"The mathematics
of basketball"
By BJ DEJARESCO

As basketball fans everywhere rabidly froth at the mouth in anticipation of a battle royale between the Los Angeles Lakers and the Boston Celtics, it brings to mind how much the game has changed since the heyday of the league; back when the Magic and Bird where dueling over titles in the 1980's. (To those too young to know, that's Magic Johnson and Larry Bird)

While the changes in how the NBA game is played are readily apparent; such as the increased international flavor of the league, or the shift to guard-centric offenses spurred on by the removal of the hand-check rule, the most important change seems to be unknown to many fans of Naismith's greatest invention.

Just like the world around us, the game of basketball is experiencing an information revolution.

Think back to the countless arguments you've had with your friends about basketball.

Despite their variety, these arguments all have one thing in common. They rely mostly on conventional wisdom. More points per game means better offensive player. More steals per game equals better defensive player. It seems so painfully obvious at times doesn't it? But if there's anything the current information age has proven, it's that conventional wisdom is not always entirely right, or is just plain wrong.

So is there a way to determine exactly how valuable a player is? Is there a way to measure his overall impact on a game, or team performance? A new breed of hoops analysts believe they have found a way to do so.

The answer, these new age basketball fanatics have found, lies in a method that economists and statisticians have been using for years: Regression Analysis.

Regression analysis in simple terms, is studying the dependence of one variable (dependent variable) on other variables (explanatory variables), and using this information to predict or estimate the average value of the dependent variable.

An simple "real-world" example would be government economists estimating how much people's personal spending (dependent variable) would change due to an implementation of a certain tax (explanatory variable) by studying and running regressions to see how similar taxes have affected it in previous years.

Basketball, just like economics, can draw on a treasure trove of data since sports statistics is something that is well documented and easily available on the internet. In theory, you could do the same thing with basketball data as you would with economic data. All one has to do, is develop a formula.

John Hollinger, an economics professor and columnist of espn.com, has created such a formula: the PER, or Player Efficiency Rating. While the actual formula is a quite complicated, Professor Hollinger sums it up (as only truly knowledgeable people can) in a few simple words: "The PER sums up all a player's positive accomplishments (made shots, steals), subtracts the negative accomplishments (turnovers, fouls) and returns a per-minute rating of a player's performance"

David Berri, a Stanford economics professor and author of the book "The Wages of Wins" does the same by measuring a player's productivity and how exactly players contribute to a win. He explains that the NBA, and people in general tends to over-inflate the value of scoring in terms of its contribution to winning the game, and tend to minimize the value of other things that contribute to a win. The key, according to Berri, is focusing on factors that are associated with getting possession of the ball (rebounds, steals) because these too have a great correlation to winning. It is through looking at the cumulative effects of these various statistics that Berri is able to estimate the amount of wins a player produces. This is what he calls the "win score".

Various organizations in the NBA have already employed a number of these "stat geeks" to help assist them in forecasting performance of draft picks, signing free agents, and even finding the most efficient starting line-up for the team. This recent trend is hardly surprising though, because the NBA is at its core, a business; A multimillion dollar business that commands massive influence around the world. The more precise one can be, the better.

The basketball purist in all of us might have an urge to resist and even dismiss these new ways of evaluating players. After all, much of the fun in basketball is watching the determination, leadership, hustle, and artistry that the great players exhibit right? We didn't need a fancy algorithm to tell us that by averaging the most points and steals, while leading the 76'ers to the second best record in the league, Allen Iverson was hands down, the Most Valuable Player of the league in 2001, did we? We all wanted it to be him. How could we not fall in love with the 6 foot nothing, paperweight going up against the heavyweights night in and night out, and dominating a game made for men twice his size?

But was Iverson really the major reason they were so successful? Or was it perhaps that he had teammates who could effectively cover for him on defense, and rebound the ball with ferocity everytime he missed on the multitude of shots that he took?

These are the questions that Berri and company are forcing us to ask, because as it is in real life, the Cinderella story is the one that makes money, despite its being a mere fantasy.

So, does this mean we should we forget looking for fairy tale stories entirely, and simply focus on numbers instead?

Surprising as it may sound, these statistical wizards will be the first to tell you that nothing can replace watching the games themselves. These new methods that Berri, Hollinger, and others have engineered are meant to enrich our understanding of the beautiful game of basketball, that's all.

Think back to when Jordan soared above the Lakers in the finals with the ball in his right hand, and with breathtaking gracefulness, laid it in with the left, signaling the end of an era, and ushering in a new one.

The numbers will forever say that moment was only worth 2 points.

But in your mind, it will forever be more than that.