For a nerd like me, this quantification of sports has been tremendous fun. Thanks to obsessive websites, even the casual fan now has access to statistical tools that would have boggled the mind of a GM 10 years ago. Sabermetrics has also transformed the act of being a spectator, so that watching a game is no longer just about cheering for our hometown team. The numbers have given us a whole new way to think about sports, elevating the conversation beyond disappointed groans, ecstatic high-fives, and subjective opinions.
But sabermetrics comes with an important drawback. Because it translates sports into a list of statistics, the tool can also lead coaches and executives to neglect those variables that can’t be quantified. They become so obsessed with the power of base runs that they undervalue the importance of not being an #######, or having playoff experience, or listening to the coach. Such variables are the sporting equivalent of a nice dashboard. They can’t be quantified, but they still count.
...Here’s my problem with sabermetrics — it’s a useful tool that feels like the answer. If we were smarter creatures, of course, we wouldn’t get seduced by the numbers. We’d remember that not everything that matters can be measured, and that success in sports (not to mention car shopping) is shaped by a long list of intangibles. In fact, we’d use the successes of sabermetrics to focus even more on what can’t be quantified, since our new statistical tools take care of the stats for us. We are finally free to think about how those front seats feel.
But that’s not what happens. Instead, coaches and fans use the numbers as an excuse to ignore everything else, which is why our obsession with sabermetrics can lead to such shortsighted personnel decisions.
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
1. BDCThis point would be more compelling if any examples were cited to support it. In fact, the only baseball fact discussed much in TFA is how a better grasp of the numbers would have kept the Giants from spending so much on Aaron Rowand.
Such as? What players who were complete ######## have been brought into a SABR-inclined organization because *advanced* metrics said they could help the team.
Lehrer's point is that the sabermetric mindset that tends to work well in baseball doesn't translate to a more team oriented sport like basketball and football. That's why he sticks to the discussion about the Mavs. It's actually a pretty damned good article if you can get your head out of the groupthink for a few seconds.
Of course Lehrer seems to be arguing against sabermetrics in basketball which is a mixed issue at best. I'm not a basketball fan but I think it's clear how in a free flowing sport like basketball the interaction between the players can make a major difference in the "sum of the parts" discussion. In baseball that is not so true as it has been said before the sport is an individual one played by teams. Jeter and A-Rod don't have to get along to succeed both individually and as a team. Lebron and Wade DO have to find a way to work together and compliment each other on the court to succeed.
Maybe, but oddly (or not so), I think it would work great for Soccer.
Back when I was following this sort of thing regularly... off the top of my head: Jeremy Giambi, Ruby Durazo, Bobby Kielty, Lastings Milledge... then again I don't think the article was so much about baseball.
Unquantifiable talents are now the new market inefficiency! Is this an example of irony? It is an interesting thought, since everyone else is doing it, why not strike out for new ways to get the extra edge. It may not be successful, but I applaud the intellectual curiosity if anyone attempts it.
Yeah, this is a good point.
Back when I was following this sort of thing regularly... off the top of my head: Jeremy Giambi, Ruby Durazo, Bobby Kielty, Lastings Milledge...
Don't knock Erubiel Durazo! He gave the A's a couple of damn good years. I think a fairer point when discussing failure of sabrmetric darlings is comparing them to how many failures there are from traditional scouting measures. Projecting ballplayers is tough no matter what decision process you use.
It's a long time back, but the Kenny Lofton experiment in Atlanta possibly applies.
I think Jack Cust is a better name for this list, personally.
I'm sure Jonah Lehrer is a smart guy. Smart enough, in fact, to hopefully understand that plenty of smarter creatures can look at the numbers without being seduced by them, or thinking they are the whole answer. I suppose this article isn't written for those creatures.
He was the best hitter on the team!
The Braves brought in Lofton not because of advanced metrics, but because he stole lots of bases (leadoff man, baby!) and hit well against them for a few games in October of 1995.
Jeremy Giambi for the Red Sox in 2003?
I'm not saying it was a bad signing, but it got made fun of by less numerically-literate types and he ended up sucking.
