frirst!

By the way, I like that even though he acknowledged that Fielding Bible people didn't look at errors and errors, but threw them in anyway. Then he said that Young since Young had few errors and a good fielding percentage, he had one of the "best statistically measurable seasons".

Isn't including fielding percentage and errors double counting?

ranked first in balls out of zone;

Wait, i thought that was Tony Pena Sr in KC!!

Good points about Neyer's column not being consistant with his Fielding Bible ballot. Neyer needs to be more careful and avoid such gaffes in the future.

Isn't including fielding percentage and errors double counting?

No more so than taking different opinions, each based on the same stats.

Let's take a lot of the stats you guys use in the Fielding Bible such as plus/minus; balls out of zone, zone rating and revised zone rating. (NOTE TO READERS: You can get definitions of all these measurements at the Fielding Bible and Hardball Times websites.). But, for old-times sake, let's include fielding percentage and errors because they may not be the most definitive stats out there, but they certainly can lend something to the conversation.

So, there were nine guys who started at least 90 games at short in the AL. I assigned each guy a point value (one being the best, nine the worst) for where he ranked in comparison only to the other nine in each category.

Two problems with this setup:

1) The categories aren't orthogonal. Zone rating and revised zone rating try to measure the same things, which aren't that different from fielding percentage tries to measure. Similarly, plus/minus and balls out of zone are treading on each other's turf. Notice that Young finished first in fielding percentage, fewest errors, and revised zone rating. Is there really enough separation between these categories that they should make up three out of six categories?

2) You should really compare players to the full spectrum of players at that position, not just the nine players who you consider candidates for the award. Being ninth out of nine candidates may not seem like a killer, but you're throwing out the possibility that one of your candidates may actually be much lower.

Orthogonal.

The one thinks that strikes me as silly is including both fielding percentage and errors. That is like saying one category (errors) has twice the weight as any other stat. (If one had to assign weights in this exercise, it should probably have lower weight than the range based methods.) If it is acceptable to throw the kitchen sink into the voting tally, then why not include PMR, DER, UZR, BP's defensive metric, etc. I don't know how Out of Zone plays affected his rankings, but it should probably be converted to a rate stat (e.g., OOZ/inning) before it is used for ranking.

It's hard for me to read so much vitriol spewed at Neyer because I love the man for some reason. When I think about it though, he very rarely says anything interesting, and I don't know how much he really adds to the conversation.

Still, considering the comment it is responding to (Neyer saying he gives the edge to Overbay over Pena because their seasons were comparable and Overbay has a longer track record of success), this is very stupid:

Oh, and one other thing: I think the awards are presented for single-season accomplishment, not for career achievement. What Overbay and Pena did in the past shouldn't be relevant to this year's awards. That would be as stupid as giving Mike Mussina a Gold Glove over Kenny Rogers because Mussina has a good reputation as a fielder and won a lot of games this year.

and so is this:

I'm not sure what the discussion about where Young will play in the future has to do with how he played in 2008. But, hey, it's your blog. The only thing that winning the Gold Glove in 2008 means to where Young plays in the future is that he will play short in 2009. Beyond that, it's wide open.

You really can't be so mean unless you're very funny, and this guy is very not. That said, he makes some valid points (especially that Neyer seems to have forgotten who he actually voted for--though it is a blog entry of course). Also that it seems to be a weak year for AL SS defensively. But he acts like what you see when watching the player is totally irrelevant, or at least that talking about it is "just opinion," i.e. no one's is more valuable than another's. Well, what about the fact that everybody who watches the guy play thinks he sucks, except, apparently, the mangers? And what about the fact that errors and fld% are super irrelevant? (Maybe the other stats are too; I dunno, but these definitely are).

The one thinks that strikes me as silly is including both fielding percentage and errors. That is like saying one category (errors) has twice the weight as any other stat.

Well. I can see an argument for it. Note I'm not making the argument, but I think I can see, if I were in a debate and had to be on the side of "include both FP and errors", this is the way I'd go:

If anything, modern fielding metrics have shown that the disparity amongst fielders in the major leagues is not that wide. There's not anywhere near the kind of variance between shortstops fielding as there is between shortstops hitting, and that's not even counting shortstops vs. everyone else hitting. The error rate in the majors has consistently dropped over time as well, perhaps indicating that defense, more than any other baseball skill, is going through Darwinian evolution (see Stephen Jay Gould's piece on this.) These two things being the case leads us to posit that fielding "true talent" (the disparity of it, which is what we're talking about) matters less and less as other components (like concentration, focus, clutchness) weigh more.

