Nate McLouth's Defense

by Charlie Wilmoth on Aug 8, 2008 3:22 PM EDT

There's been a lot of talk today--here at Bucs Dugout, at the PBC Blog, and at WHYGAVS, about this article at Baseball Think Factory that says that Nate McLouth has been fielding "terribly" this year.

Baseball Think Factory is a great site, but I'd take all this with a grain of salt. Fielding stats are subject to all kinds of year-to-year fluctuations, and this one only represents a bit more than a half year's worth of play.

No methodology is listed in the article. I'm not questioning BBTF's methodology, but I'd love to see it, if anyone knows where to find it and can send it along, because there's a lot of stuff here that just doesn't make a lot of sense to me. The numbers suggest that:

1) That the no-kneed Jody Gerut was the top CF in the NL, and 22 runs better than McLouth, despite only playing half the season;

2) That Pat Burrell and Juan Pierre have been the best LFs in the league;

3) That all the catchers in the league except Geovany Soto were within one run of being average (wouldn’t a few plays at the plate, all by themselves, make the range between the best and merely average greater than one run?);

4) That the no-longer-mobile Brian Giles has been an excellent right fielder; and

5) That Luis Rivas has been an above-average shortstop.

I'll say that again: This metric thinks that Luis Rivas has been an above-average shortstop this year. That's insanity. So let's take all this with a grain of salt, shall we? I think WHYGAVS' guess that he's "probably slightly below average" is probably closer to the truth. I've watched Jim Edmonds quite a lot this year, and while I think the Post-Gazette's "trust your eyes" approach only goes so far, any system that says McLouth has been eleven runs worse than Edmonds has some explaining to do.

UPDATE: I'm leaving the above intact, but it's more strongly worded than it should have been. I didn't mean to suggest that the metric has no value, or anything like that, only that its stranger findings probably aren't anything to get up in arms about.

-P- Credit where credit's due: this article about the trades by Jerry Crasnick at ESPN is really quite good. It gives Neal Huntington a fair shake, accurately describes fan attitudes in Pittsburgh, and even mentions that while many in the mainstream media panned the trades, "some bloggers rode to Huntington's defense." Giddyup!

Tweet Comment 11 comments | 0 recs |

Do you like this story?

Read More: nate mclouth, neal huntington

Comments

Methodology

Link.

by Vlad on Aug 8, 2008 3:32 PM EDT reply actions

McLouth

I think the use of the word “terribly” is poorly chosen by the author. Having said that Charlie, I think you cherry-pick some of the players that seem to support the idea that the data is flawed. If you look at the big picture, the players who are perceived to be excellent defensively generally grade out very well, while those that aren’t, don’t.

I do agree that on a year-to-year basis we can definitely see pretty big fluctuations in one player’s level. Being Nate’s first year as a full time starter in center I’m willing to agree that their might not be enough data to get a clear picture of Nate’s abilities. However, I don’t want to discount it, and I certainly don’t buy into the “trust your eyes” argument that many scounts and numberphobes have used for years in discounting statistical analysis and sabermetrics.

No question Nate plays the game hard. He is fundamentally sound. He does hit the cutoff man, throw to the right base and do many of the other things that we view as “playing the game the right way.” However, errors is a tremendously flawed stat and one I think is generally not useful in this discussion. And we all know that Nate has never projected to be a center fielder. I don’t think this takes away from the excellent season he is having or says that he isn’t a very good player. I think it tells us something that we all inherently know. Nate is far and away our best alternative in center at the moment, but he is playing out of position.

by David Todd on Aug 8, 2008 3:51 PM EDT reply actions

Yeah, I do cherry-pick them, to show that I think this system can make “mistakes.” And I agree that errors aren’t the best stat to look at.

When I watch McLouth, though, I see a player who is quick and sure-handed and makes good jumps. He isn’t a spectacular center fielder, but I just can’t see how he’s a poor one.

by Charlie Wilmoth on Aug 8, 2008 4:11 PM EDT up reply actions

Ultimately, it doesn't matter.

McCutchen is a better defensive CF than Nate right now. As such, Nate’s going to a corner, sooner or later. The only real question is when.

by Vlad on Aug 8, 2008 4:17 PM EDT up reply actions

Eh...

I wouldn’t say he makes good jumps. Nate takes the occasional odd route and I think on balls in front of him often isn’t aggressive enough. His arm is adequate but not outstanding. He is definitely sure-handed when he gets to a ball, though.