This kind of thing has been going on long before any FOs got stat-heavy. The number of times Dick Allen was dumped is equaled by the number of times a new team was willing to give him a shot.
If a guy can play ball, there will be a manager who thinks he's going to be the one that keeps the pain-in-the-ass in check. When he stops producing, the phone stops ringing.
That's my sense anyway. I have few connections within NBA front offices.
Enos Cabell was a favorite whipping boy of Bill James
The salary structure of baseball to some extent mitigates the problems of bad fit or poor play among unheralded sabermetric darlings. You can sign 'em cheap, and if they fail, you can sign the next one; they're not taking up a salary slot on your NBA roster and making multi-millions to have little effect on the court. The really bad personnel decisions in baseball (Gary Matthews Jr, e.g.) are almost always made in the face of sabermetric wisdom (don't spend lots on long-term contracts for aging players with a brief track record of success ...)
Keith Lockhart was mine. The man was the designated hitter in Game 4 of the 99 WS after having an OPS+ of SIXTY FREAKING SIX.
When your careless spending habits are being criticized by socialists you have problems.
Yep, people are overthinking stuff. Milton Bradley kept getting chances because every few years he would demonstrate tantalizing glimpses of batting excellence, so people kept betting on the come hoping to get a healthy/sane season out of the guy. You didn't need PaNCakE FLoPs or whatever to see that Bradley could hit.
Kenny Lofton may have been a jerk among jerks in Atlanta, I have no idea, but that would have made him a jerk coming off 3 awesome seasons in a row. A CF who averages .320, is widely assumed to play good defense and leads the league in SB isn't a guy who needs WAR to be attractive to a team.
Isn't that example of "sabermetric wisdom" also the sort of "common sense" available to front offices a hundred years ago? Remember Branch Rickey's whole thing about better to trade a player a year early than a year late, that sort of thing.
(I say that, but then why have Jason Werth; Dan Uggla; Carl Crawford; and Jason Bay sucked so badly from pretty established skill levels in prior years-is the pressure to perform to the contract getting to them?)
Uggla has admitted as much.
Which "SABR-inclined organization" brought in Milledge? Also, I don't think I've ever heard Kielty called a ######## (though I guess he was too stubborn to stop switch hitting).
EDIT: "What Dallas coach Rick Carlisle wisely realized is that Barea possessed something that couldn't be captured in a scorecard, that his speed and energy were virtues even when he missed his layups (and he missed a lot of layups), and that when he made those driving floaters their value exceeded the point score. Because nothing messes with your head like seeing a guy that short score in the lane. Although Barea's statistics still look pretty ordinary — his scoring average fell in the Finals despite the fact that he started — the Mavs have declared that re-signing him is a priority. Because it doesn't matter what the numbers say. Barea won games."
Seriously, does it matter whether basketball is not as kind to statistical analysis? What is Lehrer even saying here?
Meanwhile, Aubrey Huff, and his .297 wOBA, says hello.
The same can be said with regard all the behavioral sciences (for instance economics, or statistics based psychology). They don't work because they can't work.
I'm having a really hard time believing that NBA players get rattled by seeing a 6'0" score. If that were the case than Allen Iverson would have been the deadliest weapon in the league.
Manny seems like best example. I didn't know that Jeremy Giambi, Durazo, and Kielty were known as ########.
We’d remember that not everything that matters can be measured, and that success in sports (not to mention car shopping) is shaped by a long list of intangibles.
I love how the writer seems to believe that SABR thought advances the idea that immeasurable aspects are unimportant.
Aren't the Mavericks among the leaders in sabermetrics analysis? Focusing on those individual stats like "Which lineups perform best on the floor together"?
Let this be an opportunity for knocking down strawmen.
So is Lassus, apparently.
Best unintentionally funny line of week (bonus points for the user handle too).
The Royals have been doing this for years... [insert Francoeur joke]
Well, maybe they are on to something; they just won the championship and they did not have the biggest payroll or the biggest collection of superstars.
It depends on how you use the statistical information that you have.
