Statophile, Volume 14 | MIT Sloan Sports Analytics Conference Edition

Dorkapalooza Edition

We had the opportunity to attend M.I.T.’s Sloan Sports Analytics Conference this year. The conference offered sessions for a wide range of sports, but we focused our attention on the basketball-related sessions. After 11 sessions over 10 hours (1 1/2 days), we attempted to summarize the highlights below. We have sectioned this post by session, so the reader is able to focus on specific interests. For more detail on sessions I was not able to attend, see the posts from the excellent group of TrueHoop bloggers here.

Panel: Birth to Stardom, Developing the Modern Athlete in 10,000 hours

Moderator: Malcolm Gladwell Panel: Daryl Morey (Houston Rockets GM), Jeff Van Gundy (ESPN Analyst), Justin Tuck (Defensive End, New York Giants), Mark Verstegen (Athlete’s Performance)

“There are very few flawless players [in the NBA]. LeBron probably comes the closest.” – Daryl Morey

Moderator Malcolm Gladwell led the session with the majority of the questions aimed at the “nature versus nurture” debate. Mr. Gladwell is famous for his “10,000 hour rule” from his best selling book “Outliers“. He argues in the book the key to success in any field is largely influenced by practicing/working for 10,000 hours. This theory prompted an interesting discussion – since many athletes are quite physically gifted. The debate became how much of their success is based on natural talents vs practice time.

One of the more interesting comments came from Gladwell himself:

A lot of what we call talent is the desire to practice.

What appears to be natural talent is often simply a result of years of hard work.

Former head coach Jeff Van Gundy was certainly the most entertaining and candid of the panelists. He made the point that you want all “coachable” players, but the reality is you will not likely win with a roster full of them. There just aren’t enough of these types of players with all the qualities you’re looking for: “(A player) can be soft, selfish and stupid. (I can deal) with one of those, but not two.” Later, when the name Bonzi Wells comes up, JVG adds: “oh, I should add ‘fat” to that list.”

Daryl Morey echoed JVG thoughts, arguing you need stars to win in the NBA. Having a bunch of coachable guys is “just enough to get beat”. In the discussion around “nature versus nature,” he commented that point guards need to have a high IQ, whereas as big man does not necessarily have to: “you can be dumb and big. Can’t be dumb and small.” All you need to teach a big man? “See shot, block shot.”

The discussion turned to where analytics (data driven) can miss big opportunities. Justin Tuck, a defensive end for the NY Giants, said: “there is not enough focus on ‘heart’ during the draft.” Tuck was (I believe) the 7th defensive end picked in the draft (overall the 74th pick in the 3rd round in 2005) and none of the other players are currently in the NFL. Why? He had a deep desire to play. He loved the game. Perhaps others were slightly faster or bigger, but he had that key intangible of “wanting it.”

One of my favourite quotes from the discussion came from Van Gundy: “to be the best, you need a bit of stubbornness.” Van Gundy also gave the example of two players he had coached – one went to Yale and one went to Virginia Union. He said that the one that went to Virginia Union was his “rocket scientist” – he never made a mistake on the court. The “rocket scientist”, of course, is our beloved Charles Oakley.

Van Gundy concluded that one of the main challenges is to figure out “when their ‘love of the game’ shuts off.” Often, he notes, the ‘love of the game” shuts off before their bodies do.

See more from TrueHoop: Malcolm Gladwell on the fundamental questions about player development

The Real Resons Behind Home Field Advantage

By: Tobias J. Moskoqitz, co-author of Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won

The home court advantage in basketball averages approximately three points per game. The home team wins 60.5% of the games in the NBA. But why?

Mr. Moskoqitz dispells a few myths:

1) Larger crowds lead to better performance by the home team and poorer performance by the road team.

  • The easiest measure to test this is free throw percentage (as its consistent). The results? The home team shoots 75.9 percent. The road team? You guessed it 75.9 percent.
  • But what about close games? Isn’t there more pressure on the road team? No, the percentages are still identical.

2) Teams win at home because the rigors of travel doom the visitors

  • Examining “same city” games, he found the home teams have the exact same advantage as they do in all other game they host.
  • “Controlling for the quality of the opponent, the San Antonio Spurs, for example, fare no better when they take puddle-jumpers to play the Dallas Mavericks and Houston Rockets than when they make longer trips to Boston, Toronto, and Miami. (He does note that scheduling, especially the high frequency of road back to backs does have an impact.

