The LEAD item in the Times Week in Review, no less ... intriguing for the visibility this analytical revolution is (finally) getting.

As a Baseball Prospectus subscriber, you can guess my views. (The "playing hunches and ignoring data" habits die hard with people who blindly accept received wisdom... conservatives?)


To Play Is the Thing
By DAVID LEONHARDT

Baseball has always been the most literary of sports, but it never managed to produce an intellectual fight worthy of the term. For most of its existence, writings on the game tended toward the poetic, like John Updike's "Hub Fans Bid Kid Adieu," a farewell to Ted Williams, or toward statistical minutiae.

This summer, however, the sport has found itself in the equivalent of a theological dispute about whether baseball is a game of mystery or of data, of statistics and analysis or of intuition and human instinct.

Like any good intellectual spat, this one involves high-brow questions and low-brow insults - in this case, dumb, narrow-minded and even unloving. It also has attracted interest from fields as far from the dugout as medicine, Hollywood and Wall Street, which find themselves grappling with the same question as baseball managers: When information can be gathered more cheaply and quickly than ever before, should people rely less on their hunches and more on numbers?

"I've been sat down and told they can give me a better way to do everything," Tony La Russa, manager of the St. Louis Cardinals and the hero of a new book celebrating the hunch, said last week, describing the statistics crowd. "They really are convinced that they can sit there and crunch out a formula that negates my power of observation."

"It's been a little irritating," La Russa added, "because there's a certain arrogance with that whole group."

It began two years ago, with the publication of "Moneyball," the Michael Lewis best seller about the Oakland A's, whose general manager, Billy Beane, used quantitative tools to keep his team near the top of its division every year, despite having less to spend than many competitors.

All the while, Beane marveled at the inanity of baseball's old ways, like judging prospects by body type instead of performance. Beane's success, and that of Lewis's book, brought even more number-crunchers into front offices, often at the suggestion of a team owner who had read "Moneyball."

Last year's World Series victory by the statistics-centric Boston Red Sox set baseball's old guard even farther back on its heels. The lessons of their championship will be enumerated next month with the publication of "Mind Game: How the Boston Red Sox Got Smart and Finally Won a World Series," written by the staff of Baseball Prospectus, a Web site that is to baseball's reformation what The Public Interest was to the rise of conservatism.

The traditionalists, who still dominate the scouting ranks, many front offices and the baseball media, have mounted a counterreformation this year with two books of their own. The first, "Three Nights in August," by Buzz Bissinger portrays La Russa as a master at tinkering with players' psyches.

He tries to bring out the best from an underachieving player, and he decides which pitchers should be briefed about the opposing lineup before a start, and which should simply go out and throw.

The second traditionalist text, "Scout's Honor," by Bill Shanks celebrates the scouts of Atlanta Braves, a profession that often serves as Beane's foil in "Moneyball." The Braves have won 13 straight division titles, Shanks writes, by letting their scouts find the players with the best "makeup," a baseball catch-all for hustle, attitude and heart.

Shanks is openly contemptuous of the Lewis book, writing, "the brash disregard for scouting in its truest sense as portrayed in 'Moneyball' was just as insulting to me as it was to so many scouts around the game."

Academic research, however, is pretty much on the side of statistics. Whether diagnosing patients or evaluating job candidates, human beings vastly overestimate their ability to make judgments, research shows. Numbers and analysis almost always make people better.

"There have been hundreds of papers on subjects from picking students for a school to predicting the survival of cancer patients," said Richard Thaler, a University of Chicago economist who uses sports examples in his class on decision-making. When a computer model is given the same information as an expert, the model almost always comes out on top, Thaler said.

Baseball's new analysts say that teams rely too much on instinct and received wisdom, which leads to things like the overuse of the sacrifice bunt and the drafting of high-school players.

In a speech to a group of investment bankers shortly after "Moneyball" appeared, Paul DePodesta, then Beane's deputy, called baseball a game where you were supposed to sit on your behind, "spit tobacco and nod at stupid things," borrowing a remark from a retired pitcher named Bill Lee.

