"A New York Treasure" --Village Voice

Be Afraid

The third and final installment of our Bronx Banter Breakdown season previews:

Share: Share on Facebook Share on Twitter Share via email %PRINT_TEXT


1 Cliff Corcoran   ~  Mar 31, 2010 9:56 am

For those counting, the sixth position the Red Sox improved at is shortstop with the Marco Scutaro signing (I failed to mention that on camera): CF, LF, SS, 3B, DH, C.

2 Diane Firstman   ~  Mar 31, 2010 10:24 am


I wonder if the Sox have any regrets for the Beckett-Hanley Ramirez trade.

3 Sliced Bread   ~  Mar 31, 2010 10:25 am

Bring 'em on! I just hope the Yanks don't stumble out of the gate. First 6 games of the season (3 in Boston, 3 at Rays) could weigh heavily 6 months from now.

If you subscribe to the walk-year theory, aren't the Red Sox making a mistake rushing to retain Beckett? I don't think it makes much difference, but why not make Beckett wait, and give him that theoretical walk-year edge?

4 Sliced Bread   ~  Mar 31, 2010 10:36 am

[2] that trade has already paid off with a championship. This year they'll probably regret losing Bay's bat, as it's worth, I think, considerably more than Ellsbury's defense/ weaker bat in left.

5 ms october   ~  Mar 31, 2010 10:41 am

[4] yeah - i think so too - i don't think lf is an upgrade overall - just quite a bit defensively - last year ellsbury had a war of 1.9 and bay was at 3.5 war

6 Mattpat11   ~  Mar 31, 2010 10:42 am

I'm not at all sold on Marco Scutaro.

7 Shaun P.   ~  Mar 31, 2010 10:58 am

[6] I'm sold on his glove, but not on his bat. We'll see if Fenway helps him.

[5] I was thinking that I might be able to like WAR, but then I found out the defensive component of WAR is UZR. I have a lot of misgivings with UZR unless the sample is really, really big - and I'm pretty sure WAR uses UZR over 150 games, which to me isn't a big enough sample. Thus, I'm going to stick to BP's WARP; Bay's WARP1 in 2009 was 4.0, Ellsbury's was 1.7.

The thing that neither WAR nor WARP nor any metric I'm aware of measures is how the improvement in defense is going to help the Red Sox pitchers. That may be the biggest difference, and while I don't know if we can measure that, I bet they can.

8 williamnyy23   ~  Mar 31, 2010 11:00 am

Saying that the Red Sox’ lineup improved in 6 places is a stretch. I’ll give you SS and 3B just because what the Red Sox were throwing out there was weak (although I am not convinced that Beltre will outhit the 2009 Lowell). The other “4 slots” are iffy, however. For starters, if Ellsbury really was so awful in CF, why is it a given that he will do well in LF? Furthermore, the offensive drop off between Cameron and Bay could be very significant. The only way Boston could make up for that is if UZR is dead on in its accuracy (an argument I don’t think you can make). As for DH, has Francona stated he will platoon Ortiz and Lowell? It’s my understanding that Boston is looking to trade Lowell. Finally, while Vmart is an upgrade over Varitek, the latter did have a first half OPS of .826 (mitigating the “whole season of Vmart upgrade”). Furthermore, if defense is going to be considered in favor of Cameron, what about at the catcher’s position? Both Vmart and Varitek can’t through, but the receiving skills and pitch calling of the captain should be missed (if the Boston pitchers are to be believed).

As for the other 3 positions, I think you can pencil in a similar year for Pedroia and Youkilis, but Drew is a good bet to either drop off or miss time.

9 williamnyy23   ~  Mar 31, 2010 11:07 am

[7] Exactly...the holes in UZR re gaping. Even if you completely accept the methodology and forgive the chances of human error in compiling, the technology used is still close to a deal breaker. Last year, I read a great article explaining how the cameras used to evaluate the zones are in different places for every ballpark, meaning the judgment of angles (the parameters of the zones) can be skewed. I have no idea why so many people seem to put their eggs in the UZR basket.

10 Shaun P.   ~  Mar 31, 2010 11:35 am

[9] That reminds me of a big problem I have with relying on any of the fX tools via MLBAM. How accurate are those things? What's the level of error present? Is it the same for all parks? And not just in measurement of location, but especially for pitchers, the speed of the pitch. I think ignoring - or more precisely, not being able to account for - those possible errors in the data is a HUGE problem if one is going to rely on that data.

