences who kids idolize and what skills they develop to shape generations ahead.

The MLB, despite being virtually designed for statistical analysis, overvalued somewhat arbitrary metrics that had traditionally been valued since the 1800s until sabermetrics swept the sport around the turn of the millennium, guiding the managerial philosophy of the curse-breaking 2004 Boston Red Sox world championship team.

The efficient market hypothesis, although perhaps debatable in the stock market, cannot be debated when it comes to today’s major sports—the de facto, traditional way of playing the game today is downright arbitrary, based on shoddy assumptions and antiquated concepts of what’s considered wise.

Especially in football.

At the level of both the NFL and premier Division I college football programs, we find a failure to consider game-time strategy and decisions from first principles. Data seems to be used primarily to reinforce existing biases and belief systems.

While 2018 has generally seen an uptick in teams’ willingness to abide by what the numbers seem to unequivocally prescribe, we just don’t know if the 2018-2019 season will end up being an anomaly or a meaningful inflection point. Either way, pro football has plenty of room to improve, and the fact that the NFL isn’t played nearly as optimally as the NBA or MLB is both disappointing and exciting at the same time.

In its defense, it makes sense that football would be the last of the Big Three sports to reach enlightenment, stop worrying, and love the data. There’s a lot of money in the sport, and head coaches oversee large staffs, all trying to feed their families and advance their careers without ruining them. Worse, there are only 16 games in the NFL regular season, compared to 82 in the NBA and 162 in the MLB. Every game matters. Scrimmaging risks injury and careers. Experimentation is expensive.

But it’s time. The numbers are there to support a fanatical number of passing attempts. And the fourth down, like the 3-point line, is still a feature of the game itself—an enormous, untapped part of the game, bursting at the seams with potential energy like a boulder atop a mountain, just waiting for a push.


John Divine is a freelance writer based in Washington, D.C. He is a senior investing reporter for U.S. News and World Report


[i] The movie was based on the 2003 Michael Lewis book by the same name.

[ii] Via Basketball-Reference.com.

[iii] “How Has Three-Point Shooting Changed the Game?” NBA.com, Nov. 28, 2018.

[iv] Quotations and citations in this section are from “Meet Kevin Kelley: The Coach Who Never Punts,” Bleacher Report video, Oct. 13, 2016; and “Spirit 52: Ep. 47—Never Punt,” Sport & Story video, July 26, 2018.

[v] All statistical references in this section are via Pro-Football-Reference.com.

Next article How Do You Measure a Gerrymander?
Previous article Demystifying The Art of War—No philosophical treatise, this classic offers practical advice for anyone engaged in conflict

Related posts