Bissinger Declares Moneyball a Failure, Goes Home
by MNPundit on Oct 18, 2009 1:57 PM EDT
39 comments
Against 'Moneyball' Why the richest teams in the game--and not the lowly A's--are dominating the postseason once again.
That's the title of an article in TNR by Buzz Bussinger, author of Friday Night Lights and a noted blogger hater.
Now before going off the handle on me (or him) he makes some very good points. Namely that Beane, Riccardi and Podesta have had a lot of difficulty making it to the playoffs. That the use of statistics to value different players now just makes the big money teams sign them and makes more players prohibitively expensive--so now the old "5-tool" players and the high OBP or OPS guys are now just as out of reach as before. That it hasn't really improved the prospects of low-money teams to actually win it all, not in the long run.Whatever happens in the National League and American League Championship series unfolding over the next week or so, one outcome has already been decided--the effective end of the theories of Moneyball as a viable way to build a playoff-caliber baseball team when you don't have the money. That no doubt sounds like heresy to the millions who embraced Michael Lewis's 2003 book, but all you need to do is keep in mind one number this postseason: 528,620,438. That's the amount of money in payroll spent this season by the teams still in it--the New York Yankees, the Los Angeles Angels, the Philadelphia Phillies, and the Los Angeles Dodgers. Moneyball? You bet it's Moneyball, true Moneyball, like it always has been in baseball and always will be.
It's an interesting article that I thought was going to be full of fail at first, that turned out to be a little better. As someone who hates Michael Lewis for the tone of Moneyball, I found it interesting.
The New Republic is generally centrist political magazine.
0 recs |
39 comments
Email
Print
Comments
Reflecting on Beane
Not only have the Moneyball tactics been co-opted, but small market Oakland has seemed to trade players for the sake of trading. Consider this Beane sequence—
Mulder for Haren+ for Carlos Gonzales + for (with Huston Street) Matt Holliday for Brett Wallace. Or Andre Ethier for Milton Bradley (Hopefully Ladendorf isn’t the next Ethier.)
Meanwhile, about the only big contract they have doled out was to Eric Chavez. The “Genius,” Billy Beane makes Bill Smith look like, well, a super-duper mega Genius.
by Han Joelo on Oct 18, 2009 5:03 PM EDT reply actions 0 recs
Backlash is understandable but too strong...
I agree the major problem with Moneyball is not the content, but the tone. Lewis’s prose is really smug. I didn’t read it until just recently and I felt like I needed to take a shower after every chapter.
I certainly wouldn’t be down on Beane, though. He did win on a budget for quite a while (1999-2006) hemorrhaging MVP’s and CYA’s along the way. That was no small feat.
The rich teams did catch on to his tricks. Its the Yankees and Red Sox that near the top of the league in walks now… with Oakland in the bottom half because GM’s don’t throw away low-BA/high-BB guys like they used to. The Red Sox hired Bill James. OPS is quoted on Baseball tonight.
Yeah, the A’s have stunk for three years now, but that doesn’t change what he did before. Often, that’s all a great GM gets —a great 5-10 year run before the rest of the league catches on.
And yes, Beane wasn’t the only great GM of his time. The Braves GM in the 90s was also bleeding talent and kept reloading. Terry Ryan had a knack for spotting talent in the low minors.
by DavidRF on Oct 18, 2009 8:42 PM EDT reply actions 0 recs
Problem
Once again, the point of Moneyball isn’t the statistical analysis, it’s being ahead of the market curve. You’re right, OB and such isn’t ahead of the curve anymore. The next big thing is probably the business style analysis explored in Diamond Dollars.
"You can't sit on a lead and run a few plays into the line and just kill the clock. You've got to throw the ball over the damn plate and give the other man his chance. That's why baseball is the greatest game of them all."
~ Earl Weaver
"In God we trust. All others must provide evidence."
