FanGraphs Baseball

Comments

RSS feed for comments on this post.

  1. Is there a reason you didn’t factor in positional adjustments since Morgan plays the more premium position?

    Comment by The A Team — July 1, 2009 @ 11:12 am

  2. Too bad Whitey Herzog wasn’t a blogger about twenty-five years ago.

    Comment by Joe Anderson — July 1, 2009 @ 11:16 am

  3. This is a really interesting blurb. Would love to hear some of these other analysts’ response.

    Comment by Max — July 1, 2009 @ 11:20 am

  4. It’s also time fangraphs stops using UZR as gospel. Everyone knows fielding run metrics are in their infancy, and there is usually large variation between metrics.

    Example:
    Jason Bay’s UZR in 2009: -7.0
    Jason Bay’s FRAA in 2009: +8

    Now I know baseballprospectus loves to keep FRAA shrouded in mystery which makes it kind of meh, but a FIFTEEN DAMN RUN difference in two metrics supposedly measuring the same player ability, not to mention an ability that has very little variance compared to hitting (biggest difference between full time players last year at the same position was Carl Crawford vs. Brad Hawpe at 56 runs, I believe, and that took near historical levels of badness from Hawpe to happen).

    So when my two favorite baseball websites are saying two entirely different things about a baseball player in terms of their defense, how am I supposed to view defensive run metrics as anything more than what I do now, food for thought? And while I 100% agree Dunn is a miserable defender, saying Nyier Morgan is the equivalent to ANY guy who routinely posts 40 HR and .390 OBP seasons and could very well end his career in the top 15 of all time in Home Runs (at the very least top 20), outside of being a Kingman-esque hacker, by using a metric w/ so much variance, is a statistical cherry pick to the max, and horrifically irresponsible.

    Comment by Joe R — July 1, 2009 @ 11:21 am

  5. Here’s how I would characterize the dilemma, as a non-participant in (but active follower of) the “sabermetric community”:

    There is a strong consensus within the sabermetric community regarding the magnitude of the difference between Dunn’s offensive value and Morgan’s offensive value. Tools for assessing offensive value have matured over decades and are widely understood. Different offensive tools exist, but generally different tools produce similar stories. The community feels comfortable with its ability to project future offensive value from past data.

    However, there is less of a strong consensus within the sabermetric community regarding the magnitude of the difference between Dunn’s defensive value and Morgan’s defensive value. Tools for assessing defensive value have been developed in more recent times and are not as widely understood or believed. Different defensive tools often produce different stories. The community feels less confident about its ability to project future defensive value from past data.

    Comment by Rowen — July 1, 2009 @ 11:23 am

  6. I’ve had some harsh words in the past but I find this to be extremely interesting stuff. While I’m gung-ho about offensive numbers in sabermetrics, it is hard for me for some reason to get a grip on defensive statistics.

    Dave — given that Morgan is such an elite defensive outfielder, would the Nationals be better served trading Dunn and also employing a Morgan-esque Juan Pierre in left field? While the Dodgers were ripped for putting Pierre’s awful bat in left, I would have to think that either Pierre or Morgan would go from a great defender in center to an elite one in left field, no?

    Is defensive prowess really this important? This article makes it seem that Morgan being a below average hitter and excellent fielder versus Dunn being an excellent hitter and a horrid fielder is roughly the same thing. Is that really the case? I guess that goes against everything I’ve been taught about baseball numbers but it is something I’m open to considering. Hard to wrap my head around it.

    Comment by Garrett — July 1, 2009 @ 11:24 am

  7. Joe R and I were typing at the same time, so I hadn’t read his comment when formulating mine; but I think Joe R’s views on the value of existing sabermetric defensive tools is consistent with what I was trying to say below.

    Comment by Rowen — July 1, 2009 @ 11:25 am

  8. And to expand on that, if Dunn ever moved to the AL and became a DH (like he should), we’re talking a guy who would routinely post 3.5-4 win seasons. It would be nearly as good as having pre-decline Frank Thomas as your DH.

    And best of all, since he’s such a patient hitter and three-true-outcomesy, you wouldn’t have to worry about too much volitality outside of an Ortizian-level April/May bomb out.

