Garrett Jones vs. Nate McLouth
When the Bucs brought up McCutchen, they also traded Nate. While it might have looked like they were doing so to give Andrew the job in center, what they actually did was trade Nate so that Garrett Jones would get a corner OF position instead of Nate. They could have put Nate in LF but instead they traded him and gave Garrett an opportunity.
Looking at their numbers for 2009:
Age Games HR SB/CS Avg/OBP/SLG/OPS
Nate McLouth 28 84 11 12/6 .257/.354/.419/.773
Garrett Jones 28 82 21 10/2 .293/.372/.567/.938
Sure looks to me that the Bucs made the right decision!
Looking forward to 2010, is anyone not convinced that Bucs have the better person in LF? While Nate maybe a better fielder, Garrett’s offense could be as good or better. And they both are the same age! Plus Garrett is cheaper!
This is a FanPost and does not necessarily reflect the views of the managing editor (Charlie) or SB Nation. FanPosts are written by Bucs Dugout readers.
83 comments
|
0 recs |
Do you like this story?
Comments
Nate isn't just "maybe" a better fielder
We’re probably talking about 20+ runs’ worth of glove disparity between the two.
Yeah
Nate was a terrible CFer, but if he were moved to a corner, I think he’d be an elite fielder, possibly deserving of a Gold Glove.
by H2O on Dec 1, 2009 3:34 PM EST up reply actions
I'm not even projecting Nate all that aggressively
The numbers hate, hate, hate Garrett Jones’s glove in an OF corner. As you’d expect for a guy who spent ten years playing 1B in the minors.
Can he learn?
I know we’ve talked about this before; I’m just wondering whether, if he really has only worked on 1 position for his MiL career, there’s a possibility that he can get his glove up to near-replacement level.
The fact that he’s not slow helps; the fact that they didn’t try to increase his value by playing him in the OF as a minor leaguer does not.
Maybe, maybe not.
He runs pretty well for a guy of his size, so he’s got one of the main physical tools needed to be a better defender. The question is whether he has slow jumps and bad routes because he’s inexperienced, which is potentially correctable, or whether he has slow jumps and bad routes because he just sucks at judging fly balls, which probably isn’t.
The fact that the Twins weren’t willing to try him in the OF would suggest that the latter is more true than the former, but it’s always possible that they misassessed his abilities. I guess we’ll find out.
20+ runs on defense looks like a stretch to me. McLouth is -7 in CF with 3400 innings, which is probably enough data to get a good read. Jones is -8 so far at the OF corners without enough data, but it’s hard to see him as a -17 which would be almost worst in the league. Jones isn’t good by any means, but he never looked that bad out there. He’s not as bad as Bay.
The difference is about 10-15 runs, and about the same spread if he’s at 1B with average or slightly worse defense.
If Jones is 10-15 runs better with the bat, then the two are about even. Although Jones is less expensive so he wins. Granted, McLouth has a larger track record and is more of a known quantity.
by Adam Reynolds on Dec 2, 2009 2:29 AM EST up reply actions
20 runs might be a bit high...
…but at this point I couldn’t buy anything much less than 15. A big chunk of Jones’s numbers from last year were as a 1B, so you’re starting out with a 17-run positional adjustment versus CF right out of the box. Jones is probably about a -5 as a 1B (-3.5 UZR in his limited sample on range alone, plus a little extra for subjectively IMO having below-average hands as well). If you take his LF/RF UZR at face value, that’s a push with Nate, so you just use the standard 8 run adjustment for LF/RF to CF for those innings.
That probably understates the gap, if anything, since there’s evidence that PNC’s dimensions and the Pirates’ defensive alignment inflate the numbers for LF/RF and deflate them for CF.
Either way, though, it’s much too significant a defensive gap to just handwave away, as the OP attempted to do.
by Vlad on Dec 2, 2009 8:02 AM EST via mobile up reply actions
Good point with the PNC Park dimensions.
by Adam Reynolds on Dec 2, 2009 12:47 PM EST up reply actions
I don't think the Pirates were considering a McLouth/Jones tradeoff.
I think it was more of a McLouth/Tabata tradeoff. They could have waited until Tabata was ready, but then they might not have gotten the deal they got at the time.
Although chances are that Jones will start the year in right and Clement at 1B, I think the real competition for Jones is going to be between himself and Clement.
