Thoughts on the Sloan Sports Conference at MIT: What can baseball fans take away from the nerdery?
by Evan Hochschild on Mar 9, 2010 6:00 AM CST
in Stat Geeky
Email
Print
To me, there's nothing better than discussing sports in an intelligent, critical fashion. Some people love debating politics, or recapping the previous night's episode of American Idol or The Office. That's all good and well, but personally there's just something about athletics played at its highest levels that brings out my inner chatter box.
What's even better than me talking sports is listening to people who are smarter than myself do it.This past weekend's MIT Sloan Sports Analytics Conference was one of those summits of the who's who in American sports that had me absolutely giddy as someone who can't get enough of creative thinking in the realm of sports.Two specific panels from the Conference were of particular interest to me: the one dealing in Baseball Analytics and What Geeks Don't Get: The Limits of Moneyball.
On the baseball side of things, we know that statistical analysis of the sport has been gaining momentum not only with fans like ourselves, but in the mainstream mediums as well. When I heard Joe Morgan mention OBP last year, I nearly fell out of my seat. Truly, by hook or crook, sabermetrics has gained a foothold, one that it is not going to be giving up. Whereas certain teams had a strong competitive advantage at the outset of the statistical revolution, most if not all teams utilize advanced statistical analysis in some form or fashion in their decision making.
For those teams like say, the Astros, who probably don't take the analytical approach to heart, they run the risk of being left behind the learning curve. The game of baseball itself is complex, mutli layered and difficult to grasp on it's most fundamental levels. Finding the right combination of players to field a successful major league team is a task left to men who boast advanced degrees from Ivy League schools and scouting personnel who have earned their stripes watching more hours of amateur baseball than anyone should be made to. I would like to think that the Astros have begun to meld these two schools of thought together, but something tells me that they have not. I'm not saying that the organization hasn't improved itself over the past couple years, but I seriously doubt Tal Smith and Ed Wade are at the forefront of developing new methods to analyze and evaluate baseball players.
It's a culture of thinking, when you boil it down to it's most base level. Starting from the top, how front office personnel look at players and conduct business is ingrained on an organizational wide level. This culture begets another sort of culture on the field, whether it be winning or losing.
The Astros have seen what their previous culture got them- poor performance on the field and inefficiencies throughout the minor league levels. Bad drafts, poor minor league records and an aging major league team are the after affects of this sort of mismanagement. I hope, hope, hope that the Astros take the sort of analysis which the presenters at the Sloan Conference undertake to heart.
As we head into the second decade of this century, it's readily apparent that objective, statistical based analysis works to a large extent and combined with scouting enables organizations to take advantage of market inefficiencies to yield a strong on the field product. Our Astros may be behind, but they doesn't always have to be that way. They need only look across downtown Houston to see that....
My second area of interest in Sloan Conference dealt with the extent to which professional sports franchises can utilize statistical analysis before a point of diminishing returns occurs. The panel was moderated by Michael Lewis, and had such varied participants as Dallas Mavericks owner Mark Cuban, Houston Rockets GM Daryl Morey, and ESPN's Bill Simmons.
In all honesty, I looked forward to hearing about the exchanges of these three men much more than Rob Neyer and other baseball executives from the Baseball Analytics panel. Simmons is a thoughtful, speak from the heart sort of guy, whose writer/fan perspective is unique. He has made a career of second guessing NBA GMs and dishing out strong criticisms on how teams are run in the Association.
Cuban is best known for his theatrics on the sidelines, but the fact remains that the man is a billionaire and is incredibly intelligent. His Mavericks are at the fore of statistical analysis in basketball.
The Boy Wonder, Dork Elvis, call him what you want, Daryl Morey is the Billy Beane of the NBA. The first "stat" GM the league has seen, Morey has been GM of the Rockets since 2006 and under his watch, the team had won fifty games in each of his first three seasons, and are playing respectable (if not frustrating) ball this season sans Yao Ming. As an MIT grad, Morey was instrumental in organizing the whole sha bang.
One topic that has driven sabermetricians mad seemingly forever is the question of clutch. Can a player be clutch, or are performances in important, high leverage situations dependent much more on circumstance, match ups and other external factors?
Michael Lewis asked Daryl Morey whether or not the Rockets utilized "clutch" stats. Morey answered this way:
We don't make any decisions based on the belief of that.
This sort of falls in line with what those in the baseball saber-community have believed for years. So, not an incredibly surprising answer. Mark Cuban, for what it's worth, said the Mavs do make decisions based on the idea of "clutch".
