Colin Wyers of BP on advanced fielding metrics
by limozeen on Nov 11, 2010 3:12 AM EST
Email
Printhttp://www.baseballprospectus.com/article.php?articleid=12433
It's probably worth the subscription to read this article alone. Wyers's basic premise is that sabermetrics is about the search for reproducible truth. Defensive statistics (even ones that use the same source data) do not correlate to each other well. According to Wyers, we have fundamentally abandoned the premise of sabermetrics in an attempt to create workable defensive stats.
The conclusion: "The AL Gold Glove voters made a mistake in giving Jeter the award. But I think we make a much bigger mistake if we castigate the Gold Glove voters for their beliefs without a serious effort to give them something in which they ought to believe."
This is really worth a read; it's perhaps the best attempt at self-criticism I've seen in baseball research in a long time.
7 comments
|
0 recs |
Do you like this story?
Comments
I think Werth is "shock jocking" a bit.
Defensive statistics (even ones that use the same source data) do not correlate to each other well.
Really? +/-, UZR, PMR, etc don’t correlate well to each other? I think they do just fine most of the time. You’re going to get your wonky disagreements every so often, but for the most part they’re in general alignment. For example, guess which gold glove winning short stop all of them think sucks?
According to Wyers, we have fundamentally abandoned the premise of sabermetrics in an attempt to create workable defensive stats.
Horseshit.
I don’t have a BP subscription to read the whole thing, but if he really thinks that, then he’s nuts.
Fans are typically idiots.
by The Typical Idiot Fan on Nov 11, 2010 8:00 AM EST reply actions
Re
I think I know Colin Wyers reasonably well, as in, he is someone I’ve talked to a ton over the internet. I think you need to take that statement about abandoning the premise of sabermetrics in the context of his body of work. I don’t have time to read the piece until later tonight, if that, but I believe he means that he means there has been a general lack of objectivity in the creation of these metrics.
Normally, saberists don’t make many assumptions. They take data and they manipulate it. Or they collect data and then manipulate it. In today’s defensive metrics, we take a lot of subjective data and use it (and more importantly, ARGUE THAT IT IS) objective. See, Pedrophile schooling me where I was totally wrong a few months ago on this topic.
Colin has consistently fought against the saber community relying on stringer data, BIS Data, and STATS data as the primary input to our defensive metrics. If his point of view is that we should be more critical of this data, therefore we need to be more critical of the statistics this data creates. We aren’t – because we want so badly to create these metrics – we’ve lost our way.
I for one, should have been initially converted by Perdo, but shortly after was converted by Colin.
Come check out Bullpen Banter!
Follow Bullpen Banter on Twitter
Follow me on Twitter
Remember: baseball guys... baseball...
Some thoughts on your post
Paragraph 2 – This is where I think a lot of quantitative baseball analysis falls short – it becomes a data mining exercise that drifts away from being grounded some research and modeling fundamentals – hypothesis testing, model validation, etc.
Paragraph 3 – Yes, the “garbage in, garbage out” point that is also a bedrock concern in research and modeling.
I think the leaders in sabermetrics/baseball research are sharp minds with good understanding of these things, and able to cross over or pull in concepts from more mature fields, but there are also a lot of folks learning on the fly. The consumers of this info and analysis are left trying to sort the wheat from the chaff, without maybe being aware of some underlying weaknesses.
Footnote: in general, I find the BP people are pretty good about self analysis. I found that to be one of their strengths during the Silver era, and it sounds like it may still be the case. I let my subscription lapse, but not because I didn’t appreciate their efforts and transparency.
RE
Agreed: I think Colin is suggesting that people who understand those fundamental principles are not adhering to them (intentionally, or unintentionally) because of the allure of Fielding Metrics.
Good points.
Come check out Bullpen Banter!
Follow Bullpen Banter on Twitter
Follow me on Twitter
Remember: baseball guys... baseball...
RE
I’ll be the first to admit when I’m wrong. You were dead on then.
Come check out Bullpen Banter!
Follow Bullpen Banter on Twitter
Follow me on Twitter
Remember: baseball guys... baseball...
On a related note
here’s a great article I found over at Beyond the Boxscore about methods in prospect rankings.
When you're drowning, you don't say 'I would be incredibly pleased if someone would have the foresight to notice me drowning and come and help me,' you just scream.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