    Comment by Joe R — July 1, 2009 @ 11:27 am

  9. This is kind of the vibe I get as well about defensive numbers. Looking at the way they change from year to year, I find it hard to believe we can truely establish a baseline for each player the way we can offensive numbers. While I would be inclined to believe Morgan is a great defensive player, to determine the gap between him and Dunn is not as cut and dry as it is with offensive numbers.

    Also I still maintain that a player of Milledge’s pedigree is simply too much to pay for a 29 year-old rookie who has most likely quite a few clones across MLB. Regardless of how valuable they may be or not, is there really that much of a difference between a guy like Morgan and the potential bats of a half dozen players in the minor leagues with the close to the same skill set?

    I could liken this to the late 90s/early 2000s when a lot of valuable Jack Cust, Russell Branyan, Calvin Pickering’s were toiling away in the minors. Yes, they might have been undervalued but that was sort of the point, being able to recognize that and exploit it by underpaying for them in trades or in signings was the way to get ahead — not by paying relative value for them just because you know (or believe you know) the true value.

    Comment by Garrett — July 1, 2009 @ 11:30 am

  10. The answer is yes, and no.

    Yes in that if there was a definitive metric that measures defense as accurately as offense (for fangraphs, it’s wOBA, for baseballprospectus, it’s EqA), then we’d be in business. But there isn’t. Defensive metrics are evolving and will likely never be as accurate as offensive metrics.

    At least UZR gives the fan an alternate to the vaguely defined FRAA.

    Comment by Joe R — July 1, 2009 @ 11:30 am

  11. Its not Dunn’s fault teams play him in the outfield, if you played most starting first basemen or DH’s in the outfield they would be just as bad as Dunn in the outfield, probably worse. Dunn is clearly the superior player miscast as an outfielder. Nyjer Morgan will be out of baseball in 2-3 years, meanwhile Adam Dunn will be in the AL where he will be the perfect DH slugging 40 home runs a year and not hurting his team defensively.

    Comment by west21 — July 1, 2009 @ 11:32 am

  12. Sorry, Dave.. You are delusional here. Using Dunn as your comparison to make Morgan legitimate is pretty lame. Everyone already knows Dunn is overrated as a NL player. Making this comparison to Morgan to try to prove Nats won the trade is just stupid. It just proves that their signing of Dunn was idiotic.

    Did the Nationals get better in the short term? Maybe. I think Hanrahan, who has true set-up/closer value, is superior to Burnett, who is nothing more than lefty specialist.

    Completely writing off the potential of Milledge in the deal is also quite convenient for you as well. He’ll get better at the plate (already superior to Morgan), he’ll get better in the outfield.

    Comment by sekrah — July 1, 2009 @ 11:34 am

  13. I have to agree with Joe R and Rowen. It isn’t that UZR or defense isn’t valuable. It has value. The question is just how we calculate the value. I’m not sure there is a consensus in the sabermetric community on UZR as the be-all/end-all defensive stat, and although we use it because we know it is probably the best there is, we are pretty definitive in our belief in stats like wOBA. It just seems unlikely that Nyjer Morgan’s defense is really 3 wins better than Adam Dunn.

    When it comes to player evaluation, we need to put our trust first in the metrics that we have been tested again and again. And as of now, offense is far ahead of defense.

    Comment by Pat Andriola — July 1, 2009 @ 11:46 am

  14. I think one of the key points Dave is trying to make can be summed up in this sentence: “Nyjer Morgan and Adam Dunn are nearly equals in value, ”

    To me, he is trying to say that Dunn and Morgan, in their current situations, have the same value. Dave isn’t saying that Dunn and Morgan have the same talent level, but just that their values are equal with the Nationals due to their defensive positioning.

    I think it’s clear that if Dunn was playing first or DH, then his value but would be increased significantly. But in their current situation, Dunn is being used in the least valuable way.

    I think it’s more of a testament to how misused Dunn is rather than the value of Morgan.

    Comment by Jay Paradise — July 1, 2009 @ 11:47 am

  15. David didn’t take UZR at face value, that’s why he regressed the heck out of it. It’s not hard to believe that Morgan could be 10 runs better than the average LF.

    And comparing UZR to FRAA doesn’t make any sense to me. The data inputs are completely different, and I prefer the ones that use pbp data.

