Posted by Sean Forman on January 25, 2011
I'm a busy guy, so I don't always read all of the blogs out there that you might think I do.
18 days ago Tangotiger posted a reader letter by Pete L (who appears to be Edgar Martinez's Rich Lederer and as best I can tell has never e-mailed me directly) and added comments that disagreed with our presentation of oWAR and dWAR. Inside the Book is in my RSS reader, but the title of the post was simply "Reader Mail of the Day:Fielding WAR for DH" so I'm sure I just skipped it unread. Today, Tangotiger e-mailed about the post. I'm happy to respond to the criticism of the stat, but I feel the following in Tango's blog casts me personally in a mildly negative light and that I should comment on that first.
I know that sometimes my peers have said that, that they would have preferred that my criticism had been addressed to them directly in private. I do that sometimes, but not very often, for a good reason: when I broadcast it here, I don’t have to repeat myself. Because if I’m thinking it, others are too.
And the first time I saw the presentation at B-R.com for what the reader below is talking about, I was bothered. I had meant to tell Sean, but I forgot. But, now this reader wrote a long and well-thought out letter. So, should I now simply have a three-way dialogue with the writer and Sean? Or, just be honest here, respond the reader, and let Sean address it as he sees fit? Even if my choice in this instance will be a poor one, overall, it’s a good way to do things. Most people will benefit, we all get to watch, and things are transparent. And that may spur even more discussion by readers here, similar to Sean’s “suggestions” page from a few years ago. Given a choice, I prefer to post out in the open, and I wish alot more people would do this.
First off, I've never asked Tom to make any criticism of me private. I know that he wasn't singling me out when writing this, but since he wrote it prefacing comments on me and me alone I hope others won't read it that way.
Second, we can't have a three-way dialogue when the third party is never invited to the discussion. I feel his preface implies that I was invited to join in, but I decided not to join and therefore was ducking the conversation.
<cheap shot>I'm a little sick of bloggers like Murray Chass and Tangotiger assuming that I read every they write and will respond immediately to what they have to say.</cheap shot>
If you want my response, take two minutes and send me an e-mail. Then give me one business day. I'm one of three employees running a company with seven separate websites, so I do not do B-R stuff every single day of the week.
I've only scanned the critiques given that there are 27 long comments on the material and I have alot of other things I have to do today.
Why is oWAR put together as batting + the position adjustment + replacement level adjustment?
Some of you may be aware that there is some dispute around the veracity of the defensive numbers here, at FanGraphs, and and voted by Gold Glove voters. oWAR is simply everything except defense. Assume everyone is an average fielder and compute their WAR that way. My sole intent in splitting things this way was to present a WAR value without the contentious defensive numbers because so many people (meaning those who aren't going to Inside the Book) seem willing to throw out the baby (WAR) with the bathwater (defensive numbers) over the defensive numbers.
I think WAR is a fantastic framework for looking at player value. But people many can't get over the defensive component, so oWAR just takes it out. And then dWAR is simply WAR - oWAR.
Is it wrong to present dWAR this way (or call it above replacement) when TZR is versus average rather than replacement level?
Tango himself has suggested that replacement level players are around league average defenders. So when you compare Ozzie Smith's defensive contributions above what is contributed by a replacement level shortstop it seems to make sense that using average as the baseline is appropriate and is still, in fact, an above replacement value.
Wow, are you really this oversensitive?
Given time, I'll more fully read and digest the critique laid out in TangoTiger's blog and if appropriate I'll make the necessary changes.