Without having done any real research on the subject, my own experiences and observations watching basketball lead me to believe that there is such a thing as clutch in the NBA. Or, maybe not clutch, but there certainly are players and teams that are able to put themselves in favorable situations at important junctures in a game.
While Kobe may miss 7/10 game winning shots, it is readily apparent that the Lakers and other elite teams win close games for a reason that other teams are unable to. It's not so much clutch as it is poise, composure, and experience. Players like Chauncey Billups earn nicknames like "Mr. Big Shot" from making late second plays, but to me, it's as much about his not turning the ball over and putting his teammates in optimum situations to score that separates him from other players.
I don't think these characteristics matter as much in a sport like baseball, where the actions are much more instinctual. Jason Michaels experience isn't going to gain him any advantage in the batters box, and the inexperience of Justin Upton probably won't hurt him more now than it will when he is 30. That's just a difference in the sports, or at least I find it to be that way.
Advancing the methods by which professional sports organizations make decisions makes for great discussion for fans, but is also a necessary process for the organizations themselves. Progress is unavoidable and it's best to at least stick your toe in and see what the fuss is all about. Teams have everything to gain, but nothing to lose by doing so. Hopefully gatherings like the Sloan Conference will increase the visibility of statistical analysis while simultaneously driving the continued improvements of the underlying methods of evaluation.
25 comments
|
0 recs |
Do you like this story?
Comments
Finding the right combination of players to field a successful major league team is a task left to men who boast advanced degrees from Ivy League schools
Well, that’s overstating the case.
I’ll try to pass over the reference to Ivy Leaguers, since I realize it’s meant to be a stereotype and not a real suggestion that MLB GMs can only come from the Ivy Leagues. However, I’ll just point out that Bill James, the father of sabermetrics, has undergraduate degrees (English, Economics, Education) from the University of Kansas.
More to the point, I think that experienced smart baseball people who don’t have the Theo Epstein background can still be important to constructing a team. For that matter, GMs like Billy Beane and Kevin Towers, who are viewed as good GMs by sabermetric types, have the traditional baseball experience rather than Ivy League degrees as their qualification. I actually like the idea of combining both types of front office people, like Hunsicker / Friedman in Tampa Bay and Nolan Ryan / Daniels in Arlington.
well, I would wager
that most new GMs are schooled in statistics/math/economics/etc., and went to highly rated, Ivy-esque schools….if not, well, I owe you a Coke.
The Crawfish Boxes, Astros blogging at its finest.
by Evan Hochschild on Mar 9, 2010 9:20 AM CST up reply actions
Hmm…well, if you expand it to “Ivy-esque” schools, that expands the number of schools (or makes it more subjective). I won’t try to say whether these schools are Ivy-esque or not, but here’s the educational backgrounds of the most recent round of new GMs (well, Daniel and Friedman aren’t recent hires, but they are young GMs):
Jack Zduriencik BA Education California Univ. of Pennsylvania; MA Physical Education Austin Peay
Alex Anthopoulos BA Economics McMasters University (Canada)
Andrew Friedman, B.S. Management, Tulane University
Jed Hoyer, (don’t know degree) graduate Wesleyan University
Jon Daniel (don’t know degree) graduate Cornell
Except for Daniel and Anthopoulos, all of these GMs played on their university baseball team.
gotcha
I shouldn’t have been so absolute in saying "Ivy League", I guess. But look at those schools….2 Ivies, two elite private schools (Tulane and Wesleyan) and a Canadian school of which I know nothing…again, not all Ivies sure, but schools that only a small, small minority of people could get accepted into
The Crawfish Boxes, Astros blogging at its finest.
by Evan Hochschild on Mar 9, 2010 11:21 AM CST up reply actions
Unless I'm reading it wrong
And it’s possible, because it says that Zduriencik went to “California University of Pennsylvania”, it looks like two of those guys actually did go to Ivies. Assuming that Zduriencik went to Penn.
For what it’s worth.
I'll eliminate you like I eliminate gluten from my diet.
by tehGrindCrusher on Mar 10, 2010 6:47 AM CST up reply actions
California University of Pennsylvania....
is not the same as Penn, as best I can tell. California University of Pennsylvania athletics are NCAA Division II, and the Ivy League is NCAA Division I. Zduriencik played baseball for Cal. Univ. of Penn., which is located in the city of California, Pennsylvania, outside of Pittsburgh.
Who knew?