The primary difference between amateur analysts and the ones who work for professional sports teams is that the former can't (or won't) go beyond basic Statistics 101 tools. The latter understand the limitations of basic tools and spend more time looking at performance components and evaluating statistical outliers.
-- MWE
*Advanced* metrics. Manny didn't need advanced metrics to convince anyone he would improve a line-up.
The OP sems to be arguing that there are players with reputations as cancers or not having playoff experience (??) or not listening to their coachs that are being brought into organizations for having undervalued skills that SABR metrics have ferreted out. And like I said..."Who?"
Unless the OP is arguing about non-player personnel...which I can totally see happening.
You think Sabermetrics is hard? Try doing it in Base 8!
The 3 is silent, you see.
Um, I'm pretty sure you're already at a false premise here and you haven't even made it halfway through this sentence. I can rattle off a list as long as my arm of "amateurs" who now work for professional sports teams.
The Mavericks didn't have the biggest payroll in the league but they did spend $20 million more on player salaries in 2010-11 than the Miami Heat.
Almost every panel I attended at the Sloan conference acknowledged that the spread and evolution of "advanced" statistics has re-inforced the importance of attempting to evaluate (with some degree of accuracy) a player's health, work ethic or personality.
TFA's point that there might be guys in basketball (JJ Barea eg) whose winning abilities are imperceptible by statistic measures may or may not be true. Are there any such players in baseball? Eric Byrnes was probably most highly-regarded along these lines. He actually did play for Billy Beane, of course, and put in a couple of decent years in Oakland, and then was sent on his way because he was getting older and worse. Byrnes fetched up in Arizona, seemed to will an outscored team into the playoffs on the force of his personality alone, got some MVP votes, became an extremely wealthy man, and played very bad baseball ever afterwards. If the Byrnes case is instructive in any way, it's hard to see it; but it does become a kind of weird Exhibit A for "intangibles," because it doesn't really add up to the sum of its parts.
This fellow is not altogether getting it.
Actually, this fellow is altogether not getting it.
Because it's really an individual sport, there may not be any. Your influence in the clubhouse may be great, but if you fail in your encounters with opposing pitchers, that outweighs your influence in the clubhouse. The best-case scenario is a guy who is a fantastic influence in the clubhouse but barely ever plays, like Mike Sweeney on the Phillies.
Sticking with the Phillies, it made sense to have the clubhouse guy Chris Coste as backup catcher in '08 and '09 instead of someone like Lou Marson who may be an up-and-coming star but would probably be unhappy sitting on the bench all the time.
Not following this at all. Statistics simply evaluates the likelihood of events occurring, based on known or estimated statistical distributions. If they don't work you either need a new model or new or better data on the phenomenon of interest. They work to the degree that they can separate signal from noise, nothing less, nothing more.
Consider the case of J.J. Barea. During the regular season, the backup point guard had perfectly ordinary statistics, averaging 9.5 ppg and shooting 44 percent from the field.
44% for FG% for a point guard is well above average. 9.5 PPG for a backup is above average. In fact, the difference in the MIA-DAL series was the Dallas' coach's decision to give Berea more minutes than Jason Kidd who is a below average shooter for his position, statistically it was a no-brainer decision. All this writer is proving is that he missed the same thing Eric Spoelstra missed, and maybe by getting away with using Kidd as long as he did Carlisle was able to get the Heat to not prepare for Berea.
There are a lot of somewhat useless basketball stats, or stats that are weighted more heavily than they should be. In baseball we know that the most important stat is OBP; in basketball the most important stat is FG%. There were only 5 NBA teams with Top 10's in FG% in 3 positions - Boston had 4 and Miami had 2. Miami was an outlier because their 2 players were either best or second-best in FG% AND were both top 10 in free throw attempts. So that pretty much makes Miami a 6th team in that group.
Of all the NBA playoff series in 2011, the only one that didn't come close to conforming to the statistical analysis were Boston-Miami and Memphis-SanAntonio, and the MEM-SA series was not far off.
If they keep their shooters and get some more rebounding in 2012, Denver and Golden State will be teams to watch.
