Skip to main content
Menu
What Happened At The 2015 MIT Sloan Sports Analytics Conference

FiveThirtyEight’s delegation made the pilgrimage to this year’s Sloan conference, a kind of mecca for anyone who’s obsessed with sports, data and retelling how they first felt when they read “Moneyball.” We updated all of Friday and Saturday from Boston, where Houston Rockets GM Daryl Morey, MLB commissioner Rob Manfred and our own Nate Silver roamed the halls. Read on for highlights from the weekend.


More Sports

Sunday, March 1 12:17 a.m.

After what conference co-organizer Jessica Gelman said was a “heated discussion,” voters for the top research paper at Sloan reached a split decision and split the $30,000 prize pool between two papers. The winners:

Rosales and Spratt, both of Baseball Info Solutions, presented work suggesting that pitch framing, which has traditionally rewarded most of the credit to catchers alone, is actually a function of three independent participants: the catcher, pitcher, and umpire.

Franks, Miller, Bornn, and Goldsberry — all members of Harvard’s XY Hoops group used player tracking data to quantify individual defensive play in the NBA. The academic version of this group’s paper has been accepted at the statistics journal Annals of Applied Statistics, and can be downloaded here.

The groups behind the winning papers each received $15,000 for their efforts. Additionally, Bornn and Goldsberry, along with co-authors Alex D’Amour and Dan Cervone, received the conference’s top poster prize of $1,000 for “Move or Die: How Ball Movement Creates Open Shots in the NBA.” — Mike Lopez


Saturday, Feb. 28 4:05 p.m.

 


Saturday, Feb. 28 3:15 p.m.

Will sports betting inevitably become legal in the U.S.? It sure seems like it.

Momentum behind legalization has grown since NBA Commissioner Adam Silver wrote an Op-Ed in the New York Times in November 2014 explicitly endorsing legal sports gambling. The facade of professional sports leagues that oppose sports betting is beginning to crack. And it’s clear why: money.

Silver estimated the market for illegal sports wagering is currently $400 billion per year, though it’s likely that figure is inflated. But even lower-end estimates of around $80 billion still represent a huge market. Sports betting is already enormous in Europe, Australia and many other regions. State governments want in because of the potential revenues.

Professional sports leagues are intrigued because they see gambling interest as a ratings driver, much like fantasy sports have been. (And, coincidentally, daily fantasy sports sites – with cash prizes – bear an eerie resemblance to gambling anyway.) Gambling is already inherently analytical; but the appetite of stats-savvy fans for geeky coverage about odds is growing. Jeff Ma, a contributor to ESPN’s new sports-betting site, Chalk – said gambling analytics would meet the demand from those with a “high-brow” interest.

But there are major risks to legalization. The revelations that former NBA referee Tim Donaghy owed gambling debts and bet on games he officiated was a reminder of the long, scandalous history of how gambling can challenge the integrity of sports. Here, too, analytics can help. Ryan Rodenberg, a professor at Florida State University, suggested statistical scrutiny of betting markets would combat fraud and fixing. Several private European firms already specialize in such analytics.

The panelists were asked that if they had to bet on legalization sweeping the country, when it would happen. The lines offered by the panelists ranged from 2-to-10 years. Dan Spillane, the Assistant General Counsel for the NBA, didn’t offer a timeline, however. He just said “years, not months.” — Andrew Flowers


Saturday, Feb. 28, 1:46 p.m.

The panel: “Beating the Shift: Baseball Analytics in the Age of Big Data”

The panelists: Sandy Alderson, Dan Brooks, Dave Cameron, Ben Lindbergh, Jonah Keri

Sloan’s flagship baseball panel largely focused on teams’ reactions to sabermetric findings. Alderson, the general manager of the New York Mets, spoke about the proliferation of defensive shifts, and how it has led to changes in the way certain players are valued — specifically right-handed power hitters.

Along the same lines, no discussion of baseball analytics would be complete without some mention of strike zone analysis and catcher pitch-framing metrics. Despite the volume of research on the subject in recent years, the consensus of the group was that the market may still not be properly valuing catchers who “steal” strikes on the edge of the strike zone at a higher rate than their peers. Then again, part of that may relate to a theory that pitch-framing is a taught skill. (We’d have liked to hear more thoughts about how umpires doing better at calling an accurate strike zone has led to baseball’s aforementioned drop in run-scoring.)

Finally, Keri asked the panel their thoughts about wins above replacement (WAR). The panel agreed WAR was a valuable framework, even if its individual parts can always stand to be improved. For his part, Alderson confirmed that teams use at least some version of it, even with its imperfections, because the idea of creating a cumulative statistic is appealing. — Harry Enten and Neil Paine


Saturday, Feb. 28, 1:00 p.m.