I'll eliminate you like I eliminate gluten from my diet.
by tehGrindCrusher on Mar 12, 2010 8:42 AM CST up reply actions
I'd guess that all teams have used stat analysis forever
I suppose at the very beginning of baseball it might not be true. OBP is not an advanced stat and has been around for a long time. Walk rates, K rates, roster construction, lineup construction, long ball, small ball (in particular the wisdom of an aggressive approach to base stealing), BABIP, and the value of fielding were all debated before the internet even existed.
Is it fair to restate the clutch claim as thus: in baseball human performance does not vary with the pressure of the situation. Imagine making that claim to a nuerobiologist. Imagine telling a logic professor that the support for that included this reasoning: if they could do it then, they would do it all the time.
… certain teams had a strong competitive advantage at the outset of the statistical revolution
Is this true? Is Oakland the main argument in support of it? His success seems based mostly on choosing, perhaps in addition to coaching, some very talented pitchers.
I think there is some mixed opinion on clutch hitting among analysts identified with sabermetrics. Many years ago, Bill James performed studies which made him doubt the existence of clutch hitting as a skill. But his opinion evolved over time. Since then, he has espoused the view that no study adequately resolves the whether clutch hitting exists, because the confounding variables and sample size issues are too complex. Also, I think that he questioned the orthodox view (which he originally helped establish) about clutch hitting after watching David Ortiz perform so well in the clutch night after night for several years.
I like ol’ Pete’s re-statement of the hypothesis, because I have always felt that the denial of any clutch hitting effects goes against the grain of much pychological study of the impacts of stress and pressure on human performance.
I would love to attend this conference
Mainly for the reasons that Evan specified above.
What I found most compelling was the sheer mix of personalities present. It was just about the stat-crunchers of the world divining the next step forward in analysis. Rather, there was a clear focus on making stats accessible to the rest of fandom (the non-Ivy Leaguers).
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by Stephen Higdon on Mar 9, 2010 9:41 AM CST via mobile reply actions
I feel like I keep coming off too negative
I just see counter arguments and am putting those out there. Let me just say, I give a big thumbs up to the inquiries and examinations that are discussed. Good job Evan. Is Evan the artist formerly known as HLP?
One caveat though. All writing would have to be observed for evidence of irrationality if there is evidence of Green and Gold fever.
Evan is the artist formerly known as HLP
I’m the artist formerly known as DQ
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by Stephen Higdon on Mar 9, 2010 10:28 AM CST via mobile up reply actions
hahaha no worries, Ol Pete
My point wasn’t really put to argue anything really, but to expound upon things that are pretty wide spread views…
As far as Aaron Rodgers…I can’t deny my roots. I’ll be a Cheesehead forever!
The Crawfish Boxes, Astros blogging at its finest.
by Evan Hochschild on Mar 9, 2010 10:58 AM CST up reply actions
Without having done any real research on the subject, my own experiences and observations watching basketball lead me to believe that there is such a thing as clutch in the NBA. Or, maybe not clutch, but there certainly are players and teams that are able to put themselves in favorable situations at important junctures in a game.
And simply viewing baseball would lead you to the same conclusion. The human mind creates an explanation for patterns it notices, and it identifies patterns as a matter of course, even if those patterns are (at best) tenuous.
Much like baseball, once you look at the statistical information, it doesn’t appear that there is any such thing as “clutchness.” Indeed, the players whom we typically think of as clutch (Kobe Bryant, for instance) are pretty much random in “clutch time” from year to year. And Kobe, in particular, is an intriguing example: across time, he’s actually below-average in those situations. But he’s been in these situations a lot (the Lakers are in the playoffs almost every season, and they always have lots of ESPN/ABC games).
As a story – as a single-use explanation for single events – “clutch” works. But there’s no evidence of repeatability, and there’s no reason to believe that it has any significant effect on the game.
Besides, if you want to be really good in the clutch, just score more points (and defend better) earlier in the game.
now,
if you just want to say that “Some teams put themselves in good situations and do good things,” I agree – there are teams that tend to win more often than others. Those are the good teams, and they’re good because their players are good and do the things that they need to do to win games. But even those teams don’t show any particular ability to win close games more often than other teams.
Good teams are good because they blow out a lot of bad teams. They do the right thing throughout the game, not just when they’re down by one with ten seconds left on the clock.
I think you can also look at it like, “Who are the best players at creating their own shots?” That skill doesn’t change in clutch situations, but players who normally can get a good shot will be more likely to do so at the end of the game too.