~ Billy Beane
by AdamOnFirst on Oct 19, 2009 4:49 AM EDT reply actions 0 recs
The only thing left is defense
Moneyball was based on the premise that OBP is under valued. That’s no longer true. The next thing was to look at FIP instead of ERA. There are a few hold outs on that one, but just about every team looks at FIP now. The next thing is defense. I think UZR is valued somewhat, but a lot of teams still rely on scouting and reputation when evaluating defense. So a moneyball team could get a bargain who’s really good UZR but has a relatively low OBP. Say, Carlos Gomez.
Still, if moneyball isn’t dead, it’s on death’s door. Once every team pays attention to the same statistics, it comes down to who has the most money and who augments the stats with the best scouting. Same as it ever was.
I agree with the poster who pointed out that Beane has made a hash of things. Teams need stability. I understand trading a guy just before a big payout. But he trades guys in their second year of arbitration. There’s just too much churn with that team.
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
by cmathewson on Oct 19, 2009 9:11 AM EDT reply actions 0 recs
I think its a myth
that teams are a bunch of cigar chomping old timers who don’t know high school math. There were and still are proprietary defensive metrics used by teams. UZR is sort of a mass market measure for amateurs. It also fails a lot of the time. Modern teams have access to an enormous amount of data including video. If you think they rely on reputation, I think you’re buying a storyline. If the Twins or any other team wants to sign player A or B, I wouldn’t be surprised if they would sit down and watch a video that had most or all of every play the guy made over the last year.
FIP is flawed as well. Teams use a whole host of measures and they’d be insane not to include scouting. Some guys do more with less, but you stil want to know the guys who have more. Some guys produce bad stats for years and then become really good.
by Salty on Oct 19, 2009 11:43 AM EDT up reply actions 0 recs
The point is that evaluation of defense is still a grey area... scouting-wise or stat-wise
Measuring offense is a much easier problem. If some organization somehow figured out how to better evaluate defense then they could get a competitive edge on other teams by drafting better and making lopsided trades. None of that is new.
The whole cigar-chompers versus computer nerds rhetoric is quite stale these days. It may have been in vogue back in the 1980s, but seems a bit old-fashioned itself now. Bill James is in his 60s now, Pete Palmer is over 70. There was always a contradiction in this comparison anyways. “Traditional” scouts of any era could cite reams of stats off the top of their heads and were always looking for an edge on the competition whilst most “statheads” also talk about work-ethic, make-up and one of their biggest advances in the past few years is video analysis and batting/swing mechanics.
by DavidRF on Oct 19, 2009 1:17 PM EDT up reply actions 0 recs
Gold Gloves
Maybe upper management is no longer focused on reputation when it comes to defense, but the on-field staff still is. Gold Gloves are voted on by league managers. Derek Jeter, who every statistical measure agrees has been terrible defensively, has three of them. Rafael Palmeiro won one for a season in which he played 28 games in the field. Granted, the field managers are not the ones acquiring players, but they’re the ones who make playing time decisions with the players they have.
"There are only two things that are infinite, the universe and human stupidity, and I'm not sure about the former." - Albert Einstein
by BeefMaster on Oct 20, 2009 3:33 PM EDT up reply actions 0 recs
you know the on field people
have access to all the data plus the independent data the teams produce. They also watch games and not on television. The Palmiero thing is getting old. I forget what the logical fallacy is called when you use one outlier example as evidence of a conclusion but it really doesn’t prove much. What year was it anyway?
by Salty on Oct 20, 2009 10:24 PM EDT up reply actions 0 recs
How about Micheal Young?
The year before he’s moved to third?
by lookatthosetwins on Oct 20, 2009 11:06 PM EDT up reply actions 0 recs
I don't see much Rangers baseball
but read some Chris Kahrl or compare outcomes for holy people like Keith Law or Dave Cameron. You don’t have to go back 7 or 8 years.
by Salty on Oct 21, 2009 8:16 AM EDT up reply actions 0 recs
Having the data doesn't imply that they pay enough attention to it
Some organizations use the data as a sanity check only, and rely almost entirely on scouting. I would put the Twins in this class. Other organizations lead with the data and use scouting for things like upside, which cannot be easily quantified. I would put the Red Sox in this class. Unless it’s someone like Joe Mauer, for whom data and scouting universally agree, you simply cant listen to both all the time. The data and scouting simply don’t agree all the time. Then you have to make a choice. Organizational philosophies dictate what choice you make.