"It became clear to us that the inefficiency in decision making in baseball was vast," said DePodesta, who played baseball at Harvard and led the Los Angeles Dodgers to the playoffs last year, his first season as their general manager.

The early record suggests that the reformers have found a real edge, if not a fool-proof method. The small-budget teams that depend most on analysis - Oakland, Toronto, Cleveland - are among the only ones in the playoff picture this year.

But their record is hardly spotless, the old guard happily notes. Beane has never won a playoff series, and DePodesta's Dodgers are struggling this year. The Braves are in first place again, despite Baseball Prospectus's many predictions of their imminent demise. So are La Russa's Cardinals

The most entertaining part of the battle is the charges and countercharges. Bissinger, for example, writes that the number crunchers do not truly love the game because they do not appreciate its lore or its human ingredient, a claim Lewis called absurd.

Indeed, what makes this fight truly comparable to those that periodically roil the worlds of art history or foreign policy is that the differences between the sides aren't as great as the sniping between them suggests.

La Russa spends much of his time jotting down information on index cards and studying statistics in his office, while members of the new guard often say the future belongs to teams that combine number crunching with scouting and injury prevention.

"The 'Moneyball' kind of stuff has its place, but so does the human," La Russa said by telephone from Pittsburgh. "Really, the combination is the answer."

But reaching that happy medium is likely to prove more difficult, and more interesting, than talking about it. The Cardinals, after all, created a statistical analysis department in the last two years, but La Russa said it had "almost zero effect" on his strategy. He wishes the team had instead spent the money on new video equipment.


Copyright 2005 The New York Times Company

"As a Baseball Prospectus subscriber, you can guess my views. (The "playing hunches and ignoring data" habits die hard with people who blindly accept received wisdom... conservatives?)"


I respectfully disagree. I think that the human mind is a much more powerful computer than any computer we've built. Just look at what Jack McKeon did with the Marlins in '03.

Those tobaccos chewing folks have a lot of wisdom. I'd love to see an old-syle team beat a computer built team in the world series any day of the week. Down with the borg!

Jay

One thing that's always bothered me about the "Moneyball" people is they talk about the theory, but I don't hear a lot about the players who fit the mould. Example, the above article doesn't name any players that statistics would tell you are good.

A couple of named players I've heard, don't really stand out.

For example, when the A's acquired Erubiel Durazo from Arizona, I heard about how he was a great "Moneyball" player. That is, he was not a big name star, but he had some statistical qualities that made him valuable. In 4 years in Arizona Durazo never had more than 222 ABs. In his first 2 years in Oakland he had 537 and 511 ABs and drove in 77 and 88 RBI with 21 and 22 HR and 105 and 104 SO and a .259 and .321 BA. Nothing there really jumps out at me.

Kevin Youkilis of the Red Sox is supposedly highly touted. In 2 seasons he is a career .266 BA, 8 HR, 44 RBI and 59 SO in 274 ABs. Nothing really stands out there, either.

Also, why has Billy Beane let go so many of his star players (Hudson, Mulder, Tejada, Giambi, etc.) Wouldn't statistics tell you to keep your good players (Hudson and Mulder both being 20 game winners, Giambi and Tejada both being league MVPs)?

To me, it seems that the stats people seem better at finding players that offer good value for the money, not necessarily the best players.

A superstar player or even 2 superstar players does not make a team. (see A-Rod in TX, Griffy in Cincy) Of course if the A's had the payroll the size of Boston, we would've kept the players you mentioned above, but we don't. Therefore, we're forced into finding undervalued talent across the league and through the draft to supplement the stars we lose ea yr. One way is by using statistical analysis, which is highlighted in the book, MoneyBall.

