Bill James says so.
Per Fay:
Reds Best Defense in NLStats guru Bill James was in the press box passing out his new book Gold Mine. By his analysis, the Reds had the best defense in the National League last year in terms of Defensive Runs Saved with 52 more than the average NL team.
"Imagination is more important than knowledge." -- Albert Einstein
And that's w/ a black hole for most of the season in CF and one at 3B for a good chunk of it!
Taveras was still a slightly above average defender in CF last year. Bad routes and all.Originally Posted by SirFelixCat
No way...
See, this is still a problem w/ defensive metrics, imo. We ALL saw w/ our own eyes that he was pretty wretched out there. Better than a grandma, but far from "above avg", even if the #'s say otherwise, no?
And let's face it, a full season of Stubbs/Dickerson in CF is far and away better than WT.
They are far and away better, but Taveras was still slightly above average. His speed made up for bad routes.
The question is what did we see exactly? We ALL saw some ugly looking defense at times, a guy who got mediocre jumps and took odd routes..
But don't mistake aesthetics for effectiveness - in either direction. The eyes are certainly valuable, but they can deceive. Seeing is one thing. Valuing what we've seen in comparison to an unseen average is quite another.
I don't understand why people talk about defensive metrics like they're voodoo. What they measure, while certainly not perfect, is pretty straight forward. Given where balls where hit, Taveras got to more of them than his peers did on average -- regardless of how pretty it looked. Our eyes can see him -- but they don't see "the average" -- and that's where our biases come in.
Last edited by RedsManRick; 03-18-2010 at 11:03 PM.
Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.
No.
See, this is the problem with "our own eyes." Our "own eyes" don't see every game in person or every play on the field. Our "own eyes" feeding into our "own brain" typically remember plays that are "exciting" rather than those that are indicative of good positioning, good first step, good reads and good routes to the ball for routine plays. Our "own eyes" tend to muddle all of the plays that look routine into the background while they recall the extreme fringe "highlights" and "boners."
I can remember Willy T being pretty terrible out there and joking with GAC and Chip in chat about it to no end. However, in the end, he probably _is_ average, as Doug pointed out, due to his speed.
And, no, this isn't to say that defensive stats are the end all, they're not. But they're LEAGUES better than "our own eyes" because at least they "see" and process every play, something our own eyes and brains simply cannot do.
Now, maybe if you're a major league scout or baseball professional, I'll believe you. Maybe if you watch 150 games a year at the ballpark for the past 20 years and have a thing for always watching the centerfielder to see his first step (which somewhat precludes watching other players) you might have a point. But guys (and gals!) like you and me watching on our TVs at home are getting a tiny portrait of the picture defense paints. Again, defensive stats DO NOT paint the whole picture but they paint a lot more of it than "our own eyes."
You note that Taveras is probably an average defender because of his speed. Do defensive metrics tell you this? Wouldn't you have to supplement the defensive metric with a personal observation?
I think a complete understanding requires both personal observation and statistical analysis. In the end, it's not about avoiding what your eyes tell you altogether, but not being tricked by what your eyes tell you. I think you point out how easily we can be tricked.
How do we know he's not Mel Torme?
Taveras may be average because of his speed but that should be a red flag to many people. In Taveras case, if you are constantly having to use your speed because of poor routs and bad jumps it eventually will catch up with you. If your talking about average CF's, give me someone who gets good jumps on the ball and takes good routes over a uber fast CF any day of the week.
Yep.
They tell me he's an average-ish defender.
I do and kind of think the same thing.Wouldn't you have to supplement the defensive metric with a personal observation?
Well, to be honest, my eyes tell me that Willy T was worse than Dickerson who was worse than Stubbs but all three were better than Freel. I honestly can't get a feel for where Corey Patterson fits in there but he was pretty good in the field too.
I'm not worried about a complete understanding because I don't think it's attainable. I'll settle for a "pretty good" understanding.I think a complete understanding requires both personal observation and statistical analysis. In the end, it's not about avoiding what your eyes tell you altogether,
I supplement my "own eyes" (because I watch as many baseball games as you or anyone else here) with defensive stats. Precisely what you said above.