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
by cmathewson on Oct 21, 2009 12:42 AM EDT up reply actions 0 recs
Its a mistake to think that data leads to a single conclusion
With much larger issues outside of baseball in fields like medicine, debate over implications and conclusions of data by many real statisticians can go on for decades. They also work with much more data. Its not unusual for the prevailing wisdom for long periods to turn out to be wrong.
Its only on blogs that the notion lives that players can be evaluated conclusively on things like their BB/K ratios. I remember when Jack Zdurenciek was hired by the Mariners and all the “sabr” heads were all weepy because an old white scouting guy was picked. Then they found out that he had a numbers guy, but the numbers guy had a staff. Those people weren’t the only ones doing statistical analysis on the team either.
I remember before OOZ plays were included in “advanced” defensive metrics and Hardy’s numbers sucked. The haters all whined that he wasn’t very good to the point that it was part of interviews of the GM and coaching staff. They said they had their own internal numbers and that you had to actually watch people play which was derided. Then the different RZR/OOZ ratios were rejiggered and Hardy became one of the best defensive SSs in baseball.
by Salty on Oct 21, 2009 8:39 AM EDT up reply actions 0 recs
1999
Yes, I realize that things have changed in the last ten years, but it’s not like the Palmeiro debacle happened in the thirties or something. You say it “doesn’t prove much”; I say it proves that the people responsible for evaluating defense sometimes don’t know what the heck they’re doing.
Also, you didn’t address Jeter, who even knew that his defense was bad – he used statistical analysis to figure out where his weak points were in the offseason and did some hip-strengthening exercises to improve his movement to his weak side, and his defense has been much better this year, according to the stats.
The “they watch games in person and not on television” claim is true, but I don’t know that it helps – Ron Gardenhire sees the New York Yankees play seven times a year in person (well, ten this year). It’s not like he has the chance to pick up on any sort of trends. I’d argue that the fact that managers simply don’t see some teams very much is a large part of why Gold Glove voting is based so much on reputation.
"There are only two things that are infinite, the universe and human stupidity, and I'm not sure about the former." - Albert Einstein
by BeefMaster on Oct 21, 2009 9:50 AM EDT up reply actions 0 recs
The Jeter thing
is such a well-worn debate. I heard a different story on his hip and his low numbers. In fact I’ve never heard your version.
by Salty on Oct 21, 2009 1:58 PM EDT up reply actions 0 recs
Here you go...
Jeter article – I’d note that I just re-read it, and it doesn’t claim that he used statistical analysis. The basic point is the same, though – Jeter knew that he needed to work on his defense (particularly plays to his left), he took specific actions to improve it, and his improvement has shown up statistically as well, which I’d say lends some credence to the utility of the statistical analysis.
"There are only two things that are infinite, the universe and human stupidity, and I'm not sure about the former." - Albert Einstein
by BeefMaster on Oct 21, 2009 2:42 PM EDT up reply actions 0 recs
Stats like UZR actually require scouting information.
Someone has to be at each game charting the location of each batted ball. That’s a non-trivial task which you can’t do from TV. If you don’t like the math that’s performed to generate UZR data, the charts alone would be quite fascnating to look at.
And no one ever claimed that using statistical analysis means that you stop using your brain. Every stat will have its biases and outliers just like certain traditional scouting observations will have biases and outliers (there’s always been players who look better than they perform).
A good analyst will use multiple metrics and mix them with observation. If not every system (observational or statistical) is in agreement, well, you can learn a lot about a player by asking why that’s the case instead of just throwing away information for everybody.
by DavidRF on Oct 21, 2009 3:11 PM EDT up reply actions 0 recs
i wouldn't call charting hits "scouting"
Isn’t the guessing about zones and locations done from video (TV)?
