Comments on: Historical Catcher Block Percentage

By: Dan Turkenkopf

Dan Turkenkopf — Mon, 24 Mar 2008 12:40:49 +0000

For the 52 catchers who had at least 200 opportunities combined from 2005-2007, the mean runs above average/100 opportunities (used to get everyone on the same scale) was 0.07 and the standard deviation was 1.1. There were only two catcher (Ivan Rodriguez and Doug Mirabelli) outside 2 standard deviations, and Pierzynski and Matheny were just less than 2 standard deviations away.

The correlation in runs above average/100 opps between catchers with over 100 opps in each of 2005-2006 was .37. In 2006-2007 it was .5. Between 2005 and 2007, it was .47.

I’ve forgotten most of the stats classes I’ve taken, so I’ll leave it to you to tell me what those mean.

By: MGL

MGL — Sat, 22 Mar 2008 10:18:51 +0000

One more thing. Any chance you can either compute a y-t-y correlation for players with a certain min number of opps, or a variance (or SD of course) for all players in the entire sample (again, with a min number of total opps), after combining each player’s 05, 06, and 07 data?

That way we can estimate how much to regress the sample data. IOW, how much of the spread we see in any given number of opps is random and how much is skill.

Thx.

By: MGL

MGL — Sat, 22 Mar 2008 10:14:34 +0000

Dan, thanks for the insight and I am glad that you are using only opportunities with men on base. I agree that probably not all catchers put as much effort into blocking balls with no one on base as they do with runners on, although they generally do put quite a bit of effort with the bases empty, otherwise they would tend to get lazy when there are men on base. It is just that some may be more lackadaisical than others with bases empty, which would screw up the numbers.

I have not checked if everything zero’s out, but the charts look more reasonably balanced now. Good work!

By: Dan Turkenkopf

Dan Turkenkopf — Fri, 21 Mar 2008 11:33:10 +0000

Ah. Found it. It’s actually due to the rounding error on the average BP. If remove the pretty formatting on the average BP, then things zero out correctly. I’ll update the charts to match the new information.

Thanks for calling my attention to this MGL.

By: Dan Turkenkopf

Dan Turkenkopf — Fri, 21 Mar 2008 11:24:04 +0000

MGL: The opportunities are only for runners on base, mainly for the reason that many catchers don’t try too hard to block balls when there’s no one one.

I am zeroing out each year. The average block percentage over all catchers is listed in the first table in the post. That is to say, it’s the aggregate block percentage for all of MLB. In 2005, it was 84%, in 2006, 77%, and in 2007, 86%.

I’ll be honest, I’m not sure why things aren’t summing to 0. I’m calculating blocks above average by (Opportunites * (1-Avg BP)) - Misses, which technically gives me misses below average. But that’s really the same thing as blocks above average.

Now that you’ve pointed out it doesn’t sum to 0, I feel like I’m missing something, but I’m not sure what it is. Any ideas?

By: MGL

MGL — Fri, 21 Mar 2008 02:20:00 +0000

First of all, are the opportunities with runners on base only? If not, you can’t use .27 runs per miss. You have to know what percentage of misses or opps are with runners on base.

Second of all, what are you using as the baseline (average) block percentage for each year? It looks like you are not zeroing out per year, which you want to do. You have got virtually everyone negative in 06 and a majority positive in 07. You want everyone to be a total of zero in each year when you think you have data consistency problems from year to year, which there appears to be.