The session: “Analytics of the Tommy John Injury Epidemic”

The speaker: Glenn Fleisig

We’re in the midst of an epidemic of elbow injuries among major league pitchers. Twenty-five percent of current MLB pitchers have had an ulnar collateral ligament reconstruction (“Tommy John surgery”) and 15 percent of minor league pitchers have undergone the procedure. Over the last decade, the problem has trickled down to high school and little league players. In 1990, none of the baseball players coming to the Andrews Sports Medicine & Orthopaedic Center clinic Tommy John surgery were kids. Today, one third of them are high school age or younger, said Fleisig, the research director at the American Sports Medicine Institute.

What’s to blame for the epidemic of torn elbow ligaments? Are more players getting hurt? Or are we just better at diagnosing these injuries? Are doctors more willing to do the procedure? Or are patients more eager to have it?

The answer, said Fleisig, is all of the above. Some players assume they should go in for surgery at the first sign of elbow pain, just “to get it over,” but that’s the wrong attitude. Best case scenario, the surgery can return a player to the career trajectory he was on before he got injured, but it won’t improve performance and not every player makes it back to play, Fleisig said.

About 80 percent of major league players who get Tommy John surgeries make it back to the mound, but only two thirds of those who undergo the procedure make it back and stay there.

Most elbow ligament injuries occur due to overuse. During the middle part of the pitch when the elbow is held upright at a right angle, the joint experiences severe torque. “It’s like holding a string with five 12-pound bowling balls,” Fleisig said. (That’s why doping raises the risk of an elbow injury — “If you’re on the juice you’re making your muscles too strong for your tendons and ligaments to handle.”)

There’s a common notion that curveballs are dangerous, but the research doesn’t bear that out, Fleisig said. “We expected the curveball to have more torque than the fastball, but it turns out it has less.”

Four things determine which players get injured — biomechanics, how much a player pitches, training and recovery. “It’s not one of these things or the other, it’s all of them,” Fleisig said.

Wear and tear on the elbow is one of the most important factors, and when Fleisig’s group followed a group of 500 kids over a ten-year period, they found that pitching more than 100 competitive innings more than tripled the risk of needing a Tommy John surgery. Likewise, more than 80 pitches per game quadrupled the risk of injury, and kids who pitched when fatigued had 36 times the risk of having surgery.

In an effort to cut the rates of elbow injuries among young pitchers, Fleisig and his colleagues have teamed with Major League Baseball to create Pitch Smart, age-appropriate guidelines to avoid injury. Suggestions include limits on the number of pitches thrown and not pitching when fatigued. “The best computer we have is right here,” Fleisig told me, pointing to his head. — Christie Aschwanden


Saturday, Feb. 28, 11:20 a.m.

Screen Shot 2015-02-28 at 12.55.35 PM


Saturday, Feb. 28, 9:50 a.m.

There are bold-faced names headlining the ninth annual Sloan Sports Analytics Conference, sure. But for academics like myself the real action is in the research paper contest, where academics and researchers are hoping to create the kinds of insights that the bold-faced names one day treat as gospel. For these researchers, Sloan marks the culmination of what can be more than a year’s worth of work. The stakes are high: top prize is $20,000, with second place worth $10,000. For some of the PhD students submitting papers, that may equal their annual salary.

But until now, the mechanics of how this contest is judged have largely been cloudy (see an overview of the 2015 contest here, or my personal experience submitting a paper to the 2014 contest here). On Friday, conference co-lead Paul Campbell helped clarify how Sloan makes its picks. “We try to be consistent about what we solicit,” said Campbell. “We kind of have our perspective on the validity of the method, and making sure that the academic and mathematical rigor is there. Also, do the results make sense?”

The 2015 research paper contest began back in September, when Campbell and this year’s judging committee, comprised of various MIT student organizers and academic advisers, received 189 abstracts. Of that total, 68 were invited to turn in a formal paper for submission in mid-December. Paper submissions were placed into one of four tracks: basketball, baseball, other sports or the business of sport. The top two submitted papers from each track were given the opportunity to present at this year’s conference. In addition, 11 papers were given a poster in the halls of the convention center.

Each of the eight finalists were allotted a 20-minute presentation on Friday. The judging committee identified the top presentation in each of the four categories based on a 50-50 split of the presentation itself and the originating paper.

“We have an idea of what the best analytically rigorous paper is, but we want to see if it is presented well. It’s an equal weighting with [the presentation] and the paper,” said Campbell. The four papers still in the running for the top prize, are:

Those four finalists are given an additional 10 minutes with which to make their case, this time in front of a larger and more general audience, including Houston Rockets GM Daryl Morey and FiveThirtyEight’s own Nate Silver.

Those judges, according to Campbell, are asked to judge by something different than the last committee. “Which of these [papers] seems like the most applicable or potentially transformative within the industry?” $20,000 rides on the answer. — Mike Lopez


DSC_1481Friday, Feb. 27, 6:10 p.m.