I never argued that anyone stopped using their brain. The number of systemic biases in defensive metrics is long. And the reason that real statisticians consider factors like reliability and validity is because the value of a statistic can be very, very low. Data is sometimes “thrown away.”
The sabr crowd routinely derides observational data.
by Salty on Oct 22, 2009 11:36 AM EDT up reply actions 0 recs
This is the problem with UZR
It ultimately resolves to charting. You can’t remove the human element from defense assesments.
It’s not like there’s a RFID tag on every ball player’s shoes with a grid under the field that automatically charts hit trajectories and velocities relative to player positions. Absent this kind of a system, defense will always have a subjective element.
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
by cmathewson on Oct 22, 2009 12:29 PM EDT up reply actions 0 recs
I keep hearing about HitFX
Supposedly its close.
by Salty on Oct 23, 2009 12:38 AM EDT up reply actions 0 recs
Only partially true
The sabr crowd routinely derides observational data.
I’d say that this depends on circumstances. Ultimately, all baseball data is observational – someone has to see Joe Mauer hit a homer and record it – but that’s being more than a bit pedantic.
I think the main reason for increased acceptance of metrics using observational data (besides defensive stats, there’s also batted ball type data) is completeness. Before the last few years, that kind of observational data wasn’t publicly available in large quantities, and as a result generally tended to be anecdotal and/or suffer from sample size issues. I think the fact that there is now information on the location and “type” of every batted ball has led to a greater acceptance of it as usable data.
You and cmathewson are completely right, though, that the data is not completely objective (which probably is a more precise term than “observational”, now that I think about it). A ball the official scorer rules a hit is objectively a hit; a ball softly hit in the air over the second base bag may or may not be a line drive or pop fly and may or may not be in the second baseman or shortstop’s “zone”.
By the way, I also was under the impression that most hit charting was done via video, although I don’t see how that makes it any less valid (I would agree that defensive positioning is better observed in person unless the charter has access to video covering the entire field pre-pitch).
"There are only two things that are infinite, the universe and human stupidity, and I'm not sure about the former." - Albert Einstein
by BeefMaster on Oct 22, 2009 2:48 PM EDT up reply actions 0 recs
Video adds more potential systemic biases. Consider how different pitches look when the CF camera is straight on versus at a slight angle.
Usually when someone is derided for observational data, its a result of someone saying they saw a lot of player X and he looks bad. Sometimes that’s relevant when you don’t see a player much, but not always. I’ve seen about 25 to 30 Cards games. In about 15 to 20 I’ve seen Rasmus have a hard time fielding balls, particularly ones where he has to go back. He takes bad routes, gets bad jumps and even has balls clank off his glove. But he has a nice UZR. He may be pretty good, but speaking only for myself, I don’t buy him as a premiere defender.
by Salty on Oct 23, 2009 12:46 AM EDT up reply actions 0 recs
I'm pretty sure the charts have to be done at the game...
Most games don’t have that many cameras and it takes a while for the camera to pan to the defender. That’s always been the complaint about judging defense from TV… you miss the first step.
I’ve been following Sabermetrics for a very long time and none of the good sabermetrician writers claim they are anywhere close to solving the defensive problem. Its always been an issue… from Pete Palmer’s Fielding Runs to BP’s WARP. Lots of people are trying of course, and these guys are quite hard on each other, but the advance metrics require these charts which can’t be retroactively obtained for historical players and are often not available in the minors, either. Turns out pitching is a much easier problem and offense is practically solved.
Sounds like you’re a bit sensitive to the SABR-rattling from 10-15 years ago when they were still trying to gain acceptance. Most of the good ones are more level-headed these days all the mainstream media cites their work. Tom Tango actually collects and collates observational defensive scouting info… and one of the biggest trends amongst computer geeks these days is collecting pitcher-motion and batter-swing videos for prospects.
by DavidRF on Oct 22, 2009 2:50 PM EDT up reply actions 0 recs
The scoring was done looking at video
I don’t know if that’s changed, but I’d be surprised. I’ve seen pictures of the process. There was also a Brewers fan who had a blog who got a job with STATS or BIS and had some posts about it before it went to the great blog heaven.