    Comment by puck — July 1, 2009 @ 11:48 am

  16. Even if we disagree about the degree of accuracy UZR/150 provides, I think anyone who’s watched both players can easily see that Morgan is really good at defense and Dunn is historically awful. Maybe something in the post is overstated, but ultimately it’s obvious that the theme of the article is accurate; Morgan provides a ton of value to the Nationals. This is a team that has given CF innings to guys like Austin Kearns. Even if UZR overstates the value of defense, Morgan is getting paid league minimum and is a huge boost for the Nats in CF.

    Comment by The A Team — July 1, 2009 @ 11:48 am

  17. Good stuff here. UZR numbers’ infancy and their fluctuation are obviously red flags and skepticism. It’s obvious too the better defender of these two, but I find it ironic the bashing that Juan Pierre receives (deservedly so, he sucks), but how much different of a player is Nyjer Morgan? He is remarkably similar to Pierre, with less stolen bases and a smaller sample size of MLB data to judge. In short and simple closing, there is no way in hell Nyjer Morgan is better at this baseball game than Adam Dunn.

    Comment by Richie Abernathy — July 1, 2009 @ 11:55 am

  18. Adam Dunn is a really horrific first baseman. It’s scary, but he might actually be worse there in a full season than he is in left.

    Comment by Jacob Jackson — July 1, 2009 @ 11:55 am

  19. “Now I know baseballprospectus loves to keep FRAA shrouded in mystery which makes it kind of meh, but a FIFTEEN DAMN RUN difference in two metrics supposedly measuring the same player ability”

    Quite simply, FRAA should not even be looked at. It doesn’t use play by play data, just putout totals and some sort of context adjustment. UZR takes into account where a ball was hit, and how hard. If you get a big difference between UZR and a measure like John Dewan’s plus-minus, and you often will, that is legitimate criticism of defensive measures. But comparing UZR to FRAA is like saying Bob got a different measure of the distance to outfield fences by pacing it off than Ed did by using a tape-measure. FRAA is good for only one thing: Crude estimates of defensive performance for seasons where play by play data are not available.

    There is a lot more uncertainty in estimate defense, but that doesn’t mean we can’t compare them. I am very confident that Dunn is at least a -10 defender in left, and Morgan is at least a +10 (which means he’d be average in center). That would mean Morgan is 20 runs, or 2 wins, behind Dunn. Giving Dunn a -20 and Morgan a +20 seems to be as far as you can go, which would make them even. The truth is somewhere in between.

    And of course, whether he’s as good as Dunn or not has nothing to do with whether he’s worth Lastings Milledge. It all depends on what Milledge turns into, we know what Morgan, at age 29, is and will be. I think the trade makes sense for the Nats even if Milledge tuens into a .300, 25 HR corner guy. The reason is Milledge doesn’t fit on the roster. He is absolutely not a center fielder, and with Dunn, Dukes, and Willingham, the Nats have better options in the corners. Morgan gives them a legit defensive centerfielder, it’s something they need. Pittsburgh has a pretty good CF, and needs a potential bat in the corner. This trade, in my opinion, makes sense for both teams.

    Comment by Rally — July 1, 2009 @ 11:58 am

  20. I think Pierre gets crap because for a period he was regarded as a star. If you could pay Juan Pierre league minimum as you can with Morgan, then he’d suddenly become an incredibly valuable player.

    Comment by The A Team — July 1, 2009 @ 12:02 pm

  21. true. pierre’s contract is bat shit crazy.

    Comment by Richie Abernathy — July 1, 2009 @ 12:07 pm

  22. FRAA was recently changed to include PBP data for the years in which it is available.

    Comment by Erik — July 1, 2009 @ 12:11 pm

  23. Why is everyone bringing up Juan Pierre? Pierre=/=Morgan.

    Pierre has been a below average defender in 5 out of the last 6 years.

    Comment by Davidceisen — July 1, 2009 @ 12:15 pm

  24. I don’t think BP was vocal enough when they added PBP data into FRAA, because it’s quite clear that people still seem to think of FRAA as a solely non-PBP metric. It’s not a complete PBP metric like UZR, but for recent years, it DOES use PBP data. It was somewhat explained in the BP2009 annual, though it’s still mostly “hidden.” It DOES include PBP data though.