I think there's been something of a groundswell of support for UZR; kind of a chicken and egg thing. FanGraphs (and WAR) use UZR, people like FanGraphs and WAR, so people like UZR. Its also a bit easier to understand than the other "major" individual defensive metrics (Dewan's PlusMinus, Pinto's PMR, and BP's FRAA/FRAR), and is easily accessible - the biggest problem WARP/FRAR have is not being easily accessible.

UZR isn't awful, but just as an example - we know that it is bad, perhaps really bad, at measuring defense at 1B. Why then would I ever take a first baseman's WAR seriously? IIRC, UZR is also not great at catcher defense, so why would I ever take a catcher's WAR seriously?

11 Cliff Corcoran   ~  Mar 31, 2010 11:37 am

[9] I'd love to read that article if you have a link.

My take on CF/LF is that the two together represent an upgrade because the combined defensive upgrade is larger than the offensive downgrade from Bay to Cameron, but I agree that it's debatable.

12 Cliff Corcoran   ~  Mar 31, 2010 11:39 am

[10] Shaun, I'm with you, I completely disregard WAR as useless because of the impact of UZR.

13 Shaun P.   ~  Mar 31, 2010 12:16 pm

[8] "(although I am not convinced that Beltre will outhit the 2009 Lowell)."

Beltre will certainly outglove 2009 Lowell, by a lot, and as I said in [7], we don't have a metric available that measures how that will help Boston's pitchers, particularly their starters who all tend to give up a good number of ground balls. I believe it will help them quite a bit. Ditto at SS.

"For starters, if Ellsbury really was so awful in CF, why is it a given that he will do well in LF?"

See, for example, Johnny Damon. C'mon dude, that's just a silly agrument.

"Furthermore, the offensive drop off between Cameron and Bay could be very significant. The only way Boston could make up for that is if UZR is dead on in its accuracy (an argument I don’t think you can make)."

Oh yes one can! Bay's WARP1 in 2009 was 4.0 (638 PA). Cameron's WARP1 was 4.1 (628 PA). And again, the improvement in what the pitchers do due to having Ellsbury-Cameron, not Bay-Ellsbury isn't being measured in those numbers. Sounds like an upgrade to me.

"As for DH, has Francona stated he will platoon Ortiz and Lowell? It’s my understanding that Boston is looking to trade Lowell."

And in the meantime, they will have to play him some if the hope to trade him.

"Finally, while Vmart is an upgrade over Varitek, the latter did have a first half OPS of .826 (mitigating the “whole season of Vmart upgrade”)."

Ignoring OPS and its problems - to me its got even more holes than UZR - Varitek's WARP1 in 2009 was 0.5 (425 PA). Martinez's was 2.7 (672 PA). So yeah, that's clearly an upgrade. And while FRAR is no better than UZR at capturing pitch calling - I know of no metric that does - its abundantly clear that what it is measuring pushes things very much in Martinez's favor. And please don't pull out catcher ERA, which Keith Woolner (among others) conclusively showed does not exist.

14 williamnyy23   ~  Mar 31, 2010 12:35 pm

[10] Excellent point. While Pitch FX does away with unreliable radar guns in favor of a complex system of video at 30 frames per second, the setup process involved at each ballpark and before each game opens the door to quite a bit of error. It's hard to tell how significant that is, however, because the setup is so complex.

[11] I wish I could find the article, but intuitively, it makes sense. If you are creating a stat based on human observation of angles, then any variance in the cameras used will introduce errors. For example, one camera at a ballpark may bias UZR toward 3B and against the SS, while another could have the opposite effect. When combined, the differences between those teams fielders could be significant.

Again, I just don't get the popularity of UZR. In my opinion, it is a novelty at best.

15 williamnyy23   ~  Mar 31, 2010 12:47 pm


1)Wait...because Damon was better in LF than CF, that proves as a rule that all CF'ers, regardless of quality, will make good LF'ers? Really? Besides, how good was Damon in LF anyway? In by far is largest sample size, Damon was pretty bad last year.

2)Using WARP1 is very misleading for this comparison because it doesn’t factor league effects. Usually, WARP2 will give AL players a big boost, so selecting WARP1 bias the argument toward Cameron.

3) As for Varitek/Vmart, your analysis compares a full season of each, not 4 months of Varitek + 2 months of Vmart to 6 months of Vmart. As noted, Varitek was a good offensive player for the first three months of the season, so factoring in his second half swoon doesn’t follow that formula. In this exercise, we are only concerned with what the Red Sox did get from Varitek before Vmart came on board (i.e., the upgrade of having a full season of Vmart). As for the defense, those close to Boston insist Varitek is a master receiver. I don’t know if that’s true, but without evidence to refute it, it does need to be considered.