If Jones is playing RF, like Vlad mentioned, the Pirates are losing at least 20 runs over a player like Tabata or McLouth. Jones is likely an average 1B defensively and Clement likely below average (-10 runs on Jones).
So for the lineup to stick as Clement at 1B and Jones in RF, that would mean Clement would have to not only out perform Jones, but would have to generate 30 runs offensively more than Tabata.
I have a hard time seeing that happening unless Clement has a Legendary type year and Tabata proves to not be ready for MLB.
The odds of Tabata being "ready" are very slim
We shouldn’t be counting on him as even a starting option for 2010. 2011 maybe, if everything goes right.
I think it's safer to assume he's not than is
I realize management gushed about him and Pedro again recently, but there’s no reason to rush him through Triple A when he doesn’t have many ABs at the level yet.
His MLE for last season...
…is about a .600 OPS. ZiPS sees him at ..266/.323/.373 for 2009, while CHONE forecasts a .266/.321/.374.
He’s a good prospect, and he’s on track, but he’s just not ready yet.
So ZiPS thinks Tabata is more ready than many of these players ...
2009 ZiPS:
Andrew McCutchen: .261/.337/.362
Garrett Jones: .254/.304/.427
Chris Coghlan: .251/.315/.363
Sure, it's not infallable
But in the long run, hitting on 18 is a pretty crappy gambling strategy. That’s why it’s called “playing the percentages” – most of the time, that’s where the smart money is.
by Vlad on Dec 2, 2009 7:30 AM EST via mobile up reply actions
Percentages
I realize that there’s reams of data and many iterations of analysis to create ZiPS projections and MLE, but when they don’t pass the smell test – as they clearly didn’t with McCutchen – I think it’s reasonable to start questioning whether they do, in fact, accurately convey the percentages.
IOW, while GFJones represents what I think we can agree is an outlier to any reasonable projection, Cutch performed about as well* as any informed observer would have expected. But the projection tools noted by Mark got it egregiously wrong, which makes me loath to treat them as anything more than a bit of additional data. You seem to be saying that they are dispositive, that they define the “percentages.” I’m not convinced.
- better than median projection, probably, but well within “probable” range
Why didn't that line pass the smell test with McCutchen?
In 2008, he had a .770 OPS at AAA. In 2007, he had a .710 OPS at AA. Perfectly adequate performances for a player of his age at those levels, but not exactly the signs of an imminent breakout. It wasn’t out of the question that he’d play as well as he did, but it was definitely on the upper end of people’s expectations… just look at our own community projection for him.
Mark pointed to three examples of the projection missing on a player, which is fine, but relying on anecdotes ignores the vastly larger number of players for whom the projections were reasonably on-target.
I agree with you
but my point was that these projections should be looked at in the context of the other players they are evaluating. ZiPS thinks more highly of Tabata at this moment than it thought of Andrew McCutchen and Chris Coughlan, a player that is very similar as a hitter.
Do I think Tabata is ready right now to be in Pittsburgh? No. He hasn’t shown he can dominate the competition in AAA just yet.
However, to me, he has shown a progression that points to him probably being ready mid way through 2010. If he does very well against AAA pitching the first 2 months of the year, then I think he is going to be given a shot.
For Tabata, I think if he can keep his Ks to 10-11% of ABs in AAA, then he’s going to be all right in MLB.
by MarkInDallas on Dec 2, 2009 12:03 PM EST up reply actions
I think this take is about right, btw – I certainly don’t think he’ll be ready to come north in April, and I’m not sure he’ll be ready by July either, but I think he well could be, a possibility Vlad seems to be dismissing out of hand as “betting on 18.”
2009’s Tale of Two Cutchens gives me some faith that NH and his crew are reasonably good at evaluating when to bring up AAA guys (I’m not sure there’s a lot of data there yet after you throw out guys brought up to fill holes, but those 2 make me feel better).
For every McCutchen or Coghlan who exceeds expectations...
…there’s a Wieters (2009 ZiPS: .291/.361/.467) or a Schafer (2009 ZiPS: .244/.300/.415) who takes a step backward.
I don’t think Tabata will get the bat knocked out of his hands or anything like that, but players with walks and no power typically lose some of those walks (at least temporarily) upon promotion to MLB, once pitchers realize that they can be challenged inside the strike zone with limited repercussions. That’s why I think the key for Tabata making a successful transition is going to be a bounce in the power indicators (even if it’s only some extra doubles), and 2010 just seems a little aggressive for a timetable there to me.