    Comment by Erik — July 1, 2009 @ 12:16 pm

  25. I don’t think we’ve mentioned “skill scarcity” yet. It’s the same quibble I had with Dave’s “Is drafting Strasburg over Ackley a no-brainer?” piece, which was just as thought-provoking and entertaining as this one. Always appreciated, Dave.

    Unlike a few of the posters here, I’ll fully accept that Morgan’s defense has made him as valuable as Dunn this year. However, Dunn’s terrific power/OBP is akin to Strasburg’s 100 mph fastball, and Morgan’s off-the-charts defense is Ackley’s league-average-projection-corner-OF.

    In the comments of that piece, I argued that what Strasburg could become is irreplaceable, even if his projection (as a super hard-throwing SP) is more volatile than Ackley’s; meanwhile, Ackley’s skill-set, assuming he’s a corner outfielder, is available every offseason for a decent price in the wake of the recent defensive revolution (think Abreu or Burrell). Strasburg inarguably has the far more precious skill set, even if the projected WAR vs. WAR calculation on the two were similar.

    Back to Morgan/Dunn, here’s what I mean: You can’t just find guys who can hit 40 bombs and get on base 40% of the time off the street. It is a very precious skill set. A great defensive outfielder, I would argue, is much easier to find. The minor leagues are littered with guys who are outstanding defenders but just can’t hit enough to find regular work, like Ryan Langerhans. Rajai Davis is probably one of the five best defensive outfielders in the major leagues, and yet he made it through the entire NL when he was DFA’d by the Giants last year, even though his skill set is tailor-made for NL play.

    I can agree that Nyjer Morgan has been “much closer in value” to Adam Dunn this year than 99% of baseball fans think, which I think was Dave’s main premise…but somehow the community needs to develop a metric that better evaluates the scarcity of skills. Nyjer Morgan does something that lots of guys in baseball can (almost) do, some of whom languish on ML benches or even in the minor leagues. Adam Dunn’s skill set is far more rare.

    An analysis that includes wOBA, UZR, and EQBRR only, and not that scarcity, will occasionally lead to faulty conclusions, in my humble opinion.

    Comment by Jacob Jackson — July 1, 2009 @ 12:16 pm

  26. and his arm is atrocious, perhaps the worst OF arm in the game. I was wondering the same thing.

    Comment by Jacob Jackson — July 1, 2009 @ 12:17 pm

  27. I am surprised that Dave isn’t writing a scathing article about how, because of this move, the Nationals are sending their ‘best’ player to AAA.

    Comment by Davidceisen — July 1, 2009 @ 12:18 pm

  28. Here’s the kicker… while baseball isn’t just about who can hit the ball further and it is time we stopped evaluating players on their offensive worth alone, estimate of offensive value are much more accurate…

    For instance, the Dunn/Morgan analysis uses a lot of assumptions to derive the estimates of their true defensive skill levels.

    Lets say Dunn is in reality is closer to a 3 win player and Morgan is closer to a 1.5 win player which is a scenario that isn’t outlandish given the large error bars associated with the UZR-derived defensive numbers.

    Suddenly, Nyjer isn’t quite so Dunny even though few would argue Dunn’s defense isn’t mostly funny.

    In other words, there probably is reason to be cautious about equating player values when the gap is closed by a large magnitude of defense value.

    Comment by Terry — July 1, 2009 @ 12:25 pm

  29. “Ortizian-level”, Love it!

    Comment by verd14 — July 1, 2009 @ 12:25 pm

  30. I never said the trade didn’t make sense, I just think as a pure math nerd (and not just a baseball stats nerd), that it’s a statistical cherry pick to imply Nyier Morgan = Adam Dunn based on adding volatile #’s to their hitting and positional adjustment stats.

    And let’s just say it wouldn’t be the first time Dave Cameron has misassessed defense: http://www.fangraphs.com/blogs/index.php/second-base-is-the-new-third-base

    Comment by Joe R — July 1, 2009 @ 12:28 pm

  31. He did. He gave Morgan a +5 positional adjustment.

    Comment by JH — July 1, 2009 @ 12:31 pm

  32. Does nobody realize that Dave’s quoting a range of defensive ability every single time he analyzes a player’s contribution? It would be one thing if UZR was a massive outlier that contradicted +/-, PMR, and scouting reports, that would be one thing. This happens in limited circumstances, but more often than not, players that we count as above average defenders are guys on which there’s a pretty broad consensus that they’re good. The only question is how many runs their glove saves. For that, we use a range and make conservative estimates to try and approximate their overall value. It’s a pretty good system, and it gives us a ballpark for comparing the values of players with completely different skill sets.

