Most popular posts
- What makes great boards great
- The fate of control
- March Madness and the availability heuristic
- When business promotes honesty
- Due diligence: mine, yours, and ours
- Alligator Alley and the Flagler (?!) Dolphins
- Untangling skill and luck in sports
- The Southeastern Growth Corridors
- Dead cats and iterative collaboration
- Empirical evidence: power corrupts?
- A startup culture poses unique ethical challenges
- Warren Buffett and after-tax returns
- Is the secret to national prosperity large corporations or start-ups?
- This is the disclosure gap worrying the SEC?
- "We challenged the dogma, and it was incorrect"
- Our column in the Tampa Bay Business Journal
- Our letter in the Wall Street Journal
Other sites we recommend
Big Data in The Big D back in the Day
Before there was Moneyball, there was a little expansion football team in Dallas who invented Big Data (in sports) on their way to 20 consecutive winning seasons and 5 of the first XIII Super Bowls. They managed to win just 2 of those 5, losing the others by 3, 4, and 4 points.
Yet even those losses serve as evidence in support of the idea of “Moneyball:” Super Bowl V was so error-filled it’s an outlier known as “The Stupor Bowl,” while X & XIII were classics won by teams known for the type of big-game performances that can trump statistical strategies that need time for the math to work.
An immigrant statistician at IBM who knew nothing of football helped convert 260 scouting phrases (“he can run as fast as 2 cats“) into computer-friendly variables, input those into a giant box with less power than today’s laptops, and hired a psychologist to design a questionnaire sent to 10 scouts at 400 schools. As a result, the Cowboys were able to uncover future Hall of Famers like Bob Hayes and Rayfield Wright at obscure colleges and fill roster spots with other athletes (e.g. basketball players & track stars) who’d never played one snap of college football..
Big data may help make accurate predictions or guide knotty optimization choices or help avoid common biases, but it doesn’t control events and can be undone by cluster luck. Models are useful in predicting things we cannot control, but for players in the midst of a game the reality is different. Players don’t predict performance; they have to achieve it.
For more on this subject, please see Untangling skill and luck in sports and business, The greatest comeback ever and the limits of decision models, and March Madness and the availability heuristic.
This short video from ESPN films tells the Cowboys’ story. Inspired by what GM Tex Schramm saw at the 1960 Winter Olympics – IBM had placed chips inside skis to collect data – they created the blueprint for the modern NFL out of thin air.
In honor of the playoffs, and invoking The Rule of 3… here are two more recent examples of how data is changing the NFL:
PART II – The Comeback of the Running Back argues that increased snap counts are favoring big-play running backs, especially as the game grinds on.
No running back was drafted in the first round of the last two NFL drafts. But general managers, coaches and scouts knew that the running game would return due to a mixture of economics and fitness. And now it has.
This season, running backs carved out a new role that looks like it will last: big-play specialists designed to exploit holes in today’s new-age defenses. Running backs may never be the focal point that they once were, but they are at least putting up a fight.
As the passing game gained prominence in recent years, two things happened: Linebackers and defensive backs shed weight to adjust to the new speed of the game, and up-tempo offenses resulted in more plays and more tired defenses. Those no-huddle offenses not only wore down defenses over the course of a game, but also the course of the season. This has led to exhausted defenses with lighter players who see hundreds more snaps than in the past…
To be sure, winning teams tend to run the ball more, especially late in games as they protect leads, leading to some statistical inflation. But NFL strategy wonks say that this development is much more than that. Worn-out defenses, which are using more defensive backs to defend the pass, are having a tough time bringing down lumbering running backs…
“People were focusing on the edges of the field, defenses were focusing on playing the pass and playing ‘basketball on grass,’ and what’s happening is the offenses are going back to oversize players, getting a power forward to go into the paint with those guys,” said former NFL general manager Phil Savage…
This is a classic phenomenon in sports: an undervalued commodity becoming valuable again because of a shift in how the game is played.
PART III – The Worst QBs Over 40 (Passes) points to data that suggests winning % drops off substantially when a QB – even an elite one – exceeds 30-39 pass attempts in a game.
Tasking any quarterback with videogame style pass attempts however, may be a losing strategy. The Count looked at the career records of the quarterbacks expected to start in this season’s NFL playoffs (except Arizona’s Ryan Lindley, who only has six career starts) and found that only New England’s Tom Brady (39-24, .619) and Indianapolis’ Andrew Luck (11-11, .500) have records of at least .500 when attempting 40 or more throws.