May 11, 2000

The Daily Prospectus

Something Funny in the NL

by Joe Sheehan

A couple of weeks ago, I talked about the media uprising concerning the number of home runs being hit. At the time, there was a growing clamor that "something" had to be done, with the two most-common remedies bandied about being altering the baseball and raising the mound.

What I believed, and still do, was that the increased offense wasn't as bad a thing as the mainstream media would have you believe, and that three weeks in April were too small a sample on which to build a case, anyway. Yes, this is a good hitters' era, but panicking because the Cardinals were having a good month was a bit premature.

A few more weeks have passed, and with some of the furor having died down, I wanted to revisit my own dog in this fight, the changed strike zone. I firmly believe that the simplest and least-intrusive method of tweaking the balance between hitters and pitchers is to find some way of calling a reasonable strike zone.

The zone called by most umpires appears to be far too small vertically and has been shrinking for some time. In general, it extends roughly from just above the knees to somewhere between the belt and the bottom of the ribcage. While there is also a trend towards calling a strike zone wider than home plate, especially on the outside part of the zone, there isn't any consistency in doing so from umpire to umpire.

To me, there's a clear line between a smaller vertical strike zone and increased offense. Obviously, a smaller target is harder to hit, leading to more balls, more hitters' counts and more walks. Taking away both the high and the low strike lets hitters focus on a smaller area, one that corresponds more closely with their "wheelhouse" than the book strike zone.

But is any of this happening? Does it show up in the statistics? Just as a starting point, I looked at the two leagues' walk and strikeout rates and strikeout-to-walk ratios since 1996:

League     K/9   BB/9   K/BB

NL 2000   6.60   4.07   1.62
NL 1999   6.67   3.74   1.79
NL 1998   6.76   3.37   2.00
NL 1997   6.81   3.42   1.99
NL 1996   6.77   3.33   2.03

AL 2000   6.23   3.58   1.74
AL 1999   6.26   3.72   1.68
AL 1998   6.43   3.45   1.87
AL 1997   6.51   3.55   1.84
AL 1996   6.24   3.81   1.64

This year's data is through Tuesday.

The American League is all over the place, with a decline in walks and a jump in strikeouts in 1997 and 1998, followed by a regression in 1999. I can't conclude anything based on this other than that Roger Clemens was really good in 1997 and 1998.

In the National League, though, there is a demarcation between the 1998 and 1999 seasons. The league walk rate jumped 11%, with a corresponding decline in strikeout-to-walk ratio. So far this season, the league walk rate is up another 9%, for a jump of 21% in two seasons.

There are some problems with this quick and dirty study:

1. Thanks to Bud Selig and his long nights in the lab, we're not dealing with stable data sets. Baseball expanded in 1998, and the Milwaukee Brewers moved from the American League to the National League, so the changes from 1997 to 1998 may be masked somewhat by that. Additionally, the parks have been changing almost constantly during this time, and that may be clouding the data.

2. I haven't accounted for intentional walks. I don't know what impact, if any, doing so would have.

3. The 2000 data is based on just six weeks of baseball.

I'm not presenting this as a final verdict. There are a lot of ways to look at this, and I don't pretend this is exhaustive. But when a league's walk rate jumps this much in just two years, something is fishy. And with the strike zone at the heart of other issues, it seems to me that this is a good starting point for further analysis.

Joe Sheehan can be reached at jsheehan@baseballprospectus.com.

Joe Sheehan is an author of Baseball Prospectus. 
Click here to see Joe's other articles. You can contact Joe by clicking here