Comments on: Guerra makes an impression

By: sane

sane — Wed, 12 Mar 2008 12:19:16 +0000

roundabout,
You are likely correct, but there is an alternative possibility.
Maybe the Cashman-Steinbrenner disagreement was internal and the discussion was what COULD BE offered for Santana, rather than what WAS offered to the Twins that Cashman wanted to retract.
Cashman’s statements would have been appropriate in either scenario.

By: roundabout

roundabout — Tue, 11 Mar 2008 22:53:44 +0000

Geno:
It’s old news…
Brian Cashman sure thought Phil Hughes was being offered in a possible trade for Santana. So was Kennedy, at one point, and Cabrera. The reason most knowledgeable people who can read knows this is because Cashman said he was opposed to trading Hughes, Kennedy and the other kids, several times.
BTW, how do you know Guerra is mature beyond his years?

By: USAFChief

USAFChief — Tue, 11 Mar 2008 18:53:07 +0000

That first sentence should read…”how the raw data used…is developed and then…”

Proofread, chief, proofread!

By: USAFChief

USAFChief — Tue, 11 Mar 2008 18:51:40 +0000

Cmath, do some research into how the data used for the “advanced” defensive metrics and then get back to us. Tell me now a guy can go from bad one year to all world the next (see Coco Crisp, for example).

As I said before, GIGO (garbage in, garbage out). BIS’s PBP info, for example, is derived from guys watching every major league game ON TELEVISION, and then trying to plot balls, including velocity and trajectory, from TV. Even if you could somehow do that, not every game is plotted by the same person, so the data is very flawed.

I would agree Bartlett was a pretty good shortstop defensively, btw. Shaky hands but good range. I also would agree Harris was a bad defensive shortstop.

Trying to turn that into a specific number of runs (which most of the current metrics try to do), and then making decisions based on that data, is where you lose me.

I also think you lose anyone in a decision making role in baseball. I seriouly doubt any GM or manager in baseball is making decision solely (or even in large part) based on UZR or plus/minus.

By: sane

sane — Tue, 11 Mar 2008 18:11:23 +0000

Cmathewson,
Your examples are confirmed by consensus AND metrics (although not as obvious as Mauer vs LeCroy).
I guess the real test of methods is an example of disagreement between metrics and consensus. I am trying to think of an example but I just haven’t paid enough attention to metric results.

By: cmathewson

cmathewson — Tue, 11 Mar 2008 17:31:34 +0000

I think you can get more fine-grained than that, sane. I do think that fielding metrics come with a large margin of error. So, for example, Everett and Bartlett are close enough to make it tough to say one is better than the other range-wise. The difference, then, is errors, which is a significant difference.

And I think you can say that Punto is a much better defender than Lamb at third, based entirely on metrics. But he’d have to be Brooks Robinson to overcome his offensive deficiencies relative to Lamb. And he’s not, despite DickNBert continually hyping his glove over there.