16 Paul   ~  Mar 31, 2010 1:06 pm

My problem with UZR is exemplified by Ellsbury and Jeter:

Ellsbury: How does defense swing almost 40 runs in one year? If defensive metrics can swing that wildly from one season to the next, how reliable and thus predictive are they?

Jeter: How does a 35 yo guy go from well-below average to well-above average because of off-season exercises? If defense is a set skill that can be that easily manipulated, how can we ever expect to accurately measure it?

As for the Sox, I've got two main problems:

1) They're built for their home park. Of the last four seasons, they've had a winning record on the road exactly once (2007). They'll get their 50+ wins at home with the emphasis on pull hitters from the right side. The problem is when they aren't breaking 40 wins on the road. They lose any shot at homefield in the playoffs (hurting their chances again cause of how they're built) and if they really struggle they could hurt their chances at a playoff spot (2006). in this context, it's important to note that Bay's offense was mostly context independent. Compare that to the expected, but home-dependent, gains they'll get from Beltre and Cameron and always have from Pedroia (career road = .750 OPS).

2) They're depending on too many guys who really need to perform more than they usually have or at higher levels than is typical of recent performance. For the former I count Martinez and the fact that he hasn't caught 100 games since 2007. And with no where else to play, he'll have to. As for recent performances, I look at Scutaro and Ortiz. They'll have to either rebound or keep up career years to make a real impact on the offense.

I know I wouldn't be surprised to see the Sox miss the playoffs. There's a reason Theo called it, in a moment of clarity, a bridge year. I think the real competition is going to come from Tampa Bay and especially Jeff Nieman.

17 Cliff Corcoran   ~  Mar 31, 2010 2:10 pm

[13] I appreciate the help, but WARP factors in defense, while VORP does not. Still, those figures support my point, if Cameron is at least Bay's equal if not better when you factor in defense and Ellsbury should be at least a bit better in the field with less territory to cover, it's a net win at those two positions.

18 Raf   ~  Mar 31, 2010 2:17 pm

[16] In Jeter's or Ellsbury's case, it may be a spike. You'll them in offensive metrics, be they batting averages, home runs and whatnot, why would defensive metrics be different?

19 Yankster   ~  Mar 31, 2010 2:28 pm

[16] Nice home win stat - makes me feel more confident about the Yank's chances this year.

[0] This was the best of the three video talks and is sportscenter level. I prefer to read, but the vid is excellent.

20 51cq24   ~  Mar 31, 2010 2:44 pm

hey cliff if you love the red sox so much why don't you go blog about them?? are you gonna pick them to win the world series again??

another problem with WAR: yesterday neyer pointed out that cameron's is just about the same as bernie's. is that realistic? maybe, and i do think defense is often overlooked, but i have a very hard time believing that cameron is anywhere near as valuable a player as bernie was.

21 Paul   ~  Mar 31, 2010 2:56 pm

[18] Good point. I'd say:

1) We know how to evaluate spikes in offense (BABIP, peripheral stats, etc.) How do we do the same for defense?

2) We expect offense to jump around because we understand hitting is a skill that's 70% failure for even the best hitters. Should defense really be as variable?

3) We have much more data for offense to evaluate things like peak and expected performances. We don't have the same for defense.

22 Paul   ~  Mar 31, 2010 3:04 pm

[19] Yeah, I was surprised when I saw it too. I actually wonder if Theo is playing with fire given how he structures his lineup. They make the playoffs, but then without home field they're at a distinct disadvantage (2008, 2009). Defense may be one way to mitigate that, but then I think most smart teams build their offenses to some extent for their home parks. The problem for the Sox is it's more extreme at home and on the road. Last year they had a +.100 OPS at home and they actually pitched better at home too.

What really kills me is how Pedroia is perceived. In any other park, he's Brian Roberts - good but far from great.

[20] Don't do that. He's being honest.

23 williamnyy23   ~  Mar 31, 2010 5:47 pm

[20] I think WAR and just about every defensive stat overrates that component. I really hope Theo is using the widely available defensive metrics to make decisions because that will inhibit their chances of being successful.

24 Paul   ~  Mar 31, 2010 6:09 pm

[23] They clearly have their own system and from the rumors I've seen that they use their own cameras atop Fenway. Obviously it works only with their own players given sample sizes. But I bet that's why they didn't hesitate to move Ellsbury. The question though is why UZR had him among the best CFs in the game then called him among the worst.

feed Share on Facebook Share on Twitter Share via email
"This ain't football. We do this every day."
--Earl Weaver