    Comment by JH — July 1, 2009 @ 12:37 pm

  33. I agree with the crowd signaling the inconsistency of defensive metrics, though Dave did try to do his best to regress it as much as he could.
    A difference that the sabermetric community seems to either ignore and/or fail to be able to measure is the value of a player like Dunn from a marketing stand point.

    Now yes, I agree that the best way to get fans in the seats is to win, so that should be valued above all else. However, there is value in the kind of baseball Dunn brings to the table. Though a cliche, it rings true: Chicks dig the long ball.
    And so do kids and basically every other average fan, which conservatively makeup at least 75% of a fan base (I would bet a lot more).
    A guy who can hit home runs, and long ones at that, will put more people in the seats and more eyes on the TV screen than a player like Morgan.

    The questions then is, how valuable is that difference. I can’t wait for the day when we can actually incorporate things like this into a player’s value (at least from the organization’s stand point). Anyone have ideas?

    Comment by Ryan Hoffman — July 1, 2009 @ 12:38 pm

  34. Because people think “black, lefty thrower, lefty hitter, small frame, fast, cool name, left fielder”

    You’re right. They’re not that similar.

    Comment by Pat Andriola — July 1, 2009 @ 12:38 pm

  35. Dunn has more value as a DH, nuff said.

    Comment by Steve C — July 1, 2009 @ 12:39 pm

  36. I thought he regressed his UZR because he was a rookie and small sample size? NOT because of the Metric itself…?

    Comment by verd14 — July 1, 2009 @ 12:48 pm

  37. I agree with you. General baseball knowledge and watching a bunch of Pirates games this year will verify what UZR is basically telling you: that Nyjer Morgan is an above average defensive player.

    And I agree that “The only question is how many runs their glove saves.” It’s what I was trying to say. I feel a lot stronger in the direct transfer from offensive runs to wins than I do from defensive runs to wins due to the complexity of the defensive system and its relative newness.

    All I’m saying is that, as a community, we’re more confident in our offensive metrics than our defensive ones, and by the widest degree possible between the three main categories (i.e. offense, defense, pitching).

    Comment by Pat Andriola — July 1, 2009 @ 12:48 pm

  38. except that dunn’s lifetime UZR/150 at first is WORSE than in LF. in 900+ innings (small but not ridiculous sample) at 1B, Dunn is worth -12.7 runs/150; in 8200 innings in LF, Dunn is worth -11.0 runs. He’s clearly a pretty negative defensive force wherever he goes. The only role he’s truly suited for is DH.

    Comment by tom s. — July 1, 2009 @ 12:53 pm

  39. BP’s own writers don’t even use FRAA anymore! They constantly refer to UZR these days in many of their articles. FRAA is the most worthless of all the defensive metrics, so you really shouldn’t worry about what it has to say.

    Comment by Bodhizefa — July 1, 2009 @ 1:09 pm

  40. Verd14– exactly. Regression is due to sample size, not the effectiveness of UZR

    Comment by metty5 — July 1, 2009 @ 1:11 pm

  41. Don’t offensive numbers vary a lot from year to year, too? And pitching numbers? I don’t understand why people use that particular argument against defensive data while ignoring the fluctuating data sets of offense and pitching. It’s silliness!

    Comment by Bodhizefa — July 1, 2009 @ 1:13 pm

  42. We have a large enough sample to accurately address Dunn’s true talent on defense. Morgan’s defensive numbers were heavily regressed. I’m confused as to why you think the author and others don’t understand the volatility in UZR.

    Comment by kwk — July 1, 2009 @ 1:20 pm

  43. This discussion thread isn’t relevant to Dave’s posting. Dave is not saying that Adam Dunn would be a better DH, or that the trade was good/bad for the Nationals. He is simply saying that a sabermetric analysis (using currently available defensive metrics) would argue that Dunn and Morgan have similar value to the Nationals.

