The New York Times-20080128-Rockets Gazing at the N-B-A- Through a -Moneyball- Prism

Return to: The_New_York_Times-20080128

Rockets Gazing at the N.B.A. Through a 'Moneyball' Prism

Full Text (1793  words)

Daryl Morey has charts and spreadsheets and clever formulas for evaluating basketball players, and a degree from M.I.T. to make sense of it all. He would explain this mathematical wizardry, except that every few minutes his train of logical thought is interrupted by impulsive exclamations.

Go, Rafer! All the way -- nice!

Oh, jeez. After getting up this lead, we lost our minds!

Here we go, Luis! Here we go! Here we go!

Morey, the rookie general manager of the Houston Rockets, is a wizard in the field of quantitative analysis, a friend of Billy Beane's and a Moneyball true believer. He is the N.B.A.'s highest-ranking stat savant, the first mathematical magician to run a team.

Numbers are clearly very dear to Morey. Yet as he watched a recent game from a midlevel seat at Madison Square Garden, his refined logic was frequently overwhelmed by raw emotion. Morey became agitated when the Rockets took a bad shot. He bounced in his seat when they scored on the fast break.

In other words, he reacted like every other general manager to the ebbs and flows of the game. Morey's brain, it would seem, is not a web of wires and circuits. When he views the court, his field of vision is not overlayed by numbers and blinking lights, a la the Terminator.

Morey, 35, is not in fact an android seeking to reduce basketball to stats and logic, although his manner of speaking might sometimes suggest it.

Humans make much better decisions when they're integrating information, Morey said, momentarily sounding like an alien scientist.

Other times, he comes across as an insurance adjuster, explaining that his work is really about how to better understand and manage the risk. The statistical models are a tool, like any other, to find the right players, to pay them the right salary and, ultimately, to win games.

Quantitative analysis, or analytics, was popularized in baseball by the author Bill James and championed by Beane, the general manager of the Oakland Athletics. The practice has been creeping into basketball front offices for several years, although with far less attention or demonstrable results.

Morey estimates that about two-thirds of N.B.A. teams employ at least one expert in analytics (some on a part-time basis), but among his general manager peers, Morey is indeed something of an alien.

Of the other 29 individuals who run basketball operations (as team presidents or G.M.'s) all are former players, coaches or scouts. Morey was, by his own description, an unremarkable basketball player in high school, though he once led his intramural team to the title at Northwestern.

In my head I'm a good player, but I think the data would show that I wasn't very good, Morey quipped.

His professional resume is firmly rooted in the business world. He was a consultant for the Parthenon Group, which helped coordinate the sales of the Celtics and the Red Sox. He worked for a defense industry advisory firm. He was a statistical consultant to Stats LLC. He holds a bachelor's degree in computer science and an M.B.A. from M.I.T.

Morey looks and sounds nothing like a basketball person, but he is a rabid sports fan. (He and his wife spent their honeymoon at the 1996 Olympics in Atlanta.)

He received his on-the-job training in basketball as a senior vice president for the Celtics, serving under Danny Ainge for three years as the team's head of analytics.

Despite the popularity of Moneyball -- the 2003 Michael Lewis book that chronicled Beane's revolutionary methods in baseball -- N.B.A. teams are still inclined to put basketball people in charge of basketball decisions. Leslie Alexander, the Rockets' owner, was convinced there was a better way. So he hired Morey to replace the longtime general manager Carroll Dawson, following the path taken by Sandy Alderson when he picked Beane for the A's.

I wanted a guy who's really, really smart, Alexander said. Daryl's really, really smart.

Of course, Alexander could have done what other teams have, and simply hired someone like Morey as a consultant or a special assistant under a traditional G.M.

I want the guy doing the work making the decisions, Alexander said. Otherwise, it's filtered.

For all of Beane's remarkable success with the payroll-strapped A's, the use of quantitative analysis remains in its infancy in basketball. Part of the reason is purely practical -- basketball is neither as statistically driven, nor as statistically rich, as baseball. Unlike in baseball, in which a hitter or pitcher's success is largely self-determined, basketball players are highly dependent on one another.

The difficult thing to factor in is basketball is more of a collaborative, chemistry sport than baseball, said Chris Wallace, the Memphis Grizzlies' general manager, who worked with Morey in Boston. Obviously, you need star power and great players. But a tremendous amount of your success is dictated by your interplay and synergy with the team -- continuity, chemistry, feel for each other, those type of intangibles, which are very difficult to quantify.

Even when a player is going one on one, Wallace said, he is still not entirely self-dependent. The abilities of your team may dictate how much team defense is directed at you, Wallace said.