Because we had a lot of reason to believe that Cutch was, is, and will be more than an “adequate” player. If the Bucs had rushed him along, or if he had started ‘09 in Pittsburgh, then I would have expected a more lukewarm performance from him. But, in general, this is a guy who was always projected to be an All-Star-caliber player. To say that, after steady progress, he was going to be unready for the bigs (and that’s how I would describe the projections/translation cited above) to me doesn’t pass the smell test.
I would note that you cite a .060 jump up in OPS from level to level as evidence that it would be reasonable to expect a bigger jump down in OPS the next time he went up a level. No improvement, sure; a bump down, sure; but an over .070 drop in OPS from a guy who was only getting better? As I said, doesn’t pass the smell test.
As for the community projection, there was a staggering number of people guessing he wouldn’t come up for much more than a month, so maybe not the most prescient forecasts (although my eyeball says that a clear majority had him above the ZiPS projection regardless).
I guess what it comes down to is that I’m not willing to put a lot of weight on ZiPS* or MLE if there’s other information out there – like that a given player has always scouted well, or that a given player has struggled when moving up in the past. It’s part of the picture, sure, but I don’t see it as defining the most likely outcome.
- talking about young players here
A .699 OPS forecast for a 22-year-old...
…in his first major league look is not a bad projection. Roberto Clemente had a .637 OPS in his age-22 season, just to pull one example from team history. A projection like that doesn’t mean that a player won’t grow up to be an All-Star – just that he’s not ready to be an All-Star right out of the starting gate (as very few players are).
The small bounce in raw OPS from 2007-2008 isn’t particularly significant from a predictive standpoint. The majority of it came from an fluctuation in BA, where McCutchen’s below-normal .258 in 2007 returned to a more typical .283. There was an increase in his walk rate, but also a drop in his ISO (and increased power is what was most responsible for his 2009 breakout). There’s probably also some park stuff affecting the line, as Altoona has traditionally been mildly pitcher-friendly.
Our community forecasts typically lead ZiPS projections for individual Pirates by a significant amount, because we are all at heart Pirate fans who can’t be entirely objective about our own guys. Which is fine, but doesn’t mean that the ZiPS line for McCutchen was any more unreasonable than any other projection for a Pirate last year, since we were higher on all those guys, too.
Incorporating subjective factors is fine, but I don’t see a lot of reason to use those to move Tabata’s projected line in this case. He hasn’t hit for much power so far in the AFL (though his overall line is good), and he doesn’t have a history of unusually quick adjustments after promotion.
On Tabata's power...
The Pirates have said his power is coming along, and if you look at his Indianapolis numbers, you’ll see his HR rate did increase throughout the year.
The FO also said that there were some balls he hit in the AFL that would have been HRs in PNC and MLB in general. The dimensions in Scottsdale are 360’ in left and 430’ in center.
AMac has also said it is easier to hit HRs in MLB because the ball carries much better because of the high stands surrounding the field, which cut down on the wind acting on the ball.
That said, I don’t expect Tabata hitting 20HRs in 2010. Maybe he gets to 10-12 over a full season initially.
I look for him to be a .300/.350/.410 guy initially. That’s a huge improvement over Moss and should be enough offense to keep Jones or Clement at 1B where they belong.
McCutchen's MLE before callup...
2008 with Indy: .238/.314/.327
2009 with Indy: .257/.306/.403
If Tabata is hitting .300+ with OBP of .380 in June at Indy and either Clement or Jones is not hitting the cover off the ball, Tabata will be coming to Pittsburgh.
Maybe so
But if he does, it’ll be a bad idea.
He still needs to learn how to put backspin on his line drives. Until he does, he’ll never hit for much power, and that’s a much easier lesson to learn at AAA than in the bigs.
by Vlad on Dec 2, 2009 8:05 AM EST via mobile up reply actions
I agree with this.
However, in DK’s most recent post he states that many within the organization expect Tabata to be a mid-season call up. I sure hope he finds that power stroke before then because like you said, I think it will be a bad idea if he doesn’t. On the plus side, our defense will be that much better with Jones, or Clement, at 1st.