    He isn’t arguing the merits of defensive statistics here. In fact, he acknowledges their noise by using the conservative regression. He is arguing that defense matters, and even in the sabermetric world it gets discarded.

    Simply put, in their current roles Dunn and Morgan have a similar value for the Nationals. If you have 100 pennies (Morgan) or 4 quarters (Dunn), you still have $1.

    Comment by Erich — July 1, 2009 @ 1:21 pm

  44. I think that Dave may have missed the point when he assessed the value of both players. The reasons that people felt signing Adam Dunn to that contract was a good idea were (a) his perfomance was worth the $10mm per year, (b) he could potentially help them contend, and if not, could be traded for players who could help them contend in the future, (c) signing him proved to other free agents, who are bigger stars, that they were serious about improving the team through free agency. Clearly, they are not in a position to contend this year, and it is very possible that they will turn him into prospects that will put them in position to contend in the future. By contrast, Nyjer Morgan was acquired by trading Lastings Milledge, and most people feel that Morgan is at his peak in value and will decline before Washington fields a contending team, while Milledge has most likely not reached his peak and still has the potential to become a very valuable player. The extra value in terms of wins this year that the Nationals gain from the Morgan trade does not really equate to a significant amount of real value because they are so far out of contention. Dave appears to be valuing Dunn and Morgan based upon their performance and expected performance for this year, while those that think the Pirates got the better of the deal, and those that were thinking of not only this year, but the future.

    Comment by David — July 1, 2009 @ 1:22 pm

  45. Isn’t a lot of the criticism over the trade and not the player? The Nations already had Nyjer Morgan in Ryan Langerhans and they had just given him away for nothing. Unless they feel they’ve really upgraded with Burnett over Hanrahan the Nations are the same team as 3 days ago minus Lastings Milledge.

    Comment by NCE — July 1, 2009 @ 1:25 pm

  46. Reading fail…

    Comment by The A Team — July 1, 2009 @ 1:32 pm

  47. “You can’t just find guys who can hit 40 bombs and get on base 40% of the time off the street. It is a very precious skill set.”

    A run is a run is a run. It’s a fallacy that teams need players with (for instance) Dunn’s skill set to fill certain roles. As far as I know, there isn’t any evidence that loading up on defense first, weak offense players will lead to diminishing returns.

    Comment by kwk — July 1, 2009 @ 1:32 pm

  48. I think what isn’t being discussed is that IF you have a guy like Dunn in the field THEN it is important to have a guy like Morgan to play with him. This kind of deal makes Dunn better by allowing him to have to cover less ground in the outfield, thus increasing his value (while he plays there). This is something Kearns, Milledge, and Dukes are incapable of doing.

    Comment by Ed Nelson — July 1, 2009 @ 1:45 pm

  49. All these sabermetricians you’re blasting are simply saying this was a bad trade for the Nationals, not because Morgan is a bad player, but because in the long term, he’s worth less to them than Lastings Milledge would be. (I think everyone agrees that Hanrahan / Burnett, while being relatively inconsequential, was also a trade favorable to the Pirates).

    There are many players in the minors who could scrape up a +2 win performance in the outfield based on slap hitting and excellent defense. Even if you buy that Morgan is a +3-3.5 win player, there’s the excellent point that the Nationals already had Ryan Langerhans. I think Langerhans is not that much worse than Morgan, and based on Milledge’s potential, I’m not sure it was really worth making this move. If Milledge doesn’t pan out, then yes the Nationals “won” the trade, but it’s only a small improvement over Langerhans, during a period when the Nats won’t be contending anyway. On the other hand, if he does…

    Comment by David — July 1, 2009 @ 1:49 pm

  50. Maybe my readings of the UZR numbers are incorrect, and Pierre does have a horrid arm, but he looks pretty solid on defense to me? Am I reading these incorrectly? And they’re both high speed/average guys with little power. I think they’re pretty similar and the reasons Pat pointed out make them even more so.

    Comment by Garrett — July 1, 2009 @ 2:03 pm

  51. I worked out years ago how to accurately measure the defensive contribution a player makes, and I have never seen it employed. This makes me think the technical obstacles would be too difficult to overcome.

