The 10th Commandment

All statistics sometimes lie.

On February 27, 2009, the grandson of a Nebraskan grocer tried to explain in a letter to his investors how his companies and investments lost $11.5 billion in 2008 when his companies and investments had increased 20.3%, compounded annually, over the previous 44 years without ever losing money in any one year.

"Investors should be skeptical of history-based models," überinvestor Warren Buffet warned. "Constructed by a nerdy-sounding priesthood using esoteric terms such as beta, gamma, sigma and the like, these models tend to look impressive. Too often, though, investors forget to examine the assumptions behind the symbols. Our advice: Beware of geeks bearing formulas."

Sound advice.

In deciding how much weight, if any, you should give to QC's statistics (or any other statistics), QC strongly encourages you to consider a few things. First, QC's statistics are not simply a bunch of numbers randomly pulled out of a hat. QC's statistics are derived from a theory of competition (Growth Theory (a/k/a Endogenous Technological Change). In considering any sports statistics or quantification formula, a starting point is to ask what, if any, theory of competition do the numbers reflect. To paraphrase the vikings in an old credit card advertising campaign, "What's in your theory?"

Second, QC openly admits that, like the history-based models of the nerdy-sounding priesthood, QC's statistics are history-based models. Thus, QC's statistics are an abstract description of reality and should not be confused with reality itself. Like the formulas of other geeks, QC's statistics may fail and, from time-to-time, will fail.

For example, QC has already identified two scenarios where QC's statistics may fail. The first group, "negative black swans", is when a more productive team turns the ball over more often than its less-productive opponent (See 9th Commandment). The second is what QC calls the "Delhomme Exception." The Delhomme Exception occurs when a team's superior statistical productivity is derived from a single, disproportionately productive play, such as occurred in the 2004 Super Bowl when the Carolina Panthers lost to the New England Patriots, 32-29, despite superior statistial productivity and no negative turnover margin. However, Carolina's productivity (9.788 YPA) derived from an 85-yard TD pass from Jake Delhomme to wide receiver Muhsin Muhammad, the longest gain from scrimmage in Super Bowl history. Other than that one play, a "positive black swan", the Panthers' productivity was just about the same as the Patriots' productivity.

However, the Delhomme Exception actually is consistent with QC's application of Growth Theory to football. Returning to the kitchen analogy, suppose that in the regular course of running his restaurant, Chef Walsh goes to the Par-Sells Grocery Store every week to buy the ingredients he needs for his signature West Coast Bull Steak. With the assistance of an extremely faithful stockboy, Kurt Warner, Chef Walsh purchases his key ingredient, bull meat, which the Par-Sells Store imports from Lombardi Brothers Meats in Sheepshead Bay, New York. At his restaurant, Chef Walsh's staff provides graceful service to the patrons. By repeating this process on a nightly basis, Chef Walsh's restaurant business thrives and makes $1 million in profit for the year.

On December 31, Chef Walsh enters the Par-Sells Grocery Store and follows his standard routine. But as he checks out, an alarm sounds and balloons appear. The owner, William Parsells, steps forward and says, "Congratulations Chef Walsh! You are the Par-Sells Grocery Store's 1 millionth customer. You have won $1 million!" Chef Walsh stammers, "Gee, that's great. Thank you Mr. Parsells." As Chef Walsh leaves, the stockboy, Kurt Warner, says to Mr. Parcells, "Boy, that Chef Walsh is really great! Have you ever had his West Coast Bull Steak?" Mr. Parcells answers, "Never. That recipe is bull something, but it isn't steak. You have to respect a man who shops for his own groceries, but I don't go in for that fancy-pants stuff. Our tuna is better than anything you can get from Lombardi Brothers. And it's better for ya!"

At the end of the year, Chef Walsh has $1 million from creating, making, and serving his signature West Coast Bull Steak to his customers. He also has $ 1million from winning the Par-Sells Grocery Store contest. But clearly, the second scenario does not reflect the true productivity of his restaurant over the course of the year.

Concededly, Delhomme was not just "lucky" in the same way that Chef Walsh was "just lucky" when he won the Par-Sells Grocery Store 1-millionth shopper contest. After all, Delhomme's touchdown pass to Muhammad involved a great deal of skill. If you watch a re-play of the pass, you can see that, before he throws to Muhammad, Delhomme looks at another receiver for just a split-second, which fools the New England defense. Still, Delhomme's touchdown pass was a "lucky" play in that the longest play in Super Bowl history is by definition a very rare, probably singular, event. Assuming that on average a Super Bowl contains approximately 125 plays from scrimmage, there have been approximately 5,375 plays from scrimmage in the Super Bowl. Thus, the odds of any play resulting in the longest gain from scrimmage is 1-in-5,375 or .0186% chance.

Other black swans may include: (1) kickoff, punt, interception, and fumble returns for touchdowns; (2) blocked and missed field goals and PATs; (3) blocked punts; (4) exceptionally long successful field goals; (5) extraordinary punting; (6) a single trick play; (7) on-side kick recoveries; (8) critical penalties (or incorrect officals calls); (9) unusually long TD runs; (10) weather; (11) overtime coin flips; and (12) other atypical acts or events that a coach cannot replicate through his play design.

QC has attempted to disclose and discuss in detail in these so-called 10 Commandments the assumptions behind QC's statistics and to illustrate the assumptions through narrative. Only through such effort might (might) statistics achieve the powers of language. And it is only when statistics achieve the powers of language that statistics have any value.

James, the high priest of sports' nerdy-sounding priesthood, wrote: "I didn't care about the statistics in anything else. I didn't, and don't pay attention to statistics on the stock market, the weather, the crime rate, the gross national product, the circulation of magazines, the ebb and flow of literacy among football fans and how many people are going to starve to death before the year 2050 if I don't start adopting them for $3.69 a month; just baseball. Now why is that? It is because baseball statistics, unlike statistics in any other area, have acquired the powers of language."

© 2009 QuantCoach.com
QuantCoach.com is not affiliated with the National Football League or any of its members.
The material contained on QuantCoach.com is for entertainment purposes only.