I was going to ask about this
Vlad, I understand that you, personally, wouldn’t be planning to bring up Tabata if you were GM. But you seem very certain that NH won’t bring him up in 2010, when all the evidence is that he will – is that right? Or are you arguing independent of what NH may do?
I'm just saying what I would do.
I can’t speak for NH. I suspect that he’d be open to promoting Tabata at midseason if Tabata’s performance warrants it (as with Cutch last year), but I don’t really know.
on the point of OPS-es early in one's career
while .700 is perfectly acceptable in his rookie season, it is really to the Pirates benefit to bring him up when he can do more than be singles hitter with doubles power. Tabata will be playing a corner outfield spot in Pittsburgh, and you would want more production than that.
Also, on the topic of bringing someone up, there should be factors taken into account such as that by the time he develops power, like Cutch this year, will it be his 3rd season? Will he start getting a helluva lot more expensive? Why not let him mature at AAA for a year, and sort out the OFs we have right now? Will bringing him up be a big step forward for him in acclimatizing to the bigs, so he can contribute next year?
I would leave Tabata in AAA for the year, at this point- that said, if he shows power, and if the positions in Pittsburgh are clearer (read, the roles of GFJ, Clement, Moss, Pearce), I wouldn’t be averse to bringing him up.
I am also tempted to believe he might stay down for the year, because the Pirates would like to see more before deciding what to do with the players already in Pittsburgh. If there’s nothing to be gained in bringing Tabata up in June, why do it?
First, to be clear, I absolutely don’t want to see Tabata brought up prematurely – if he’s not ready in mid-summer, leave him in Indy. That said, I see the calculus as follows:
If Tabata isn’t brought up this mid-summer, we have to wait until next mid-summer – there’s no sense in leaving him in Indy all year in ‘10 then bringing him north in ’11, because you will simply have thrown away 3 months of free ML adjustment time. Once it’s 2011, may as well leave him down for another couple months to gain the extra year of control. So leaving him down this year means not bringing him up for a whole year, not just a couple months.
Most players will have an adjustment period – even Cutch looked pretty bad for a few weeks last year, after pitchers had adjusted to him. Given that 2010 is a lost season, I’d just as soon see the adjustment happen that year.
On some level, leaving Tabata down in 2010 is saying that 2011 will be a bit of a lost year as well. I don’t think 2011 is a playoff year regardless, but it would be nice to be in a position where the best of the old players (Doumit, Duke, Maholm) are still around while the best of the new players (Cutch, Pedro, Tabata, Lincoln) are in place (and familiar with the majors) from the first game. No guarantee Pedro arrives in ‘10 either, but he’s at least as likely as Jose, no?
If Tabata's ready at the start of 2011...
…and he wasn’t ready at the middle of 2010, I don’t see how you wasted anything by not bringing him up. If he wasn’t ready, he wasn’t ready.
I agree.
I’ll also say if he’s hitting .285 at Indy like McCutchen did in 2008, that won’t be enough. He’s got to be hitting at least .300 with Ks down to where they were in AA to warrant a call-up.
What I’m guessing is that Tabata will be hitting .320 in June, and that will bring him to Pittsburgh. I may be wrong, but that’s my prediction.
I also think that Alvarez needs to get his Ks down substantially as well. I actually see less chance for Alvarez to be brought up than Tabata for that reason. I don’t think Alvarez will make enough contact to be effective in MLB unless he gets his Ks down in AAA.
Power is the key
I am not sure a 320/360/420 gets Tabata to Pittsburgh necessarily, because a BA driven OPS isnt ideal.
That said, I think we all basically agree that the one reason to leave him in AAA is to let him mature, if only physically.
With Alvarez, there is the added constraint of his option years. I believe this will be his second, and he will have one more. I can certainly see him coming up in 2011 only, and not in 2010, especially with the Iwamura-LaRoche uncertainty.
What I don't understand
is why you wouldn’t want to see full seasons from Milledge, Jones, Moss, and Clement before deciding to bring Tabata up. Tabata’s only 21; there’s plenty of time for him, whereas if the other guys don’t stick this year they’re pretty much done. If Tabata takes any of their places, we’re basically giving up on them. I don’t see any good reason to do that. If Moss is repeating his 2009, maybe you give up on him, but not the other three, so I just don’t understand calling up a 21-year-old Tabata next year.