So what explains home court advantage?
Social influence on referees

  • While FT% percentages are the same, the frequency that the home team got the the line is higher – 1.0 to 1.5 times a game more. Is this just a matter of the home team being more aggressive?
  • To help answer this, the types of fouls and turnovers were examined – where a high degree of judgment is required. These include loose ball and offensive fouls.
  • The evidence? “Offensive and loose balls fouls go the home team’s way at twice the rate of other personal fouls.”
  • The authors also looked at fouls that are more valuable – ones that cause a change in possession. “These fouls are almost four times more likely to go the home team’s way than fouls that don’t cause a change in possession.”
  • The chance of a visitor being called for a travel is 15% higher than a player for a home team.

The author emphasizes they are convinced that the vast majority (if not all) officials are “upstanding professionals, uncorrupted and incorruptible, consciously doing their best to ensure fairness.” He cites several studies in the field of psychology and concludes: “when humans are faced with enormous pressure — say, making a crucial call with a rabid crowd yelling, taunting, and chanting a few feet away — it is natural to want to alleviate the pressure.” As well, “they may also be taking a cue from the crowd when trying to make the right call, especially in an uncertain situation.”

What Optical Tracking Data Says about NBA Field Goal Shooting

An interesting session from Sandy Weill. He presented initial observations from new data sets from Stats LLC. Stats LLC have put several cameras in three different areas and record all player movement as well as referee position. This allows for significantly improved analytics – including things like “how wide open was player x?”

Some initial observations:

  • Tight defense reduces FG% by around 12 percentage points.
  • Field goal percentage drops by 1% for every 1.5 feet from the basket.
  • A quick catch and shoot tends to result in a higher FG% – even with close defenders. He notes that perhaps there is a rhythm to shooting.
  • Shots on possession changes beginning with defensive rebound or turnover are still slightly better chances (FG% +3%) – even after controlling for shot type and distance.
  • Tip attempts have a 22% lower FG% than a jumper from the same distance.
  • A tightly contested 3-pointer is equal to a loosely contested 13 footer. A tightly covered 24-foot 3pt attempt is equal to an open 30-footer.
  • He also looked not just at the proximity of the primary defender, but the number of defenders. The second closest defender makes a difference too. (Me: Spacing and ball movement matters!)

See more from Queen City Hoops’ Brett Hainline: The “Data Holy Grail”

How Much Trouble is Early Foul Trouble?

By Philip Z. Maymin, Allan Maymin and Eugene Shen
Certainly this session piqued my interest, since I may have expressed some positive views on Amir Johnson the odd time in the past.

One interesting assumption, which I thought was a great rule of thumb, is that foul trouble = Q + 1. E.g. if the game is in the third quarter, 3 + 1 = 4 fouls would be considered foul trouble.
The three authors examined play-by-play data from 2006-2009. The focus was on the team’s starters as they’ll have the most impact on when to take them out.

Their conclusion: Coaches are usually right. Taking your starter out when they have foul trouble (as defined above) is usually the optimal strategy.
See more from CelticsHub‘s Brian Robb: The “Data Holy Grail”

Moral hazard in long-term guaranteed contracts: theory and evidence from the NBA

By Arup Sen and J. Bradford Rice

I’ve dubbed this session “Turkoglu Theory.”
The authors presented several great data points to confirm what many of us were fairly sure of: players exert much more effort in their “contract year” and subsequentially see performance drop after signing the big deal. #LikeABosh

“performance … in the final year of a multi-year deal is approximately 10% higher than in the year prior.”


“We find that relative to the year prior (i.e., the final year of the previous contract), player productivity falls by 17%, lending further support to the hypothesis of decreased effort subsequent to signing a new contract.”

Basketball Analytics

Moderator: Marc Stein (ESPN) Panel: Mark Cuban (Owner, Dallas Mavericks), Kevin Pritchard (Former GM of Portland Trailblazers), John Hollinger (ESPN), Mike Zarren, (Asst Exec Director of Basketball Operations, Boston Celtics)

This was the best session of the conference. Mark Cuban is quite open with many of the tools the Mavericks use – they certainly invest significant resources into analytics (and it’s paying off).

A few of the best nuggets by the panel:

  • Cuban: “Analytics is about risk management.” It helps GMs avoid the big blowup trades and/or free agent signings.
  • Cuban thinks mid season trades are easier as you know how the team is working or not working. Leading up to (and coming out of) the draft, scouts help you fall in love with players and you believe you have good team. Once the season gets going, the shortcomings become clear.
  • Cuban: “I think the first misunderstanding about trades is that the NBA is an efficient market for trades. It’s not.”
  • Prtichard: Regarding the draft: guards and wings get it earlier (learning the NBA game) while bigs take longer (uses Aldridge as an example). The challenge is to assemble a team and try to get everyone when they’re peaking.
  • Pritchard: “You’re not trying to beat the S&P 500 [like a portfolio manger], you’re trying to ‘win’ the S&P 500” re: compiling a team.
  • Zarren: Not enough people look at the variance of various stats. (E.g. a 20 ppg scorer that scores 30 half the games and 10 the other half is much different that one that scores 20 consistently.)
  • Cuban: also need to weigh the time of game and situation. (E.g. a player that scores frequently through 3 quarters, but disappears at crunch time.)
  • Cuban is a fan of Adjusted +/- for lineup building purposes.
  • Cuban: w/ basketball “there is no moneyball solution to say ‘the numbers say you have to draft this guy’”
  • Panelist: (forget who) estimated that 10 NBA teams are not doing analytics.
  • Cuban: “the worst position in the league is a 40 win team.” Hear, hear.
  • Cuban: Some GMs worried about what they are thought of in media (implying this may affect some trading decisions).
  • Audience member asked the panel to respond to a comment by Minnesota Timberwolves GM David Kahn: “Analytics are less important for rebuilding teams” … Cuban: “I’m glad they think that.”

