Wednesday, May 08, 2013

What Data Scientists Can Learn From Moneyball

A colleague asked me about how to get started with data science and how to influence the organization to understand the analytics and utilize it in decision making.

You can get many answers to these questions just by watching Moneyball, the movie that put Data Science on the map. Starting with the basics, organizations need executive sponsors to recognize that utilizing data and analytics in decision making is a game changer. Billy Beane, GM of the Oakland A's and played by Brad Pitt is that type of sponsor. Second, you need talent and tools. Assistant GM Peter Brand, "Yale, Economics, Baseball" is what the team needed.

The movie doesn't go into the technology, but provides a view into the challenges and organizational changes both men faced in trying to implement their strategy. Below are some memorable quotes:
No. No. Baseball thinking is medieval. They are asking all the wrong questions. And if I say it to anybody, I'm-I'm ostracized. I'm-I'm-I'm a leper
Peter points out that the art of data science is to ask good questions. He also shows, and feels the difficulty explaining and selling analytics and data based conclusions with his peers.

Billy Beane: No. What's the problem?
John Poloni: Same as it's ever been. We've gotta replace these guys with what we have existing.
Billy Beane: No! What's the problem, Barry?
Scout Barry: We need three eight home runs, a hundred twenty R.B.I's and forty seven...
Billy Beane: ... We got to think differently.
Are you analysing the right metrics? It's not about replacing players or getting wins - it's about scoring runs. Again, if you're asking the wrong questions, it can bring you to the wrong conclusions. And more importantly, the whole organization needs to change the way it thinks
... Of the 20,000 notable players for us to consider, I believe that there is a championship team of twenty-five people that we can afford, because everyone else in baseball undervalues them. 
This is a classic, needle in a haystack data mining problem. It also shows that data scientists need to consider economic factors when performing their analytics, so in this example there are two contraints. It's not just twenty-five people that they can afford, they also must be talented and therefore must be undervalued by other teams.  
Major league baseball and it's fans they're gonna be more than happy to throw you and Google boy into the bus if you keep doing what you're doing here. You don't put a team together with a computer, Billy.
This is another reminder that change management is hard. There are a good number of people in the organization that are used to making decisions based on their intuition and experience. In some cases, they may be using data, but they may not be using the right analytics or asking the right questions. When the data scientist comes out with a new perspective, they will be the first person to challenge the analysis, followed by its messanger, and then finally the overall strategy.
I'm saying it doesn't matter what moves I make if you don't play the team they way they're designed to be played.
Bottom line is you can have a great sponsor, smart data scientists, and the right analysis but if line managers don't utilize the strategy, tools, and data provided to them then it is all for naught.


3 comments:

  1. You had me until you said, "...Moneyball the movie."

    The movie is nothing more than a pale shadow of the book. All of the real lessons are in the book, and if someone cannot take the time to read the book (along with Michael Lewis's epilog in the 2nd edition), then they really shouldn't be trying to talk about data science and how to make it work on organizations.

    ReplyDelete
    Replies
    1. Rick - Really good point. Thank you. I confess that I haven't read the book but will put it on the reading list. Also, a bit easier to fish out movie quotes.

      What I liked about the movie is seeing the human reactions, especially with the scouts.

      Delete
    2. The second edition of Moneyball has an epilog by Michael Lewis (the author) that brings up some excellent points about the human dimension. Again, read the book.

      Delete