You are right
if Clement (or Ankiel), Jones and Milledge are all hitting the cover off the ball. No question about it. You only bring up Tabata if there’s a spot open for him and he’s doing very well at AAA.
But I would say the chance of all of those guys having great years is pretty slim. But if they all are, then there will be some pretty exciting baseball happening.
by MarkInDallas on Dec 2, 2009 11:26 PM EST up reply actions
But even if they’re not “hitting the cover off the ball,” I’d still give them more than half a year. That’s my point. Say they’re all just hitting league average. Who are you going to bench to make room for Tabata? If they cared enough to trade for Clement last season, they’ve got to give him a full year to prove himself in the bigs. After breaking out last season, you can’t bench Jones if he’s holding his own this year. And Milledge is still only 25, so I don’t understand giving up on him, either. I’m having trouble envisioning a scenario in which we would want to call up a 21-year-old to take any of those guys’ spots. Even if they all totally suck, I think you’ve got to keep playing them in the hopes that they’ll improve in the second half and accrue some trade value. The only reasonable shot Tabata has at a midseason call-up is if Moss is awful again and one of Jones/Clement is hitting really well but playing awful defense and some team is willing to trade something valuable for him. Otherwise, I’d argue that Jones, Clement and Milledge all need at least 500 PAs in ’10 so that the Bucs can evaluate their futures with the team.
And don’t get me started on Ankiel. Why anybody would want a terrible one-year stopgap at a position we already have more than enough options is beyond me.
Since league average is what you would hope to get from Tabata, if Clement if playing OK defense and hitting league average, you are probably right.
I don’t think you’re understanding me. What if Clement is hitting 240/320/400 at the end of June? My argument is that you should let him keep playing because 1) he needs to get a full season at the ML level so that the Pirates can evaluate him and determine his future with the club, and 2) sending him back to AAA or to the bench at that point kills both his career and his trade value. He’s 26, so it’s now or never for him. He needs a full season. My point more generally is that no matter how Milledge, Jones, and Clement are doing, it does not make sense to bench them at any point this year in favor of a 21-year-old. Why rush such a young player when you have older players with just as much talent who also deserve a shot?
I think that Huntington would prefer to light a fire under players fighting for jobs. Notice how Andy suddenly sprang to life once Walker was called up. It was clear at that moment – whoever does better is going to play. That was a message not only to Andy but also to Walker – to have a job you must outplay someone else. He tried that same thing between Moss and Pearce, but neither was up to the challenge. If Clement has the kind of numbers you mention in June and Tabata is ready, I see the same kind of situation developing. Tabata won’t be anointed by default, but will be given a chance to outperform Clement. If Clement can’t perform better, then he doesn’t deserve to be on the field.
by MarkInDallas on Dec 3, 2009 12:38 PM EST up reply actions
My point
Is that I don’t think you bring him up to start the season in 2011, ready or not – at that point you’re giving up 6 months in 2017 for 2+ months in 2011, which makes no sense, unless you expect 2011 to be meaningful. But, again, it’s frustrating to give up on 2011.
Bringing up Tabata in 2010 also gives away 2017, but you get 6 solid months in 2011 instead of a few months of adjustment plus the rest of his year (all of this, obviously, is based on assumptions about readiness and adjustment that may not apply – but the most likely development path is a few months’ struggle when he reaches the bigs regardless of ostensible MiL readiness).
I think you have to also view the big picture
Which is, prospects that reach the majors in good time and prepared to succeed helps them in their efforts to sign future amateur players. That is one of the few advantages they have over the Yankee organization.
Huh.
Do you really think this would make a lasting reputational difference? I mean, we’re basically only discussing this issue for the sure-fire guys: a Pearce or a No Relation is brought up if/when the org thinks they’re ready, not with an eye on the 6th year of control. And hasn’t Pedro effectively done an end-around via the major league contract?
IOW, it seems to me that it’s such a small percentage of players who would/should even consider this issue – it’s like a 5th round draft pick writing All-Star incentives into his initial contract.
Or maybe not. I dunno.
I do know
that this is something they do in fact use to try to get their more valuable prospects to sign quicker. Whether it works or not on an individual level I don’t know. But you can bet if Tony Sanchez comes up a year earlier than Grant Green, the Pirates will point to that for the next guy they are trying to sign quickly to show the importance of it.
i am sure you are right here
if Tabata’s not ready by June, he wont be brought up. In that event, the only scenario in which he comes on Opening Day 2011 is if the FO believes he is ready and they believe he would be a key piece in us contending in 2011. If not, he comes up in June 2011.
