In a must-read for insights into how upper managers really work, the New York Times today has a profile of Zeus, the all-knowing football program. The program seeks to apply rigorous statistical methodology to determine Gameplay decisions. Zeus allows specific team versus team scenarios. Coaches when asked essentially if their poor decisions should be replaced by better ones, were understandably reluctant to resign.
“In his model, you should go for it on your own 20-yard line because you’re going to convert 80 percent of them, but that 20 percent is going to lose the game,” said Bengals Coach Marvin Lewis, who has met with Bower and his partners. “To say you could just plug it into a computer and say you should go for it here, I don’t think you want to explain that to the owner on Monday.”
Lewis inadvertently reveals himself and fellow entrenched managers as entirely the problem. By and large, upper-managers are there to make cautious decisions and keep their careers and vesting going, so they don’t have anything to explain on Monday. Lewis, after all, is scheduled to make $3 million this year.
Jim Schwartz, defensive coordinator for the Tennessee Titans goes one better — vaguely hoping his “instincts” will still have a place in the coaching world. Instinct being of course your own algorithm to assess the likelihood of winning:
“It’s never going to take the place of instinct and due diligence,” said Jim Schwartz, the defensive coordinator for the Tennessee Titans, who said he liked the concept when he talked to Bower about it. “But every single person uses some sort of statistical analysis. This is just taking it to another level.”
Sports have defined rule sets whose decisions could be easily optimized. And still coaches don’t use it. How much harder is it to evaluate managers when there are infinite possible outcomes? “Betting the company” and taking risks are things for 2-and-20 fund managers; they can have spectacular flame outs time and again and still start new funds. Brand managers generally can not. Compensation plans have no means of evaluating a risk-adjusted return, which is a common benchmark institutions use to guide against gunslinging 2-and-20ers.
The worst team in the league, the Cardinals at 3-13 had the lowest (best) score of only .42 games lost for not making bad decisions, which is entirely consistent with what should be predicted: they had nothing to lose and were free to play more optimally. Zeus ranked all teams on their regular season decision-making. The highest (worst) score was Tampa Bay Bucs at 1.42 games lost to bad decisions. Full list here.
The ‘72 Dolphins can drink their regular season perfection champagne in Peace, with the Patriots coming in Zeus-adjusted at 15.33-.67 last year.