The point is that the possibility of separating signal from noise is based upon the law of large numbers. This law of large numbers requires that a cohesion between the micro and macrocosms. It turns out, however, that in all of the cases we are talking about it is precisely this cohesion that does not exist and hence it is impossible in an axiomatic sense for statistics to separate the signal from the noise.
Comments like this make me want to stab someone in the neck.
The Mavs are the most stat-oriented team in the NBA. Lehrer slammed Roland Beech's player ratings on 82games.com (the 3 our of 4 best players crap, never mind that the Mavs had the next six players on the list after James, Wade, Dirk, and Bosh). Beech was hired by Cuban and HE SAT ON THE BENCH TO HELP CARLISLE WITH IN-GAME MATCHUPS AND STRATEGY. In other words, the only reason Barea probably started after game 3 was because a stathead crunched the numbers and found it was a better matchup for the Mavericks.
Here is what Cuban told John Hollinger about Beech and Carlisle.
http://espn.go.com/blog/truehoop/post/_/id/30227/carlisle-pushed-all-of-the-right-buttons
Amazing, it's it? Apparently they're going to implement that later. Which is of course ridiculous, how do you not have it from day one? Grantland is the cool kid's table of internet sportswriters.
Yes, but Lehrer also posted his article at Wired, which does allow for comments.
I really think you need to start over and explain exactly what you mean. You are into statistical theory here but it is not at all clear what you are getting at.
Thanks for interpreting for me, Dan. I was thinking along the lines of the differences between amateur cooks and professional chefs, or between home-grown application developers who put freeware on the Internet vs professional software engineers developing enterprise-quality applications. Almost every professional (in any discipline) was an amateur in that same discipline at one time.
It doesn't always take a massive quantity of information to separate signal from noise. Bill James referred to the concept of "signature significance" in one of the Abstracts in referring to an event that wasn't statistically significant in and of itself nonetheless being a strong indicator of quality because normal players/teams simply don't *do* things like that. I believe the example he used was the Clemens 20 K/no walk game, so it must have been the 1987 Abstract. James argued that 20Ks and no walks was far more indicative of quality than something like 17 Ks and 2-3 walks would have been, even though on a gross level it wasn't a whole lot more likely.
-- MWE
To whom?
Most of the time, the outsider simply doesn't know whether the 4.50 ERA pitcher is actually a 3.50 talent who would benefit from listening to the coach, or a 5.50 talent who is smart enough to recognize that if he doesn't listen to the coach he's going to be out of a job. Some fraction of performance is coaching, probably a lower fraction that talent, to be sure. We do the statistical community no favors by failing to recognize that there IS value in so-called "intangibles", just as the folks represented by the author of this article do themselves no favors by failing to realize the value that statistical analysis DOES provide.
-- MWE
Spot on.
Note that I didn't go as far as some would and say "if it can't be measured it doesn't exist". I am simply saying from a decision making angle if it can't be measured it's not worth considering.
Actually, someone smarter than myself could perhaps write a best seller titled "Epistemology and Sabermetrics: How do we Know What we Know?"
Who are you thinking of here? I'm not terribly familiar with the professional analysts in other sports, but in baseball, many of the professional analysts have been or are again amateur analysts--Tango, MGL, Tippett, Woolner, Olkin, Kalk, Click, James, Wright, Law, etc., and there are fair number of others who were never regularly employed by a team but have done consulting projects. So I'm not sure I understand what distinction you were trying to make.
What are you thinking of as basic Statistics 101 tools?
Without you being more specific about either of these two things, it's easy to think of a number of examples that "disprove" your point. But I'm curious where you were actually going with it.
This is because nobody associated with that site (Bill Simmons in particular) knows anything about baseball (or hockey). I wonder if they realize how dumb they sound talking about those sports.
I see you commented while I was typing my post. I have definitely seen people with database skills (as Dan mentioned) getting hired by baseball teams, as well as people who generally have good sense with how to handle data as opposed to just pushing buttons to do a linear regression. What I haven't seen is very many people applying advanced statistical techniques in baseball or the people who do that getting hired by teams. Russell Carleton is the one exception that I can think of there.