But....
…if you take the reports that Nate McLouth refused to move to a corner, then it’s unfair to compare the two fielding wise, right? Jones is a RF / 1B, he should be putting up bigger numbers, and at the same time shouldn’t be compared with the glove.
I personally think they’re about equal value if one believes in Jones. After Nate’s great 08 the big questions were: Can he do it again? Will he regress? Is there room to grow? McLouth regressed some in ’09, but still a 20-20 guy and a valuable player with the low average.
In my opinion, GJ will regress some this year, but not fall off a cliff as some here predict. 27 HR 85 RBI .825 OPS… something like that I think is realistic.
It's not unfair
In that you’re comparing both players to replacement-level talent at their position. They’re commodities in and of themselves, just like all players, and you can compare them (relative to baseline) just as easily as you could compare, say, a shortstop and a starting pitcher. If you couldn’t, nobody would ever get traded.
Anyway...
I’m sure Nate would have moved to a corner for the good of the team. His position was simply McCutchen should have to prove himself being the rookie. He never said he would refuse to move.
Michael Young initially balked at moving to 3B with the Rangers in favor of Andrus. However, after Andrus played some game there, Young was perfectly understanding of why it was better for him to play 3B.
Exactly
And he still played and realized in the end that Daniels was right that it would be best for the team. In the end, it was more about the way it was asked and the fact Young thought he shouldn’t have to move for the new guy.
Both McLouth and Young have poor career UZR ratings, and both were completely oblivious to how they were hurting their team with it. But in the end, I woudl guess McLouth would listen to reason just as Young did.
That's a stretch
But it’s a stretch because no one even knows what was really asked. So it’s based on a what-if of a what-if. Essentially, it’s a dumb thing to even really discuss.
However, Michael Young still was pissed and you’re underplaying it by quite a bit.
As I’m here in Dallas and followed that whole Michael Young saga from the eye of the storm, I brought it up as an example of a player that was upset about being asked to move, and yet in the end did it anyway for the good of the team. I didn’t mean to underplay the fact that Young felt insulted by the ask. The fact he was indeed very pissed strengthens my belief that McLouth would have moved if he was shown the team would be better off.
There's always other factors to consider
McLouth just signed a new deal, the Pirates weren’t expected to have a good W-L record, and the call-up happened during the year — there’s no way to know how any of these things would have affected McLouth.
I think that the crux of the matter is
the McLouth trade didn’t happen because McLouth was going to be baby and refuse to play a corner position. It happened simply because Huntington liked the team’s chances with Morton and company better than with McLouth. If Nate’s disappointment was indeed a large part of the decision making, that is not a positive point for Huntington. It shows a failure to communicate correctly with the player and bring him on board to better the team.
by MarkInDallas on Dec 2, 2009 12:12 PM EST up reply actions
I just like the deal
Because it was a good one. I’m not going to worry about the reasons behind it because there’s no way to know.
Which is why.....
I’ve argued that in watching Jones play about 70 games last year is defense, especially at first, or even in right, doesn’t seem to be anywhere close to as bad as UZR might suggest.
And while discussing it, can we talk about how WAR doesn’t take into account defense very well or changing economics at all?
Sorry.....
this comment was part of the bottom thread talking about the usefulness of UZR and UZR/150 btwn Stratman01 and Vlad.
Personally, from watching Jones last year...
…I’m pretty surprised that his UZR isn’t WORSE. Literally every game where I saw him in the outfield, he had at least one botched play. He was the worst-looking Pirate OF I can remember since Brad Eldred (who was just as clumsy, with instincts that were just as bad, but who also had the range of a walrus in a tar pit).
I don’t really see a problem with WAR re: defense and/or economics.
All the same...
I do think Jones will out-perform McLouth in 2010. McLouth is consistently .260/.350/.450 and I would see that as the bottom of what I would expect from Jones in 2010. I will predict .290/.370/.500 from him and ending the year at 1B.
Vlad I love you but.......