Although, Durazo's size led some to believe he's a HR hitter, that wasn't the reason we got him. The case for Durazo was his high OBP (on-base %). His discipline at the plate was the key. He rarely struck out and he walked a lot. His OBP was .396 in `04, 7th in the league. Eric Chavez was 5th. You can't score runs if you don't get on base.

The A's lead the league in fewest strikeouts and are among the league leaders in on-base percentage. They are getting on base and scoring runs.

The best players have been replaced by players who can also get the job done:

OLD.................YOUNG

SS, Tejada........Bobby Crosby (2004 ROY & cute)

1B, Giambi........Dan Johnson (ROY Candidate; good plate discipline w/ surprising power)

SP, Mulder........Danny Haren (#3 pitcher in rotation & injury-free)

SP, Hudson........Joe Blanton (rookie, 2nd best ERA behind Johan Santana in August; also won Rookie of the Month in June.)

The new players might lack star appeal, but they are quietly getting the job done as evident by their 74-56 record. And they will only get better.

Man, I could write 20,000 words here, or just direct y'all to Baseball Prospectus!

Allen is on the money... Canmark, it's significant that with those two specific players, the numbers you cite (except for BA) are what the performance-analysis types -- let's call em seamheads -- refer to as COUNTING STATS, which are generally not as significant per an individual's ability as RATE STATS -- on-base pct, slugging; strikeout/walk ratios and HR/IP for pitchers. (otoh, RBI, runs scored, and a pitcher's wins, for example, depend heavily on the individual's team.)

In Oakland's particular case, you're absolutely right -- since Beane's payroll CANNOT allow him to keep many stars, chasing undervalued talent is his game. Wouldn't you say it's worked, since the A's keep contending? Harden, Blanton and Haren seem to have replaced Hudson and Mulder very nicely.

Jay, I'd point to the Times article above; data-based analysis is always a more reliable methodology than human memory, which deceives constantly (as we find especially as we age). McKeon did a fine job, but I doubt you want to replace your computer with his brain ... and he had lots of help from Marlins staff, some of whom use data analysis.

Someone on another bulletin board remarked that Moneyball has much in common with the Bible, as that it's a widely referenced book that many of its outspoken critics have either never read or utterly misinterpreted.

[ August 30, 2005, 07:32 AM: Message edited by: Bill W ]

But isn't that my point... that Moneyball finds you the best value players, not necessarily the best players.

And it wasn't until Beane disciple Theo Epstein broke the bank and bought the second highest payroll team in the AL that the Red Sox finally won the World Series.

Meanwhile, Oakland's last World Series came when they had the big-name talent: Ricky Henderson, Jose Canseco, Mark McGwire... ie. guys with gaudy stats (HRs, RBI, SBs).

And also great OBPs (Rickey being the premier leadoff hitter of his day).

Billy Beane famously said "the [outcome of short series in the] postseason is f*cking luck." There are other ways to put it, but the most a GM can expect in designing a team is to get to the playoffs.

Obv not every team has to play 'moneyball' since they don't all have the same payrolls (and Beane would love to have Epstein's), but they can play smart ball, which is the subject of the forthcoming Mind Game book on the Red Sox cited in the Times column above. I'd recommend it, the Prospectus annual, or many of the sabermetric websites for getting an in-depth appreciation of such analysis.

Well, well! With pitchers' won-loss totals being thrown out the window by the BBWAA in (properly, imho) awarding the Cy Young Awards last week, dinosaur sportswriters are tearing their hair out. It appears that careful statistical analysis is expanding its presence in the media, and will continue to.



http://www.nytimes.com/2009/11/22/sports/b...l/22awards.html


I wonder how long Joe In Philly will continue to cling?

"My favorite (stat), besides facing individual batters, is FIP (Fielding Independent Pitching), which is kind of like walks to strikeouts and home runs given up. So I try to get ahead of the count without leaving it run down the middle in a person's power zone, get ahead in the count. That helps me not walk guys, and then, when I get two strikes, I try to strike guys out. And that's how I try to pitch, to keep my FIP as low as possible."
—Zach Greinke