True talent mattering less is important. It does so because when everyone regresses to the mean in one aspect of the game, unless that aspect becomes more important in the framework of the scoring environment, disparities are worth less in the end equation of relating runs scored/saved to wins. If the best fielder can reach 90 balls that the worst can't in one context but in another context the disparity is only 45, for the latter to mean as much as the former would require that saving each of those 45 balls was worth twice as much in its context as the 90 were in theirs. That's not true, and we know it.

So we know that as talent levels shrink towards each other, the value of a positive disparity (or negative value of a negative) in true talent means less. We've just agreed that fielding isn't becoming *more* important in the overall context, so we'd have to agree that it hasn't become *less* important, right? Well, to keep it the same amount of importance, we have to weight something other than true talent more than we used to so that we end with the same value in an expected value format. What is there? The aforementioned other components. And what's the best way to measure those components like how much you're paying attention? Errors. If you and I have the same talent but you screw up twice as much as me, we should measure your screwups. And so, we have to weight errors more, meaning that it's completely justifiable to count both errors and fielding percentage.

(Again, I don't believe much of the reasoning in the above.)

Isn't including fielding percentage and errors double counting?

No more so than taking different opinions, each based on the same stats.

See, the issue here is that they're the same thing - one as a counting stat, one as a rate. It'd be like if I came up with a ranking system of evaluating hitters based upon, say, hits, batting average, OPS, home runs and RBIs. Hits and batting average don't just measure similar things, they measure the same thing, just expressed differently. It's the same here.

See, the issue here is that they're the same thing - one as a counting stat, one as a rate. It'd be like if I came up with a ranking system of evaluating hitters based upon, say, hits, batting average, OPS, home runs and RBIs. Hits and batting average don't just measure similar things, they measure the same thing, just expressed differently. It's the same here.

Nope it's actually worse here. Hits and BA do more or less measure the same thing (which is dumb), but at least in this situation you are benefitting the players that had more AB's (e.g. if too guys both have a BA of .300, the guy that played 500 innings will have a better score, than the guy that played 400). There is at least some merit to that kind of thinking.

But counting fielding percentage and errors actually penalizes players for having more playing time. That is just insanely stupid.

I also want to point out that I disagree with everything written in #9. Especially the bit where he weites "We've just agreed...". I hate it when people do sh!t like that.

See, the issue here is that they're the same thing - one as a counting stat, one as a rate.

My point is that if one person says "I trust the Plus/Minus numbers more than the others, and they say X is the best, so I think X is the best", and another person says, "I don't understand any of the fielding numbers, but I trust John Dewan, and his numbers say X is the best, so I think X is the best," and you take both opinions and count them equally, you're essentially double-counting Plus/Minus. So an approach like...

The Fielding Bible polls 10 guys for their best players at each position. You then give guys points for how high they rank on each voters list and compile the total. Guy with the most points wins your award.

...might be double-counting in much the same way as using fielding percentage and errors as this writer does.

I'm not advocating using fielding percentage nor errors, FWIW.

I also want to point out that I disagree with everything written in #9. Especially the bit where he weites "We've just agreed...". I hate it when people do sh!t like that.

Teehee. Honestly, as I wrote in the Lounge, I went into some sort of fugue due to lack of sleep. I had a brief comment to make, I started watching the game and not paying a ton of attention to what I wrote, and out of nowhere, there were three paragraphs. That said, while reading it again it doesn't strike me as a ridiculous argument. It's got holes, no doubt, some more obvious than others, but it's defensible as a reason to include both. I certainly think it's a more defined and reasonable (however reasonable you may think it is) rationale than Evan Grant is likely to provide. I grew up in Dallas, and my mom still lives there. Evan's always been the redheaded stepchild of the Metroplex sportswriter scene to me.

Neyer stoops to giving props to an Angel and a former Angel to dig Michael Young's grave and I still love pointing out how Neyer insisted before the 2007 season that Michael Young was the most interesting player in the American League West. I really wish the BBWAA would let Neyer in as a member so the terrible writers could again all be lumped together.

Oh, snap! You done told him!

Man, remind me never to get on Halofan's bad-side

If you're a sportswriter, a media figure of any kind, a sabermetrician, a non-sabermetrician, or are in any way related to or sympathetic to the Red Sox, Yankees, Rangers, A's, or Mariners, you're already on Halofan's bad side.