Like most people I see, you are completly misusing UZR. Even IF you assume that UZR is a good statistic to measure defense, which I think the jury is still out, you cannot use a single year of UZR to make any valid point about defense. Further, As MGL (the creator of UZR) has said on numerous occassions, UZR doesn’t tell you if a player is a good or bad defensive player, but only whether the player has a good or bad UZR. Moreover, I think there are serious problems with UZR and UZR/150 with regard to PNC park. I don’t know why that would be and Michael doesn’t either but it is being looked at.
You’re point is valid as I think there is no question McClouth is a better defender but I am simply making a point that the avenue you used to get there is not legit.
You can use any amount of information to make decisions...
…provided that you treat those conclusions a level of confidence appropriate to the information used to make the decision. Yes, Jones’s sample is much too small to be particularly useful (as was already directly stated in the post to which I was replying, and as I’ve noted in numerous other posts on the site), and yes, there are issues with advanced metrics in PNC park (as I noted in my post) – but it’s the best information we have available right now, and as such, a sensible starting point for speculation. If you need to say whether a particular hitter is a good hitter or not, you’re still better off with one month’s contextless offensive numbers than with none at all, right?
Since this is a casual exchange of guesstimates, rather than a serious analysis, I don’t think my post was appropriate. I wild-ass guessed the gap at 20 runs. Maybe it’s bigger, maybe it’s smaller. As they play more games, we’ll come up with a better idea of the real number – and it’ll become a proportionally less interesting question to discuss, since there’ll less room for subjective interpretation.
One thing I'll note
And this isn’t intended to pick sides on this specific matter:
As someone who was very prescient about the Iraq War has noted, certain kinds of bad data can’t simply be discounted to derive good data from them – they have to be thrown out entirely. Now that post was talking specifically about data provided by liars (“Fibbers’ forecasts are worthless”), which is not what’s being discussed here, but given the weaknesses of UZR, I’m not sure it isn’t in that category – a month or two of offensive data doesn’t tell you much, but a week’s worth tells you nothing at all*. And a couple months’ worth of UZR isn’t much more informative than a week’s worth of batting.
I guess you could argue that, if that limited amount of UZR comports with other information (your own eyes, past scouting), then you can use it to get an order of magnitude number, but I’m not sure even that’s valid – UZR is so variable (look at Bay’s numbers for the past 4 years – is he a -11 guy or a -18 guy? Or even a +3 guy!).
- small caveat that a guy going 25 for 28 over a week has at least told you that he’s capable, on some level, of hitting major league pitching – but it has to be that extreme to give information
If the precision of the estimate...
…were going to be the difference between us going to war or not, there’s no way I’d give a recommendation based on a sample that small.
But since we’re just bullshittin’ on a website, and “there’s no reasonable way to guess” doesn’t make for a very interesting conversation…
Which is the real Jason Bay?
Is he a wOBA .397 guy, a .387 guy, a .413 guy or even a .326 guy?
The answer is, he was all of those guys at one time.
We look at .397 and .387 and think there’s not too much difference there, but that is 6 runs difference (41 to 35) when calculated on a scale like UZR/150. That’s basically the same difference as Bay’s -11 and -18 UZR seasons.
Bay’s +3 UZR season is easily understood in the context that he had a knee injury that began his problems in the field starting in 2007. Before that, he was an average to slightly below average fielder. But his fielding did not significantly lower his value. After the knee injury, Bay had problems in the field, which he acknowledged. Since that time, his UZR numbers have been poor. He especially seems to have had problems with the Green Monster in 2008 (UZR/150=-24), which you would expect, then improved to UZR/150=-11 in 2009.
All of these numbers coincide with what I have seen with my eyes, so I don’t see any weakness of UZR being exposed here.
My point was simply that Bay’s year-to-year UZR jumps by more than 50%, representing almost half the gap that Vlad is citing between McLouth and Jones. It’s a noisy number, and so doesn’t admit saying this like “McLouth is definitely more than 15 runs better.” The (potential) range is vast, and almost certainly includes numbers lower than 15.
I’m aware of Bay’s injury history – the +3 was silly to include, I admit. Where do you get a breakout between his AL/NL 2008? Part of the reason I thought Bay’s numbers moved too much was that most people thought he looked much better in the field in 2008, and so the -18 surprised me – but if he was -12 as a Pirate and -24 as a Sox, then that would be less weird.
On fangraphs, click "show partial seasons".