I rarely interview people, but I think I have most of the rest of the skills a writer needs. I should have figured out that I liked to write before I was 30. I think that has stunted my career; such that it is.
I think the point some are making about Stats 101 vs. more advanced tools can be illustrated in the following - if any of you have read the book on fielding metrics by Michael Humphreys ("Wizardry: Baseball's All-Time Greatest Fielders") - the first several chapters try to explain in detail what statistical methods are being used by the author to come to his conclusions. I think it is sort of interesting and somewhat followable - but my head still wound up spinning and I jumped to the rankings results charts, since that to me is where the fun is!
Point being, I guess my brain really struggles with more advanced mathematical analysis. There, I admitted it! I think the level of sophisticated analysis that ML teams are looking for (i.e., "professional") is way beyond my simplistic view of things and, as others also mentioned, superior database management is absolutely imperative for a team looking for a market edge.
That being said, there is some beauty in simplicity sometimes. I think Voros McCracken would admit that he kind of stumbled on BABIP as something very basic - but he noticed a pattern there that apparently no one else was paying attention to.
As to players who were coveted almost solely because of their SABR tendencies in spite of a cancerous personality or lack of intangibles like "character" - I would have thought someone would have mentioned Barry Bonds by now. He kind of seems like the elephant in the room on that topic. I wonder - had he played back in the '60s when we were less enlightened - would he have been shuffled from team to team like Dick Allen was? I think by the time Barry's career came along there were much better methods of statistical evidence available showing his incredible performance level that made a team like the Giants think having him on your side was worth all the headaches that came with the package.
#70 (Drew) - Keep this in mind - I seem to remember Bill did a column once where he a)said he wasn't necessarily convinced Wade Boggs was a HOFer (didn't drive in enough runs or hit well enough in the "clutch" or something like that) and b)said something about what a great actor Adam Sandler was - thus showing complete lack of knowledge on two different topics simultaneously! I think Bill's better on basketball than any of his other stuff, and perhaps he's suffered some damage from being a die-hard Celtics fan living in Los Angeles!
Tell me about it - Grantland is driving me nuts. I hate the way half the articles are written
Um, no.
Sandler is closer to funny than the other three listed.
This sounds very reasonable and even-handed, but is really false equivalence of the worst kind. Failing to learn what statistical analysis can teach is willful ignorance, usually resulting from laziness or prejudice (in general -- not commenting on this particular article). Whatever analysts may fail to appreciate about "intangibles" -- if it is anything at all -- is tiny compared to the loss of knowledge entailed in rejecting modern analytic methods.
And please let's not confuse "intangibles" with the idea that players can learn skills (from coaches, or others). I don't know any saberist who rejects the idea of learning, and believes all baseball skills are present at birth. Of course players can learn. That appeal to "intangibles," however, invariably boils down to an evidence-free argument for ignoring statistical evidence simply because that evidence leads to a conclusion someone finds unpleasant.
Um, no.
Kevin James and Will Farrel have done nothing funny at all. Ben Stiller hasn't done anything in a long time.
Sandler at least gets an occasional chuckle still.
Ben Stiller has written, directed and produced movies funnier than either of the others. Left to his own devices, Ben Stiller makes Zoolander and Tropic Thunder. Left to his own devices, Adam Sandler makes Mr. Deeds.
The Extras episode featuring Stiller as his filmmaking self is very sharp. Laughing at yourself is an oddly rare commodity among comedians.
So you're saying Stiller and Sandler are equal?
BTW, am I the only person who sees a ridiculous amount of parallels betwee Zoolander and Jeter. I'm half convinced they modeled the charachter based on him.
1. First name: Derek
2. Metrosexual pretty boy / image conscious
3. Has casual sex with a string of hot girls
4. Dumb (more a jock stereotype than specifically Jeter)
5. His big weaknes: He can't tutn left.
Just say you love crepes.
You must be Registered and Logged In to post comments.
<< Back to main