Walking into a conference at Sloan today I walked by yet another guy in a sports coat — and then did a double take, because this guy’s blazer sleeves were rolled up…and he was a 13 year-old. There are some teenagers running around Sloan but none looked younger than Sam Hafetz and his friends, Manu Hurskovitz, 14, and Jonah White, 14. After calling their parents for permission (hi, Mr. Hurskovitz!), I dragged them to our podcast table. There, Jody Avirgan asked what brought them to Sloan (it’s their second year attending), why they love sports analytics, and what they’d do if they became GMs of the Celtics. — Chadwick Matlin


Friday, Feb. 27, 4:43 p.m.

The panel: “Commissioner’s Perspective: Growing Soccer with Don Garber”

The panelists: Don Garber, Grant Wahl

MLB and MLS share two letters and the pickle of how to balance tradition and innovation. Baseball’s struggle comes from within, as Major League Baseball Commissioner Rob Manfred showed earlier Friday. Major League Soccer isn’t as conflicted about changing rules and trying new technologies, MLS Commissioner Don Garber said. Its burden, unlike MLB’s, is its peripheral place in a global game.

Garber said he wanted goal-line review technology, extra time put on scoreboards (instead of only a ref with a “Timex that probably cost 20 bucks” knowing how much time remains) and a whole lot more. “If I were king, we would have instant replay, we would have cameras on our players, we would be putting them on goalposts.” He’d put a microphone on the field. Players would wear GoPro cameras. He watched hockey players wear GoPros at the NHL All-Star Game and thought it was cool.

But Garber can’t have all those things. Other sports’ U.S. pro leagues just need to get the owners to agree, but MLS needs the approval of IFAB, the International Football Association Board — or, as Garber called it, the International Federation of Somebody Who Has Something To Do With the Rules That’s Not Me. Garber’s message to IFAB: “Let us be the Guinea pigs.” He worries that the world’s most popular sport could lose its lead “just because of our structure. We should be able to use the power of our influence to lead.” — Carl Bialik


Friday, Feb 27, 4:35 p.m.

IMG_0032At last year’s Sloan conference, Dean Oliver was our ESPN colleague, leading analytics at the Stats & Info Group. This year, he’s here as the Sacramento Kings’ director of player personnel and analytics. I spotted him Friday huddled with a few of his peers from other NBA franchises. Oliver has been in the sports analytics business for three decades, and has seen it grow from a field wrestling with a lack of data to one with more data than it knows what to do with. He spoke with me about the similarities in working for teams and working for sports media, and about what it takes for a franchise to succeed at using analytics. — Carl Bialik


Friday, Feb 27, 3:28 p.m.

The panel: “Commissioner’s Perspective: 1 on 1 with Rob Manfred”

The panelists: Brian Kenny, Rob Manfred

Rob Manfred has a long history with Major League Baseball. And Major League Baseball has long tried to avoid letting its history weigh it down. In a wide-ranging interview at Sloan on Friday, one month into his tenure as league commissioner, Manfred sounded like a man trying to make sense of how to reform a game without hollowing it out.

A few days ago, Manfred said that there was a universe in which baseball could shave eight games off its regular-season schedule “sometime down the road.” A reduction in the current 162-game schedule could make the sport’s playoff timing a little more flexible, and might increase fan interest in each game. At Sloan, Manfred said he chose the 154-game mark because it would take the majors “back to a number that’s already in our record books.” Could he see MLB going even lower, to 150? No, because then “you’re going to go have a record book with 150, 154, 162 …” Only in baseball, a sport hallowed enough to get the Ken Burns treatment, could the record book be more important than the ledger. Integrity is paramount. (Or as Manfred, who has worked for the league for nearly two decades, put it when talking about whether to reform gambling laws around sports betting: “Integrity, it’s Rule One.”)

Yet this is a commissioner who clearly wants to find ways to change the game. Manfred has introduced rules to speed the pace of play, and said Friday he’s very happy with the replay system MLB added last year. He said that in the future — “past Rob Manfred” — the league could have a team outside North America, and before that, maybe even one in Mexico. Now that would be historic. — Chadwick Matlin



Friday, Feb 27, 2:20 p.m.

The panel: “Basketball Analytics: Push the Tempo”

The panelists: Shane Battier, Mike Zarren, Sue Bird, Mike D’Antoni, Pablo Torre

Are basketball teams now so saturated with data and analytics that it’s hard to use them for a competitive advantage?

Mike Zarren, assistant general manager for the Boston Celtics, raised an interesting point about what qualifies as analytics in an analytics age. “If I know how well a player slept last night, is that analytics?” The breadth of topics discussed — injuries, biometrics, pace, traditional positions, rest, incentives, shot selection, team chemistry — reveal what a truly broad spectrum of questions and answers fall under the umbrella of basketball analytics. However the field is defined, it all serves the same master: talent. Shane Battier, the poster boy for the adoption of analytic ideas at the player level, summed up the mission perfectly: “It’s about creating space to allow talent to do what they do.”