Bay’s UZR/150:
2006 before injury: +3
2007 while injured: -11
2008 with Pirates: -14
2008 with Red Sox: – 24
2009 in contract year: -11
OK, thanks
But do you agree that -14 in ‘08 vs. -11 in ’07 is a bit surprising? Or is that just UZR noise and/or UZR’s heavy weighting of assists (remember how long UZR showed Moss a great RF because he had 3 assists in the first 10 games? Then it drifted down, down, down to a bit above average)?
Yes, but you can say that same thing about hitting stats as well. If Ronny Cedeno hits 3 HR the first 10 games of the season, that is going to color his hitting stats for a while. Nobody expects he’s going to hit 9 HRs a month – especially if he doesn’t have a single walk in that time. Take out McLouth’s first 12 games of 2008, and the rest of the season he hit pretty much his career average.
If we look deeper into Bay’s UZR stats, we see in 2008 his range factor was in fact better than in 2007, but his arm (measured by assists) was very low – accounting for half of his -14 with the Pirates. Then when he went to Boston, his range factor dropped off a cliff, and somewhat came back this year.
The main difference between hitting stats and UZR stats is just simply the number of chances that a fielder gets is about a third the number a hitter gets, so a large number of assists in few games takes longer to even out than a large number of HRs. But from my experience of keeping tabs on range factor for different players, I’ve noticed it seems extremely accurate to what I view with the eye.
Gotcha
It all comes back to sample size. But it also reinforces the idea that UZR is a very fuzzy number – in some sense it would be less distracting to only show it rounded to the nearest 5, because showing it to a tenth of a run gives a misleading sense of its precision.
As an architect, I experience a similar issue when drawing existing buildings on a computer – the CAD program is all too happy to offer me thousandths of an inch, when what I’m working from are rough measurements of a building that’s neither square nor consistent – nothing sillier than an 1/8" dimension shown on a 100 year old church, but it’s easy to end up there if you don’t recall where the measurements came from in the first place.
Yes but...
Like I said, the chances of getting an outfield assist are spaced very far apart – similar to RISP for example. Just because Jones had a low RISP this year doesn’t mean he won’t do well next year. ARod tanks in the playoffs one year, then does well the next. There aren’t too many conclusions you can make about his ability there.
However, Range factor is based on many more chances, and if you look at the spray charts of balls hit to the outfield, you’ll see it doesn’t really take too many chances to get an accurate reading – enough to say this guy is good or average or bad.
Range factor and hitting stats tell you what someone did, not what what they are going to do. Making a conclusion as to whether someone is a good hitter or fielder you do based on those stats and what you see with your eyes in certain situations.
If you based it on Holliday dropping the ball in the playoffs, you might think Holliday is a bad fielder. But when you see he generally has a fairly positive range factor, about average on errors, and a slightly negative arm rating, you get an idea of his strengths and weakness as it relates to defense.
When it comes to quantifying that, you can definitely say “Bay has allowed more balls to drop in his area than the average left fielder”. You can also say “If Bay had caught the average amount of balls that came his way, the Red Sox opponents would probably have scored 11 less runs on the year and the Red Sox probably would have won 1 more game”.
But you can’t say “This means Bay will cost his team 11 runs next year” anymore than you can say “Bay will hit 35 HRs next year”.
All you can do is try to make a prediction based on what he did in the past.
I hate no edit button
I of course meant to say “Mitchel” as in Mitchel Lichtman and not “Michael.”
Your point is undertsood
and I know it was a guesstimate. I have no issue with that. My problem is more with the criminal misuse of UZR (not that I am talking about you necessarilly) that I see occuring everywhere—even on Primer. Mitchel has said over and over that basically there are 2 ways to use UZR as a stand alone stat: 1) compare the UZR of one player to a similar player in a given year (i.e., CF to CF for 2009) or, 2) use 3 years worth of data to get a general sense of a players defensive value.
The problem I see everwhere are posters saying something like, “his UZR was -17 last year. He sucks on defense.” I am not pointing a finger at anyone, just a general observation. That is a complete misuse of an otherwise advanced metric that is likely better than what was available before it. The problem with any statistic is how the data is interpretted. The misuse and misinterpretation of the data is going to lead to a faulty result and as a consequence the validity of the statistic is going to be doubted rather than the interpretation of the stat which is usually the problem in the first place.

by 