Whenever the "I use my eyes" "argument" gets thrown around, it feels like a connotation is attached that anyone who mentions stats isn't using their eyes. I _AM_ using my eyes as much as I can by watching the games on TV. But when someone says "I use my eyes," why should I "de facto" trust that? I don't know what their eyes have seen. I don't know what skill they have in evaluating defensive baseball. To me, it seems like a dismissal at best.
The was another comment earlier today about how Jonny Gomes looked good in the outfield because he made a few tough catches. Well, tough catches aren't the entirety of defense. Good positioning, good reads and good routes yield infinitely better defense than replay worthy highlights.
It's not so much being tricked as much as it is the capacity of the human mind. How many defensive plays does an average MLB game have? 30? 40? Multiplied by 14-15 games a night for 500-ish events to remember per night? 3500 events per week, 14,000 a month 50-60,000 over the course of a season? The average person can remember and process all of those events? Heck, even WITNESS them?but not being tricked by what your eyes tell you. I think you point out how easily we can be tricked.
I think not.
For example, the only defensive memory I have of the aforementioned Jonny Gomes last year is him backing up a play on one of his first games after he came up. He ran all the way in and backed up 3b so nicely I almost spilled my drink (which would've been a shame.)
Great! He hustled and did a good job. But that one occurrence does not make him a good defender. It means he did a good job for that one play. Every bad defender makes a good play every now and then. Every good defender makes a bad play every now and then. It's all the plays in the middle that separate the good guys from the bad guys. The vast majority of those plays aren't especially memorable.
Regarding the limitations of "our eyes". I think one reason this conversation can get contentious at times is because people fail to distinguish between skill and performance. Our eyes are really good at identify ability. Scouts can tell really quickly how well a guy is able to read the ball off the bat and how his routes look. When you break it down to the component pieces, eyes can tell pretty quickly what a guy is capable of doing.
But performance is ability applied over time and context. And that's where our observation skills fall short. As Nate explained so well, we don't see every opportunity. We don't see every player. Our memories don't weight things fairly, preferring the exceptional to the routine. And we can't see how the value of those events that clearly. A botched running catch may be a lot more embarrassing to watch than the play where the guy simply pulled up short, but at the end of the day, they affected run scoring the same way. That calculus is really hard to do in our heads.
And despite the pithy convention that "defense doesn't slump", it does. Just a like a really good hitter can have an off year, so can a really good defender. And vice versa of course.
Defensive metrics have a long way to go. But just because their conclusions conflict with our existing assumptions doesn't make them any more (or less) accurate. There's a great Bill James quote, which I can't seem to find, to the effect of: "If a stat doesn't challenge any of your assumptions, then it's useless. If it doesn't confirm any of them, it's probably wrong." It's a good rule of thumb.
Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.
I wasn't accusing you of not using your eyes or anything of the sort. I think we're on the same page with everything you mentioned above, differences in wording aside. I was just trying to point out that there's room (and a need) for both forms of analysis since each type has limitations. I don't know if that was clear from your post.
You make an especially good point about the casual fan. Since our ability to make personal observations is so limited, reliance on those observations is very misleading.
How do we know he's not Mel Torme?
High five!
(can we get a high five emoticon?)
Posting Permissions
|
Board Moderators may, at their discretion and judgment, delete and/or edit any messages that violate any of the following guidelines: 1. Explicit references to alleged illegal or unlawful acts. 2. Graphic sexual descriptions. 3. Racial or ethnic slurs. 4. Use of edgy language (including masked profanity). 5. Direct personal attacks, flames, fights, trolling, baiting, name-calling, general nuisance, excessive player criticism or anything along those lines. 6. Posting spam. 7. Each person may have only one user account. It is fine to be critical here - that's what this board is for. But let's not beat a subject or a player to death, please. |
Reply With Quote