I remember posts over at baseballfever forums where a bunch of guys, including Tango IIRC, talk about how the results aren’t even that accurate over a 3 year period and are more valuable in reviewing career performance.
I usually say nothing, but challenging the idea that defensive metrics are very good, especially UZR is usually an instant debate and, well, debates are tedious. UZR’s widespread acceptance is related more to easy availablity if you ask me.
by Salty on Oct 23, 2009 12:54 AM EDT up reply actions 0 recs
To quote Phillies GM Ruben Amaro:
"I do not buy numbers defensively. At all. But that other business? I don’t buy it a lick. I think defense is subjective."
by lookatthosetwins on Oct 21, 2009 3:16 AM EDT up reply actions 0 recs
Yeah, it seems most teams evaluate player value much more accurately than ten years ago. I think most teams are even trying to properly value defense, now. Any sort of WAR type evaluation factors defense and positional value. Although that is one area where there’s probably still some room for a team to gain an advantage through superior analysis.
So, with a “correction” of sorts in how teams evaluate, I think cmathewson has it right, and now it’s just a matter of money, scouting, and player development. Same as it ever was.
Nobody can predict what the next big thing will be, but I suspect it will probably be non-statistical. Some team will develop some biomechanical program that reduces pitcher injury, or personality profiling, or something like that…
by jianfu on Oct 19, 2009 9:56 AM EDT reply actions 0 recs
Cloning Joe Mauer will be the new market inefficiency
I think the Twins will give Mauer the biggest contract ever but in return get the rights to the first 10 Mauer clones.
by DJL44 on Oct 20, 2009 1:13 PM EDT up reply actions 0 recs
That's why Punto's contract seemed so high
cloning rights
by what_would_gil_thorp_do on Oct 20, 2009 1:32 PM EDT up reply actions 0 recs
Why do we want clone's of Punto?
He should have payed us for the rights to do that.
by lookatthosetwins on Oct 20, 2009 2:40 PM EDT up reply actions 0 recs
Bill Smith did it for Gardy
Gardy wants to field an all-Punto team
by what_would_gil_thorp_do on Oct 20, 2009 4:45 PM EDT up reply actions 0 recs
Are you telling me we're never going to escape Punto?
If they just keep cloning him…he’s just going to be here…forever.
I need to lie down.
"Baseball doesn't owe me a thing. I owe my whole life to baseball." -Kirby Puckett
by fischean on Oct 20, 2009 2:54 PM EDT up reply actions 0 recs
Isn't Punto a clone
I thought Punto was the clone of Denny Hocking, or Jeff Reboulet, or Al Newman or Ron Gardenhire himself.
by DJL44 on Oct 20, 2009 3:41 PM EDT up reply actions 0 recs
well played, sir
my floor is looking pretty dirty... BETTER GET OUT THE BROOM!!!
by natetheskate on Oct 20, 2009 4:31 PM EDT up reply actions 0 recs
That's a new one: Player breeding
Let’s find Mauer a good mate. Let’s see, Jenny Finch is taken…
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
by cmathewson on Oct 20, 2009 6:52 PM EDT up reply actions 0 recs
Cat Osterman is single.
Only 3 days younger than her.
by MNPundit on Oct 20, 2009 8:10 PM EDT up reply actions 0 recs
Ha.
We can then sign the fetus to a 27 year, 30 million dollar contract.
by lookatthosetwins on Oct 20, 2009 11:08 PM EDT up reply actions 0 recs
Fetus' rights
Hmm – the twins would have to jump on the pro-life bandwagon. They’ll need fetus’ to get rights as a person to enter into legal contracts, presumably with the authority of its parents. Seems politically risky.
Cloning is safer – I think you could argue that you own the clone, and that it isn’t an independent legal entity, but rather the result of research and development. Presumably, you could control it for life under as peperty, and possibly even keep any other teams from copying your experiment for 20 years under intellectual property law. Any lawyers around who can help out here?
by snolls on Oct 21, 2009 8:58 AM EDT up reply actions 0 recs
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2009 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