    All you would have to do it measure the time it takes for every fly ball to reach the ground. Using that combined with the location of the landing point, you can follow the data to see what velocity balls hit to what areas result in an out, a single, double, triple, etc. With a large enough database, the categories of fly balls hit can be quite specific by combining them by landing point and time aloft. Once you calculate the value of each fly ball in play, you can pretty easily compare the value of each fielder’s contributions by comparing the result of the plays they made or didn’t make vs. the league average for such a ball.

    For ground balls, you simply measure the angle of the hit against the time it takes the ball to cross the base paths, which of course is just about the only constant thing in each park. With this data, you can know how much value can be expected from a ground ball hit at a 37 degree angle which takes 0.47 seconds to reach the base paths. If the second baseman converts that ball into an out, you can then compare that to the result of the average for the contribution the fielder made.

    I’m not certain how to quantify infield plays, but I figure with this system you can measure about 90 to 95% of the value that each fielder contributes to preventing runs. This of course will help in quantifying how much each pitcher relies on his defense, how “lucky” each hitter is, etc.

    I suppose as I said the technical obstacles of actually performing all this work would make such a project infeasible for the sabermetric community, but with each win worth over $2MM in the major leagues, I can’t imagine that they are not devoting resources to this type of quantifiable analysis.

    Comment by Dan — July 1, 2009 @ 2:10 pm

  52. My take is Dunn helps put a few extra fans in the seats for a team who has no shot. Dunn or Morgan are not long term options if the team ever hopes to compete. Milledge although a long shot at this point still has a future to grow into a better talent. Even if they just waited for Milledge to regain more trade value.

    There is a big difference between signing a player who is not costing you talent and trading away talent to get someone. Even if they did add the same win value to your team you lost something in the deal.

    Comment by Troy Patterson — July 1, 2009 @ 2:17 pm

  53. Agreed Bodhizefa. It baffles me.

    Comment by Erik — July 1, 2009 @ 2:19 pm

  54. Exactly what in the post you linked is misassessed?

    Comment by NadavT — July 1, 2009 @ 2:42 pm

  55. While I agree that players with Morgan’s skill set are generally undervalued when compared to players with Dunn’s skill set by the hoards of sabermetric fandom, calling this particular trade anything but a win for the Pirates is re-dic-u-lous.

    Morgan is a fantastic defender and Dunn is awful. BUT one of the key factors holding back the development of defensive metrics is the lack of ability to account for positioning and how much the player(s) next to you help. The Pirates have a fast, young, exciting outfield in which each player helps make the ones around him better. Morgan plays a good deal of leftfield because the Pirates have other capable defenders in center. Example: If I’m playing right field and I know that the centerfielder can run down anything in the gaps, I’ll cheat over towards the line and make more plays that might be viewed as out of my zone. Dunn plays in a historically awful outfield, and doesn’t get the help that any other LFer in the league does….I have much more faith in team defensive numbers than individual ones. While I’m not debating the relative merits of Morgan being really good and Dunn being really bad, I can’t completely buy into those numbers being the only reason for similar Win Totals. I think defensive numbers are too context driven, at this point in time. How much of a better SS did A-Rod make Jeter?

    There are a couple other facets of the trade that have been over-looked here. One is upside. Morgan is a useful player, but it hasn’t been mentioned what Washington gave up. Morgan is closing in on 30-years-old and his only real tool is his speed. Players with late starts generally don’t age well, and players with one skill generally don’t age well. Morgan is both. The number of players on the wrong side of 30 who continue to play GG caliber CF is extremely short. Morgan today is likely the most valuable he will ever be. Milledge, for all his inconsistancies, at least has a chance to be a very good player, for a long time. His raw defensive ability isn’t far behind Morgan’s, and his offensive ability dwarfs anything Morgan could hope to accomplish. Zips has Milledge as a .347 wOBA hitter RoS. Even if you think Milledge will never be mature enough to maximize his abilities, and think Morgan makes the Nats two wins better…what good does that do? Giving up Milledge’s potential for the right to break the 60-win plateau, I don’t see how that makes sense. Both of these teams are awful, and need as much high ceiling talent as possible. Washington gave up a chance, even if only a very teeny-tiny chance, of having a franchise corner stone. 1% of a star player should be worth more to a last place team.

