/cdn.vox-cdn.com/uploads/chorus_image/image/20753759/20130903_gav_sv5_016.0.jpg)
Joe Maddon's recent decisions to play Delmon Young (who hits lefties well but righties poorly) rather than Matt Joyce (who hits righties very well but lefties poorly) against right-handed pitchers have been puzzling. Luckily for all of us sitting alone in our mothers' basements, Zachary Levine of Baseball Prospectus was on hand to ask him the question directly. Read Maddon's entire response here.
Let me draw your attention to what Maddon said near the end of the quote:
I don't know if I'm right or wrong. I guess nine innings of baseball always show you that or not. But going into the game, I like what Delmon's done more recently.
I almost always give Maddon the benefit of the doubt, but there's a lot wrong with this statement. Put aside the fact that he claims to be making his decisions based on what players have done over the past few weeks (along with their mental place right now, which he admittedly might be privy to and which could make a real, quantifiable difference). Put aside that the expected difference between Young and Joyce based merely on their split tendencies and overall numbers is too large for all the myriad secondary considerations to really come into play. Instead, take exception to the lack of understanding or the misrepresentation of sample size and how it pertains to a decision making process.
Nine innings of baseball does not always show you that a manager is right or not. In fact it never does. Anything can happen in five plate appearances, or in one start. It might be a representation of current true talent level, astutely identified by an astute manager, or it might be dumb luck. That's why decision makers need to take the long view. They need to take a consistent approach and to trust what they know. Most of all, they need to properly self-evaluate. When a manager thinks he can gauge a single decision purely on whether or not it works, he's already lost half the battle.