There's been a proliferation of batter/pitcher confrontation results presented in the daily paper in an ongoing effort to provide the fans with seemingly relevant and important information. 16 for 38, 7 for 9, 0 for 3. What does it all mean? Aren't some of these cases just too small a sample to draw any kind of conclusions whatsoever?

Let's look at all batter/pitcher confrontations from 1999–2002. We'll exclude any PA with position players as pitchers, pitchers as batters, bunts, and IBB. We're going to split up the data into two sets, to determine if performances in past years can predict performances in a subsequent season. The 1999–2001 confrontations will be the before, and the 2002 confrontations will be the after. We'll ensure that the same confrontation pairs exist in both the before and after groups. This gives us 20,208 confrontation pairs: from the three PA confrontations between Brent Abernathy and Rolando Arrojo (two before, and one after), to the eleven before PA and three after PA of Todd Zeile and Masato Yoshii, and the 20,206 pairs in-between.

Let's concentrate on those confrontations with at least seventeen before PA and nine after PA. This reduces our data set to exactly three hundred pairs. This sample will be much more manageable, as well as being presumably more relevant. After all, not many people will argue that the two times that Abernathy/Arroyo faced each other would be predictive of the third time they faced each other. With our three hundred confrontation pairs, we get an average of twenty-two before PA and eleven after PA.

So, if a batter has faced a particular pitcher twenty-two times in the past, how predictive are the results of those twenty-two PA? In other words, do they tell us anything about how he is going to do the next time he faces that same pitcher? Let's see.

In terms of performance, the biggest mismatch was between Luis Gonzalez and Andy Ashby. How incredible was it? In eighteen PA, Gonzo had four HR, one triple, two doubles, two singles, one walk, and struck out once. That's an OBP of .556, and a SLG of 1.471! Even Barry Bonds would be jealous of numbers like that. Most people would say that Gonzo owns Ashby.

How about the flip-side? Let's take a look at Mike Mussina and Jason Varitek. One single, one walk, and seventeen outs (of which nine were strikeouts). We don't think we need to figure out Varitek's OBP and SLG.

So, we've got some really lopsided confrontations here. But, how significant are they? How much attention should we pay to these batter/pitcher confrontations? Are hitters who have shown a tendency to own pitchers (and vice versa) exhibiting some real trait? Is the identity of a pitcher crucial to a batter's performance (and vice versa)?

If this was real, if from 1999–2001, Luis Gonzalez really has Andy Ashby figured out, if Mike Mussina knows how to pitch to Jason Varitek, then we would expect these lopsided confrontations to continue at least somewhat in 2002, right?