March 6, 2010

MIT Sloan SAC, Limits of Moneyball

This should be fun and informative. Michael Lewis, author of Moneyball, moderates Mark Cuban of the Dallas Mavericks, Jonathan Kraft of the Kraft Group, Daryl Morey of the Houston Rockets, Bill Polian of the Indianapolis Colts and Bill Simmons of ESPN discuss the decision making process behind the numbers.

Update: Michael Lewis wanted to put together a panel to argue against Moneyball, since everyone here seems to be of the same mind, that they accept it. Michael likes the book the way it is, so he wants to find out what others think is wrong with the ideas in the book.

Update: Lewis asks Bill Simmons what he doesn’t like about the people in this room, the geekification of sports.

Update: Bill wants people to make the formulas accessible to every day people. Division throws him.

By the way, this was one of Bill James strengths.

Update: Mark Cuban is next. He wants more, Daryl won’t tell him his formulas.

Update: Kraft deploys the mental toughness vs. stats tripe. He thinks statistical analysis is more useful when the salary cap comes into play. Kraft says the subjective parts of football analysis is more important.

Update: Polian disagrees to an extent. Personnel decisions have to bring in undervalued assets, especially under the salary cap. They do use statistics to decide which pool of players get on the board.

He says game management doesn’t use the player statistics.

Update: There’s a feeling in baseball that the more a front office uses stats, the more they try to hide it. Lewis says Morey is the only person he met in sports who doesn’t try to hide it.

Update: What are the risks of going down the stats path too far? Morey says front offices might follow these things too blindly, like what happened on Wall Street. Morey says if the Mavericks won the NBA title, it might bring down the league office. 🙂

Update: Bill Simmons wants to know what happens when all the teams have all the same information. Cuban says there’s a lot of off field statistics that effect a team.

Update: Kraft thinks there’s no ceiling on information. You need players who fit your system, you need to sign them at a reasonable price, and you need to coach them. There’s too many moving pieces.

Update: Polian talks about the need to bring in young people with new ideas, otherwise, you end up in a cocoon.

Update: Cuban says that he can tell by the lineup a basketball team puts on the court how statistically informed they are.

Update: Morey can tell the same thing by how a team plays defense, by what shots they are trying to take away.

Update: Polian says coaches are ahead of front offices in statistical analysis in football, due to having 22 cameras, one on each player for years.

Update: Kraft just said some of his head coaches were smarter than others. 🙂

Update: They’re discussing the fourth down play, Pats vs. Colts. Polian is giving all the reasons on that particular play the overall odds were wrong, including injuries and fatigue.

Update: Polian still thinks it was the right decision from a football standpoint.

Update: The panel was just asked about what they can exploit with statistics, and refs came up. Mark noted that there was a ref who didn’t call the 3-second violation, so the Mavs took advantage of that.

Update: Cuban is talking about protecting morons so you can dump them on other teams. Mark said maybe they all should just be outed to improve the NBA. The Mavs do a lot of psychological evaluations to see if a player will fit on the team.

Update: Both Cuban and Kraft are talking about evaluating how players will handle in-game stress.

Update: Matthew Glidden also tweeting conference.

Update: Morey won’t pay more for a clutch player. Mark Cuban will.

Moneyball Panel

Longshot of Moneyball panel. That's the back of Mark Cuban's head, and Lewis is reading a piece of paper.

Update: Bill Simmons wonders if owners are star struck, hiring former players to be GMs. Cuban says no. The team as an asset is too valuable.

Update: Polian says the NFL should use former scouts as replay officials since they are used to looking at video. NFL officials are used to looking at live action.

Update: Why are teams bad long term? Cuban says the Pirates owners make a lot of money getting yelled at by a city, so why do they need to win?

Update: Polian is talking about how the Red Sox and Patriots meld all the talents of their front office people to create a great product for the fans, from parking to the team on the field. Some organizations are waring internally.

Update: That’s it, another great panel.

2 thoughts on “MIT Sloan SAC, Limits of Moneyball

  1. bureaucratist

    Wow, this sounds super interesting, I wish I were there. A few comments:

    1. The Moneyball lesson has been widely misunderstood, in my opinion, even on this panel, it sounds like. It is not the stats trump received wisdom/anecdotes/impressions line. It is that teams at severe financial disadvantage will have to find what the market undervalues in order to be successful. Ten years ago that may have been OBP (or whatever). It no longer is. The disadvantaged teams will now have to find another way to recognize undervalued commodities in order to succeed THAT MAY OR MAY NOT have the same kind of heavily mathematized characteristics.

    2. The geekification of formulas: I have enough of a mathematical background to understand most fangraphian analysis, broadly, at least, if not specifically. Even if you make it accessible to everyone, however, it is still a real issue as to whether it improves understanding and enjoyment the way the game is actually experienced. Example: One of the first sabrmetric stats I learned was runs created per game. Intuitively, very easy to understand: The number of runs a lineup of nine identical players would score, based on the given player’s stats. Nothing geekified there. But it still doesn’t contribute to the actual experience of baseball. There’s no way to apply that concept to bottom of the fifth, one out, down 4-2, runners on first and third, Placido Polanco at the plate batting seventh. It doesn’t match up in any intuitive, natural way with the experience of the game, even though it is distinctly ungeekified yet mathematically sophisticated. This is my primary objection not to the sabrmetricization of baseball, but to its OVER-sabrmetricization.

    3. Mental toughness is tripe, with mental toughness standing in for a whole host of other de-metricized characteristics: There’s a sort of ad hominem argument to be made against the idea that mental toughness is a load of hooey, and it goes like this: Statistical analysis shows that, given the sheer weight of enough numbers subject to enough analysis, mental toughness/clutch performance are revealed as mirages. However (and here’s the ad hominem part), given the sheer weight of folks on the other side of this argument, folks who have often devoted their very lives to questions like these, who will go to their graves defending mental toughness/clutch play as crucial to an understanding of their games, the weight of the numbers on the other side might cause a devotee of numbers to wonder whether there is something his/er analysis just hasn’t picked up. Physicists have yet to discover gravity waves; but everyone is pretty sure that gravity is still there.

    ReplyReply
  2. soccer dad

    This is kind of a surprise. Before the Super Bowl there was some discussion of how the Colts snapped up shorter players, thought to be too small for the NFL.

    http://online.wsj.com/article/SB10001424052748704259304575043380897055398.html

    And here’s a similar article from a couple of years ago.
    http://online.wsj.com/article/SB120915463828645643.html

    In other words, Polian has taken a Moneyball approach – finding the undervalued talent – but didn’t brag about it in a perfect forum. Maybe he didn’t want to give away a secret.

    Also it’s interesting that Polian points to the multidisciplinary approach taken by the Red Sox (and Colts) as a reason for success. This is something I’d actually credit to John Hart. In Cleveland he was the guy who assembled a front office out of people from many different disciplines. Theo Epstein copied the idea, but Cleveland did it first.

    ReplyReply

Leave a Reply

Your email address will not be published. Required fields are marked *