Zarren returned to a well-worn focus at this conference — communication of insights — and defended that arena as the place where a competitive advantage still exists: “You have to use it, it has to affect the decisions you make. I don’t think there is a saturation of that yet.” — Ian Levy


Friday, Feb 27, 12:21 p.m.

The Panel: “Valuing Franchises: How Sports Teams Break the DCF”

The Panelists: Lyle Ayes, Aswath Damodaran, Joe McNulty, Randy Vataha, Abe Madkour (moderator)

The recent sales of the Los Angeles Dodgers and Clippers for over $2 billion have opened up a new paradigm in sports franchise valuations. As shocking as the price of those transactions may have been, the mood at this Sloan panel was buoyant. In fact, panelists seemed to be most worried about prices getting so high that billionaires would be priced out of the market. As Lyle Ayes, managing director of the investment bank Evercore’s sports advisory practice said, “how many people can pay $4 billion for an asset?”

Panelists thought the seemingly inexorable rise in franchise valuations was driven by the increasing value of media and content rights. Aswath Damodaran, an NYU professor who focuses on valuation (and FiveThirtyEight contributor), commented that across the entertainment industry, owning content is becoming king. Ayes cited the NBA’s massive new TV deal as evidence of this trend. He noted that advertisers put a large premium on live content like sports because viewers are relatively captive during the event. Interestingly, none of the panel members thought that a team’s performance had a large impact on valuation. The most important factor, according to the panel, was metro area population and GDP. The New York Knicks can command significantly more from their local TV rights for bad basketball than the San Antonio Spurs can command for good.

Despite the increase in the real earnings of teams as media deals improve, panelists (with the exception of Ayes), broadly agreed that sports franchises still do not make sense as actual businesses. While they are relatively low-risk and uncorrelated with other potential investments, almost any analysis of the current cash flows — or lack thereof — will not find them to be great investments. As Damodaran noted, the supply of franchises is relatively fixed, while demand has been growing. The panelists did not see this dynamic changing any time soon. — John Ezekowitz


IMG_0029Friday, Feb 27, 11:55 a.m.

Daryl Morey has been as instrumental to the rise of the Sloan conference as he has been to the rise of the Houston Rockets. Morey, the general manager of the Rockets, has steered the team to third place in the Western conference — behind MVP-candidate James Harden, whom he acquired in a now-legendary 2012 trade — and helped start the Sloan conference in 2007. At Sloan on Friday, I boxed him out to ask a few questions about advanced basketball analytics, specifically player-tracking data from companies like STATS’ SportVU technology. While he can’t divulge the details of the Rockets’ private statistics, Morey’s remarks about the publicly available numbers are especially insightful because the Rockets are one of the most stats-savvy teams — not just in the NBA, but in all of sports. — Andrew Flowers


Friday, Feb 27, 10:40 a.m.

The panel: “Innovators and Adopters”

The panelists: Shane Battier, Michael Lewis, Daryl Morey, Jeff Van Gundy and Jackie MacMullan

Pity Carmelo Anthony and Kobe Bryant. Out for the season with injuries after performances well below their high standards, they’re now punching bags in Boston, at least according to the first session of the Sloan Sports Analytics Conference on Friday.

The other panelists treated the retired Battier more or less as Lewis depicted him in a New York Times Magazine article in 2009: the platonic ideal of the intelligent NBA player, one who incorporates insights from advanced statistical analysis to optimize his game for team success. (Battier initially resisted that framing, saying “it was about winning,” before eventually letting on that yeah, he was a pretty smart player.) LeBron James, with whom Battier won two titles in Miami, was the more typical player, open to occasional tastes of analytics-based tips.

Anthony and Bryant, though, were depicted as the anti-Battiers, in a question by moderator MacMullan (who, like Battier, works for ESPN, which owns this website and sponsors Sloan). MacMullan noted their selfishness and focus on scoring over other ways of contributing to their teams. (To which my boss, Nate Silver, would respond that Anthony’s shooting makes his teammates better.) Battier made clear how much he relished having those two stars as foils, learning their tendencies so that he could neutralize their strengths when playing defense. MacMullan pointed out that Battier blocked more of their shots than any other player’s. Anthony also topped the Battier leaderboards for balls stolen and offensive fouls drawn. And the pair led another personal leaderboard Battier innovated: They gave him the most “looks of disdain” when they found out he’d be guarding them. — Carl Bialik


CORRECTION (Feb 28, 9:47 a.m.): A previous version of this article misstated the ages of Sam Hafetz and Jonah White.

Filed under , , , ,

Comments Add Comment