The relationship between balls and walks

One thing baseball announcers like to look at a lot is pitch counts, and the number of balls versus the number of strikes a pitcher has thrown during a game. Obviously the best pitchers throw less pitches, throw many more balls than strikes, and of course miss a lot of bats while keeping the ball on the ground.

The assumption would be that more balls translates into more walks, and in general this is obviously the case. However is this always the case? Do some pitchers have an ability to prevent more walks than their number of balls would suggest? To find this out, I decided to divide a pitchers balls by their walks aloud.

FanGraphs has ball/strike data going back to 2002, so I decided to evaluate data from 2002 to the present. In order to limit for sample size, I used all pitcher that have pitched 500 or more innings during that time. 167 pitchers qualified.

This is what I found:

From 2002-2009 the average Ball/Walk ratio (among pitchers I studied) was 17.89. In 2008, the average Ball/Walk ratio was 17.82.

There was a pretty large variance. The lowest number was under 9, the highest over 32. Most pitchers were clustered somewhere in the middle, with 81% of them between 14 and 21.

The pitchers clustered at the bottom are guys that are pretty universally described by one word: shaky. Oliver Perez, Victor Zambrano, Daniel Cabrera, Brandon Backe, and the list goes on and on.

 

The top of the list is where most of the finesse pitchers are. Brad Radke sits at the number one spot by a pretty confertable margin, with David Wells, Curt Schilling, Jon Lieber, Greg Maddux, Carlos Silva, Paul Byrd, Josh Towers, James Shields, Roy Halladay, and Mike Mussina occupying the top 11.

 

So far this does make some sense. The top pitchers are described as smart, finesse pitchers. They are human – they do miss sometimes – but when they fall behind in the count they lock down and they rarely ever walk anyone. At the bottom you have your shaky (and for the most part bad) pitchers that throw a lot of balls, but also walk a ton of guys, and have no ability to lock down and command their pitches.

Most of the good K pitchers are clustered in the middle of the pack. They don’t have as much command as the top guys, and they aren’t nearly as shaky as the bottom few.

Another observation I made is that some of the pitchers near the bottom of this list are actually pretty good. Scott Kazmir is at 14.36, Kerry Wood 14.43, Rich Harden 14.49, Matt Cain 14.6, and Derek Lowe 14.9.

Now this information isn’t all that useful. It basically confirms that BB rate is a completely luck independent stat. But we can put it to work for us in some cases. For example, Felix Hernandez.

2007: 53 walks, 1064 balls.
2008: 80 walks, 1172 balls.

Felix didn’t exactly lose the strike zone in 2008, he was just very shaky and inconsistent. There’s no reason to believe he wont revert back to his career B/BB ratio around 17.5% (almost league average). There is even a good chance at his age that he will improve that. Felix has plenty of upside where BB rate is concerned, just like he could put up a big K rate or GB%.

Matt Cain is also interesting, and highly confusing. Over his career he consistently is among the lowest ball counts in the league, but his BB rate is all over the place. While he is always a very inconsistent pitcher, there is some hope. His B/BB rate was around 15.65 in 2007 and around 14 in 2008. He has the ability to improve that BB rate over the next few years if he can mature.

A lot more interesting stuff here. For example, here is a graph detailing the relation ship between strike/ball ratio and ball/walk ratio (both are “better” high).

Here is the spreadsheet with 2002-2009 statistics on balls, strikes, and walks.

http://spreadsheets.google.com/pub?key=pkaci5b7xQ0iZohDUoBU7cQ

Overall the most important thing out of this is pretty simple. Just because you throw more balls, doesn’t mean you are going to walk more batters. There is a skill involved in preventing walks independent of pure control.