Be the House

Great article by A's assistant GM Paul DePodesta written during last season that was drawn to my attention by Management By Baseball. Lots of insight here related to innovation in business and the changing of paradigms. Fortunately for DePodesta major league baseball was, and in most cases still is, an industry loaded with low-hanging fruit for people who focus on outcomes and ask the naive questions as DePodesta puts it. My favorite quote is:

"These discoveries ranged from broad philosophical ideas, such as the fact that 90% of the player population in major league baseball is replaceable by someone who makes less to the very minute detail, such as pitch counts or control of the strike zone."

That first idea is perhaps the most liberating news for small market teams to come out of the "new knowledge".

Although I love the article, a couple points that should be noted that those not familiar with sabermetrics might get the wrong idea about. First, the idea of applying a Markov chain model to baseball, in order to derive values for baseball events given the 24 base-out situations, did not originate with DePodesta but dates back to the 1960s. Additionally, most of the crucial insights or "discoveries" that the A's have employed were well known to sabermetricians beginning in the mid 1980s (e.g. the non value of bunting and overvaluing of speed). Second, DePodesta and Beane started their retooling in 1999 and did not incorporate scouting until 2002 and they are now doing things the right way, but to be fair a good portion of their success is due to a trio of pitchers drafted under the old regime, Tim Hudson in 1997, Mark Mulder in 1998, and Barry Zito in 1999.

One other key insight in the article is the following quote:

"I was on a quest to find relevant relationships. Usually it wasn't as simple as 'if X then Y.' I was looking for probabilistic relationships. I christened the new model in the front office: 'be the house.' Every season we play 162 games. Individual players amass over 600 plate appearances. Starting pitchers face 1,000 hitters. We have plenty of sample size. I encouraged everyone to think of the house advantage in everything we did. We may not always be right but we'd be right a lot more often than we'd be wrong. In baseball, if you win about 60% of your games, you're probably in the playoffs."

It's hard to underestimate the importance of looking at baseball this way. Analyzing event data to create strategies that work more often than they don't and then applyng them over the long haul in a disciplined way will without a doubt lead to success given the sample sizes involved. However, this approach has been criticized by the likes of Joe Morgan as a strategy that may result in more wins in the regular season (and hence playoff births) but not in playoff series wins where the advantage - it is said by the critics - shifts to teams who can play "small ball". I'm never sure what to think of these assertions. While never having studied it, I would assume that the playoffs represent a constrained run environment and so strategies that maximize single runs at the cost of scoring in bunches like the bunt or stolen base would take on added importance. There certainly might be some merit to the critics argument.