Same goes in baseball. Good hitters will come through eventually, no matter if it’s in the second inning or in Game 7 of the World Series.
by David Coleman on Mar 9, 2010 10:33 AM CST up reply actions
That's a good way of phrasing it.
Kobe’s good. Pujols is good. If you put them in situations where they need to be “clutch,” they’ll probably come through more often than their peers, simply because they’re good players.
At the same time, players like, say, Shane Battier are relatively unlikely to be described as “clutch” in basketball, because they’re much more unlikely to be in a situation where they’ll get the last shot of the game.
That’s not really the case in baseball, except that good players are simply going to get more PAs than bad ones (usually).
No evidence of repeatability is a bold statement as is no reason to believe. You could just as easily say that studies have been done to disprove it, so it is a myth is a good story. Do you believe that and might your mind be creating a pattern that doesn’t exist? I have to say the requirement for career long repeatability is a poor standard.
When players say they are “in the zone” do you think that’s their imagination?
No evidence of repeatability is a bold statement as is no reason to believe.
It is, but fortunately we have statistical evidence to back it up.
You could just as easily say that studies have been done to disprove it, so it is a myth is a good story
I’m really not following you here.
I have to say the requirement for career long repeatability is a poor standard.
Well, we don’t have to look at career-long trends. Just from year-to-year, players have huge swings in the “clutch” stats. And keep in mind that basketball stats tend to be fairly consistent once adjusted for time and/or pace – more consistent than baseball stats, anyways. So there’s a great deal of repeatability with most player’s skills to rebound/score/pass from year-to-year, but very little evidence of repeatability for their ability to do those things in a defined “clutch” set of circumstances.
When players say they are "in the zone" do you think that’s their imagination?
Yes and no. Players really can be on a “hot streak,” but evidence suggests that this has little predictive ability. Whether or not you were “in the zone” on your last five field-goal attempts doesn’t seem to affect whether or not you’ll be there on your next one.
People create stories to explain what they experience. It’s human nature. The issue we have is that we see guys like Kobe hit a few big shots at the end of games and we say “Well, he does that because he’s clutch,” rather than looking at the data and seeing that he’s really no better than anyone else in hitting those shots. But “Kobe is a good player who shoots well” doesn’t sound nearly as interesting as appealing to “clutchness.”
He’s “clutch” because he’s done “clutch” things, much as he’s a “champion” because he’s won several championships. But neither of those things, when separated from other variables, appears to have much predictive power.
BPro
Just did an excellent piece on this topic. I don’t have the link because I’m on my phone, but for those who have the means, I highly suggest checking it out.
The essential conclusion is that we infer/create meaning after the fact. IE everyone feels locked in after the fact, but never do we hear “I was dialed in during BP, look for a monster two days.” etc. Not proof that it doesn’t exist, but a valid qualifier to reading discrepancies between player’s reports and what the numbers can find.
The Crawfishboxes
A good friend of mine used to say, "This is a very simple game. You throw the ball, you catch the ball, you hit the ball. Sometimes you win, sometimes you lose, sometimes it rains." Think about that for a while.
by Stephen Higdon on Mar 9, 2010 11:34 AM CST via mobile up reply actions
My pet peeve about most discussions / studies of “clutch” performance is that it treats a player in isolation without considering his opponents. If hitters can be clutch, so can pitchers. Admittedly basketball is different than baseball in terms of potential variables, but conceivably an opposition player can be clutch on the defensive end just as the shooter can be clutch. What happens when two clutch players oppose each other in a clutch situation. (If Sandy Koufax strikes out Mickey Mantle in a crucial situation in the World Series, does that mean Koufax is clutch but Mantle isn’t?) In my mind, this makes a determination of the clutch question intractable…or at least, I don’t know how to get around the problem.
I think the “in the zone” idea is different from clutch performance. I think this goes to the idea of streaks, etc. In my own personal experience, I have a lot of streakiness (I will go through periods playing tennis when I can’t lose points, and vice versa…the same applies to my performance in other games, like pool). My feeling is that streaks of good performance result in confidence feedback (i.e. I become more confident as I succeed) and that streaks of bad performance cause me to battle frustration. That’s why, for instance, I try various mind games on myself when I’m performing poorly in tennis. I also think there is feedback vs. the opposing player (if I gain confidence as I am playing, sometimes the opponent is losing confidence). Admittedly, I’m not a professional athlete, and their overall level of talent is so high that confidence may be less of an issue. But I do think that a confidence feedback loop contributes to hitting streaks, players “in the zone” etc.