Morey, who became the general manager last May, is fairly guarded about the models he uses -- understandable in a sport in which the science is still being developed. Basketball as a sport has not yet settled on more sophisticated ways for assessing a player, as baseball eventually did with statistics like on-base percentage and slugging percentage.

But the search is on for something more meaningful than points per game. Morey and others in his field are filling hard drives with information, categorizing player types, shot selection, substitution patterns, the result of every pass and the success rate of the transition 3-point attempt (which, statistically speaking, happens to be a good shot).

We track everything imaginable, Morey said. Each pick-and-roll, what's the result of it? Each guy on the floor, how efficient they are. A lot of it, we end up not using. But we track it so that we have it available in case the question comes up where it becomes relevant.

Whatever revelations Morey has found for assessing players, they remain proprietary for now. But at the team level, he said, there are four statistics that are now widely accepted as indicative of a team's success rate: effective field-goal percentage (a combination of 2-point and 3-point percentages), rebounding and turnover rates (which determine how many more possessions a team gets), and free-throw edge (in attempts, not percentage).

The Rockets presumably have a small army of people crunching this data. Morey will only say that the team has made a significant investment of people and millions of dollars.

He is equally secretive about his methodology. But Morey said he could determine, for example, which players created the best scoring chances for their teammates. According to his work, the Rockets' Tracy McGrady (in the last season they tracked him) led the league in passes that led to high-percentage shots.

A traditional eyes-and-instincts scout might say that such a thing is self-evident, with or without the use of a high-tech abacus. Morey has no problem with that.

The data information is often just teasing out what the traditional folks have always known and believed, he said. But maybe it gives you a better sense of the risk around some of those decisions.

Risk is perhaps an underused word in the N.B.A. With small roster sizes, a salary cap and a luxury tax (which serves as a virtual ceiling on payrolls), franchises can ill afford to make a mistake when signing a player. Just as important as assessing a player's skills is assessing his value -- or what do they produce per dollar, Morey said.

In baseball, Beane has become a master of winning on a small budget. In the N.B.A., the San Antonio Spurs have successfully navigated the salary cap and maintained a relatively low payroll while winning four championships since 1999. At the other end of the spectrum are the Knicks, who appear to be headed for a seventh consecutive losing season despite annually having one of the league's highest payrolls. (The Knicks, incidentally, do not employ quantitative analysts.)

If you make a mistake in that area, it kills you for a chance to win championships, Alexander said, referring to player contracts. I wanted to have a system that was the most efficient and make the least amount of mistakes.

Analytics played a part in the Rockets' trade for small forward Shane Battier in July 2006 (when Morey was serving an apprenticeship under Dawson). Houston gave up Stromile Swift and the draft rights to Rudy Gay, a lottery pick, for Battier, a player with modest career averages: 10.3 points and 4.7 rebounds. But he is an accomplished defender, a solid 3-point shooter, a willing passer and a selfless teammate. His salary, $5.9 million this season, is slightly above the league average. He is the sort of glue player that is essential to a team with two superstars (McGrady and Yao Ming) and championship aspirations.

A savvy scout would recognize Battier's intangibles. Statistical analysis lent credence to those observations. Sometimes, the data and the observations come into conflict, Morey said. But more often, the information complements and better informs the team's personnel decisions.

In Boston, Wallace said Morey never promoted analytics as the holy grail but worked with a collaborative spirit that put everyone at ease.

He's the perfect individual to meld the traditional scouting, the eye-balling as you may say, with the more high-tech version of quantitative analysis, Wallace said.

That fusion was used in every acquisition the Rockets made last year, from Aaron Brooks (the 26th pick in the draft) to Luis Scola (obtained in a trade with the Spurs).

Alexander insisted that placing his team in the hands of a business-school graduate, rather than an N.B.A. veteran, was not a risky decision. Indeed, he saw it as a move to minimize risk. How will Alexander know he was right? How will Morey know his methods make practical sense? How will other teams know whether to make the leap?

With a very old-school, easy-to-grasp bottom line.

I do think at the end of the day how people judge it will come down to whether or not we win titles, Morey said. I think that's appropriate.

[Illustration]PHOTOS: Daryl Morey, in his first season as Houston's general manger, uses quantitative analysis when making basketball decisions. (PHOTOGRAPH BY BARTON SILVERMAN/THE NEW YORK TIMES); Houston decided to trade for Luis Scola, above center, and draft Aaron Brooks, left, last year based on a combination of analytics and old-fashioned scouting. (PHOTOGRAPH BY EUROPEAN PRESSPHOTO AGENCY; DAVID J. PHILLIP/ASSOCIATED PRESS)