It’s also important to keep in mind that PNC has a lot of centerfield to cover. Nate may be a decent centerfield in a normal park, but a decent centerfielder at PNC looks worse than a decent CFer elsewhere. There doesn’t appear to be any ballpark adjustment.

As Vlad mentions, though, it’s all moot as he won’t be staying there when McCutchen arrives.

by matskralc on Aug 8, 2008 5:27 PM EDT up reply actions

Defensive Metrics

I’ve always been sceptacle of defensive ratings, but I usually find Baseball Info Solutions (which i found though HardballTimes.com) -

http://www.hardballtimes.com/main/article/fielding-stats-at-the-hardball-times/) -

to be a pretty good metric. It has McLouth as a slightly below-average CF, but apparently plays a great LF, which I’ve read on here is the harder corner to play at PNC? Is that true? (I’m an A’s fan, but I come on bucsdugout regularly since my dad is a diehard Pittsburgh fan)

Anyways, if you check out the greatest positional seasons since 2006 according to their RZR (Revised Zone Rating) you’ll see the usual names (Mark Teixeira, Mark Ellis, JD Drew, Ichiro, Melky Cabrera).

Anywho, check it out. I enjoy it.

by WhatElse on Aug 8, 2008 9:05 PM EDT reply actions

trusting the eyes

Nate doesn’t appear to be on roller skates the way Nyjer Morgan did, at least. Morgan made some great plays, but that was freak speed covering up for the instincts of a hockey player. Maybe Nate has profited unfairly from that comparison in my mind, but I’m generally pretty relaxed about balls that get hit to center this year.

by KPatrick on Aug 8, 2008 9:44 PM EDT reply actions

Trusting the eyes

I think the phrase only means that, no, he can’t possibly be the worst CF in the league, having watched him almost every day. I don’t think it’s necessarily a substitute for metrics but, as for Rivas too, it’s easy to spot the flaws. I can’t tell you why the metric is flawed, but know a flawed metric when I see it.

by azibuck on Aug 9, 2008 1:04 AM EDT reply actions

a comment from Baseball Think Factory (Dan Szymborski)

Why can’t Luis Rivas play above his normal level and field well at shortstop for 23 games? Isn’t that like saying that offensive statistics should be taken with a grain of salt because Ramon Vasquez is hitting 325/393/492 or Jerry Hairston 344/393/484?
Of course fielding performance is going to fluctuate year-to-year, just like hitting and pitching performance does. Now, fielding measures don’t give us the granularity of hitting statistics, but they’re still going to be prone to all the rules of math such as small sample size and such.

And again, it’s 23 games. That’s 1/6th of a season. That’s like a month’s performance for a starting pitcher. Randy Johnson’s monthly splits for ERA this year are 4.79, 3.19, 6.82, 3.60, 1.35. Is his 6.82 ERA a weakness of ERA? of course not. Should Jack Cust’s consecutive OPS months of .649 and 1.004 be a strike against OPS? Nope.

As I note, fielding stats aren’t precise to the degree that the other aspects of the game are, but the fact that there are fluctuations, sometimes extreme, especially over extremely small stretches, isn’t a point of evidence against advanced fielding measures.

my blog // calling BJ Upton lazy is lazy // past results do not guarantee future performance

by Sky Kalkman on Aug 9, 2008 3:46 PM EDT reply actions

The issue is not whether Rivas could “play above his normal level” for 23 games. The issue is that I saw the 23 games, and he was a disaster. I think pretty much everyone here would agree with that.

by Charlie Wilmoth on Aug 10, 2008 2:25 AM EDT up reply actions

that's fair

how about this, though. over 23 games, might a fielding metric have some built in error that it’s off by a total of two plays?

take, for example, batting average over 23 games. if a player manages to get an extra bloop single and a seeing-eye ground-ball single, his average will be about .050 points higher. is that a problem with batting average? it’s not, even though we wouldn’t expect that player to hit .050 points higher going forward.

i definitey don’t think fielding metrics are perfect. a full season tells you about as much about a fielder as a month or two of offensive stats. but to totally ignore them, usually because of what people (both smart people and dumb people) observe isn’t a good idea, either.

my blog // calling BJ Upton lazy is lazy // past results do not guarantee future performance

by Sky Kalkman on Aug 10, 2008 12:51 PM EDT up reply actions