The "defensive clutch" thing is a good point.
Mostly, I think it’s just that people automatically look at the offensive end of any sport. On the other hand, I’ve heard a few people refer to, say, Hakeem’s block of Starks in the ‘94 Finals or Carl Landry’s block on Deron Williams in the ‘08 Playoffs as “clutch.” Closers don’t get labeled “clutch,” but I suspect that any closer who “choked” routinely wouldn’t be a closer for very long. Of course, the viability of “saves” as a meaningful stat is a discussion in itself.
I think this goes to the idea of streaks, etc. In my own personal experience, I have a lot of streakiness
Part of the issue, I think, is that people look at “streaks” and assume there’s a “trend,” and thus “something changed.” In reality, any random assortment of events will have “streaks” in it. A random assortment of cards will probably have a run or two of single suits or of a particular face. “Random” doesn’t mean “alternating,” after all.
Shit, I do it, too, and I like to think I have a pretty good understanding of the way probability works.
But this is what I mean when I say that people formulate a story off of an invalid “pattern.” We see a streak of diamonds and we assume that the deck wasn’t shuffled properly. We see Berkman strike out twice and we assume he’s in a slump. We see Kevin Martin hit a few threes and we say he has “adjusted to the system.”
They’re probably just random events, and we’re just throwing our preconceived notions onto them.
randomness implies that the individual’s mental state can’t affect the order. I think that there can also be a self-fulfilling prophesy effect too. A player gets three hits in a row or hits three shots in a row. In actuality, that may just be a random ordering of events. However, if the player interprets it as “I’m in the zone,” he may actually put himself in the zone, so to speak.
The existence of studies is light evidence. The very long history of the use of statistics is that study construction and conclusions that are derived don’t equal the conclusions being true many, many times. They often involve massive amounts of data. Professional statisticians clash in their discussions and believe me, sides often claim to have proved one thing or another. Statistical debates in medicine and economics sometimes last decades or longer. There was a large amount of statistical evidence that the most common form of stomach ulcer was caused by excess acid production. It is now known to be not true. The efficacy of anti-depressants remains controversial. Spending money on education in inner cities is still debated – witness the book The Bell Curve (sadly financed here in my home town). It made a complex statistical argument that was widely acknowledged till professional statisticians read it.
The point I was making was that the same inclination to believe may be affecting you rather than those who disagree.
There is also significant evidence that players can be “in the zone.” Its existence is acknowledged in scientific disciplines which would all have to be wrong if you were right that it was anomalous luck.
I wanted to compliment you...
… on your excellent replies, Only_A_Lad. One of the guys on Niners Nation linked this article, but I’m recommending that folks read your comments specifically. I can’t decide if it’s depressing or reassuring that other blogs have the same arguments about the validity of statistical analysis generally, or of treating “clutch” as a skill specifically. Regardless, your reasoned comments on these subjects are appreciated. I wish I had half the patience you apparently do.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recent FanPosts
TCB Tweets!
Current Series
3 game series vs Rockies @ Minute Maid Park
|
Colorado Rockies
Friday, Apr 6, 2012, 6:05 PM CDT |
||
| Sat 04/07 | 6:05 PM CDT | |
| Sun 04/08 | 1:05 PM CDT | |
NL Central Standings
| W | L | PCT | GB | STRK | |
|---|---|---|---|---|---|
Chicago
|
0 | 0 | .000 | 0 | Lost 0 |
|
Cincinnati
|
0 | 0 | .000 | 0 | Lost 0 |
|
Houston
|
0 | 0 | .000 | 0 | Lost 0 |
|
Milwaukee
|
0 | 0 | .000 | 0 | Lost 0 |
|
Pittsburgh
|
0 | 0 | .000 | 0 | Lost 0 |
|
St. Louis
|
0 | 0 | .000 | 0 | Lost 0 |
(updated 2.9.2012 at 12:00 PM CST)
In Association With
Big League Stew - MLB - Yahoo! Sports
- Pennsylvania senator cries foul over Nats plans to take back park from Phillies fans
- Photo: Comerica Park readies itself for Hockeytown
- Podcast: Dirk Hayhurst talks about Italy, the reality of the big leagues and his new book
- Dennis ‘Oil Can’ Boyd claims he used cocaine before a majority of his games
- Does Billy Beane’s contract extension signal an optimistic dawn for Oakland fans?
Certain photos copyright © 2012 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.
