How To Run A Baseball Team, Part 2
Big Ball to Gain A Lead, Small Ball to Keep It
Avoiding so called ‘smart ball’ will score your team more runs throughout an entire season. This has been proven. However, the goal of a baseball team is not precisely to score more runs than your competition over a season, but to score more runs than your competition in each individual game. The invention of the Pythagorean theorem in baseball, despite all of its benefits, has muddled this simple and obvious baseball truth.
To take an extremely simplified situation, if a two-team league existed where one team scored 1000 runs in a season while the second scored 500, this would appear to be a mismatch. However, if the first team consistently won blowouts while the second team won many one-run games, this would be a close race indeed. Some in the sabermetric community take it for granted that winning percentage in one run games is based on luck, and while that is true most of the time, it is not universal. An effective and efficiently deployed bullpen will aid in one-run games, but that is a separate topic. As hard as this may be for some to read, strategic employment of small-ball can lead to greater success in close games.
Take another example: In the top of the eighth inning, your team is leading by one run. You have a runner on first base with nobody out. The average value of this situation is roughly 0.953 runs. A successful sacrifice bunt would reduce that value to 0.725 runs. Clearly, the right play is to swing away, right? Wrong. The average value of the first scenario, with a runner on first and nobody out, is inflated because of a higher probability of scoring a greater number of runs. The second scenario actually gives you a higher probability of scoring one run (23% to 17.6%). Winning by five runs is the same as winning by two, which is why in this scenario a higher probability of scoring one run is worth more than a higher probability of scoring multiple runs. This is reflected in the Win Expectancy Matrix. A Team in the first scenario (runner on first, nobody out) will win 66.4% of the time, a team in the second scenario (runner on second, one out) will win 70.2% of the time.
‘Big Ball’ is clearly the correct strategy to secure an early lead, but to help avoid blowing late leads you need to learn how and when to use certain strategies. For an organization to have a solid grasp of situational baseball, you need…
Statistically-Oriented Coaches
If I were running a baseball team, I would not want a coaching staff full of yes-men, planning strategy on which tactic would upset their boss the least. For a team to be managed effectively in-game, coaches must understand the madness behind the method.
Not only should coaches understand statistic principles, but they must also agree with them. If you have read ‘Moneyball’, you may remember a section of the book that highlights then infield coach Ron Washington complaining about the team’s unwillingness to run. That is unacceptable. For a coach to be on my staff, they must fit the system.
Leveraged Bullpen
Let’s call it a ‘leveraged bullpen’ since a ‘bullpen by committee’ suffers from such a negative connotation. Of course, bullpen by committees only fail because teams who resort to them only do so because their relievers are terrible anyway.
The idea of sending your best relief pitcher in only when you are ahead in the ninth inning is stupid. There are far more important situations in a baseball game. If you were a Yankee Fan, would you not cringe to see LaTroy Hawkins brought in to face Manny Ramirez late in a game with two men on base, all because Mariano Rivera is being saved in order to rack up an arbitrary statistic?
The Cleveland Indians - one of the best run organizations in basbeall - have found a happy medium. They bring Joe Borowski, a mediocre reliever, in to mop up the saves while their ace relievers like Rafael Betancourt and Rafael Perez are used flexibly as the situation calls. Your best relievers should be kept available for the most important situations, as defined by the Leverage Index.
Related posts
Comments
1 Comment »
Subscribe
Run expectancy is the way to go, but swinging away _could_ still be the right play in that situation. Maybe a successful bunt adds to win probability, but the average at bat of that particular player adds more (or the chance of bunt success could be less). Going by the “Book” sure gets complicated in a hurry late.