New Sports Owners: The Challenges and Opportunities

Moderator: Jessica Gelman (VP of Customer Marketing, Kraft Sports Group) Panel: Wyc Grousbeck (Co-owner & CEO, Boston Celtics), Brian Burke (President and GM, Toronto Maple Leafs, Jeff Moorad (Owner, San Diego Padres), Joe Lacob (Owner, Golden State Warriors)

Having Bill Simmons on a panel always makes for an even more interesting session.
Key points:

  • Moorad: our goal is to “break even”. Feels that the asset will appreciate in value. (Me: scratching head)
  • Simmons: Three reasons why he thinks teams are purchased: 1) Owning is an investment. 2) Hero complex. White knight. 3) Ego thing. Attention.
  • Lacob: Disagree. It’s a dream. Fun business.
  • Lacob: Its really hard to get these deals done. Have to strike when situation comes up. So hard to value teams. But the opportunity to buy rarely comes up.
  • Grousbeck: 23 out of the 24 championship teams included a “top 50 all time” player and two other big all-stars. Thus, he endorsed the “big 3 concept” from the beginning.
  • Grousbeck: His plan started in 2003 to add a Kevin Garnett-type player (which happened to be Garnett!) 4 years later in order to win in 2010. [Liston’s note to Raptors: that’s a 5+ year plan. Copy it]
  • Lacob: corporate owners don’t work – they can get better returns elsewhere.
  • Lacob: On bloggers: “they are not real fans, because they don’t have season tickets.” (I’ve added it to my “Really Stupid Things New Owners Say” book)
  • Burke: During his interviews to owners over the years “There are only two hands on the steering wheel and they’re both mine.”
  • Burke: Once you approved the budget, “get out of the god damn way.”
  • Burke: “we’re in the game to have a parade” Re: not trading for experience and giving up youth. Also, second place is worthless.

A few takeaways:
I was a little shocked to hear successful businessmen try to convince themselves sports franchises are a good investment. Two of them essentially admitted they likely overpaid and these assets are “hard to value.” Mr. Moorad paid $500 million for the San Diego Padres and bragged about improving the club from a $18 million loss to breakeven. Certainly a nice achievement, but last time I checked paying $500 million to “hope to break even” (Moorad words) over the foreseeable future is not a sound investment. He added, we think the asset will appreciate. This will ONLY happen if you find another billionaire with an ego. Yes, some rich owners buy franchises simply as a hobby – to “have fun.” The challenge is you have a limited audience. Traditional investors are quite hard to come by. And Lacob even admitted, due to scarcity value of franchises being available for sale, you have to “strike” when they come available. Which means you have to overpay even further. There is potential for what’s known as the “winner’s curse“. Thus, you have to hope there continue to be plenty of billionaires looking for a money-losing hobby who will overpay even more. Maybe, but given the economic backdrop, how deep can the losses be for the “hobby” to be worth it?

Other notes:

  • Growth of the conference has been substantial: 1,500 attendees this year versus 1,000 a year ago.
  • Largest group? From Canadian Brock University in St. Catharines (47 people)
  • Attendees from various teams:

  • No one from the Toronto Raptors.

  • Best panelist (by far): Toronto Maple Leafs GM Brian Burke

Quote of the Conference?

Current Houston Rockets GM Daryl Morey on a pre-draft interview with Marcus Banks:
Mr. Morey: What do you really want to do with your life?
Marcus Banks: Be a male fashion model.

Sources: TrueHoop, MIT Sloan Sports Conference, Scorecasting.com
Questions? Email me: [email protected] or find me on Twitter.

One great event. One great charity. The Team Up Foundation’s one4one event featuring all the athletes from the Toronto Maple Leafs, Raptors, Marlies and FC all in one place. Enjoy cocktails, entertainment, food, and mixing & mingling with your Toronto teams at the hottest sports bar in North America! See www.liston.ca for details.

To Top