    There were other players involved as well. Hanrahan is a legit power-righty who is capable of pitching in high leverage situations. His K/BB ratio and FIP have improved over each of the past three seasons. Washington just gave him away for a Sean Burnett, a replacement level lefty with poor control. The only thing Burnett has on Hanrahan is a prettier ERA. Washington sold mind-bogglingly low on a valuable reliever with just over a year of service time. I guarentee that every GM in the pennant race was furious Hanrahan was just given away.

    Even if you think Morgan has been as good a player as Dunn this year. Where does he play? Does he take ABs away from Dunn? or Dukes? or Willingham? There’s no big advantage there. Morgan is likely a better player this year than Milledge, so the Syracuse Chiefs just got a little better for the stretch run.

    Comment by LHamilton — July 1, 2009 @ 2:59 pm

  56. If there are +2 win outfielders sitting around in the minors for free, don’t tell Atlanta because the Phils are scrapping.

    Comment by don — July 1, 2009 @ 3:11 pm

  57. Very nice work, Dave. I love rabble rousing! Great post.

    Comment by Chomps McGee — July 1, 2009 @ 3:16 pm

  58. So if we assume that Morgan is one of the best true-talent centerfielders who has ever lived (Despite, you know, being considered a corner outfielder by his last team) and Dunn is actually worse than the -15 runs or so that defensive analysts say is the realistic low point for defensive performance, Dunn is only a little over half a win better? Yes, truly Morgan = Dunn.

    You know, I actually don’t like Dunn at all and was incredibly impressed with Morgan this year, but stupidity still needs to be called out. Somebody please tell me that Cameron doesn’t actually get paid for his work here.

    Comment by cpebbles — July 1, 2009 @ 3:59 pm

  59. Hit f/x

    Comment by BIP — July 1, 2009 @ 4:41 pm

  60. I’ve heard a lot on either side of the deal. I’d say that the arguing over how good UZR is as a statistic is a viable argument, but not so much that it should affect this particular trade. There’s obviously ample evidence that Dunn is terrible in the outfield, and -20 runs seems fair. There isn’t enough evidence to say that Morgan is ridiculous in the outfield, but Dave’s suggestion of approximating at +10 seems fair. A +10 outfielder is excellent, but not earth-shattering like Morgan’s current numbers are suggesting.

    That being said, the trade itself isn’t particularly good or bad. I’d be happy to see the Nats run with Morgan in center field and take his 1 win defense. Considering he hits at replacement level, you have a 1 WAR player, perfectly acceptable. The reason why people were calling the Pirates the winners is the potential expectation of Milledge’s play. Milledge as a corner OF could be worth 2 WAR (i.e. average ballplayer) if he gets his act together, and that’s why the Nats were considered “losers” of the deal. It’s potential vs. a weaker but known commodity.

    That being said, I can accept Dave’s point here. Dunn and Morgan have been about even and could be expected to be even as we go through the season. Defense matters, pay attention to it, and all that usual jazz. It’s not surprising, just worded in an intriguing fashion.

    Comment by Michael — July 1, 2009 @ 4:54 pm

  61. “All these sabermetricians you’re blasting are simply saying this was a bad trade for the Nationals, not because Morgan is a bad player, but because in the long term, he’s worth less to them than Lastings Milledge would be. (I think everyone agrees that Hanrahan / Burnett, while being relatively inconsequential, was also a trade favorable to the Pirates).”

    - If you accept that Morgan is pretty equal to Dunn and that Milledge would be better than Morgan. Then, isn’t the argument then that you should’ve never signed Dunn in the first place because Milledge was/is apparently better and Dunn took his spot in the lineup?

    - For all the people talking about what Dave “meant” by this post…it seems pretty clear that his intention wasn’t even really comparing Dunn and Morgan, but instead saying that if people called Dunn a great signing then they should be saying the same thing about Morgan.

    - Finally, I found it surprising how many people claimed that signing Dunn was necessary cause people like to watch HRs? I’m sorry in my 30 years on this earth…where I have been more and less enlightened in baseball…I’ve never went to only to see guys hit HRs. I want the team to win…hitting HRs is just as boring if your team is last in the MLB.

    Comment by Myk — July 1, 2009 @ 5:00 pm

  62. You know who else was considered a corner outfielder by his previous team before being traded? Franklin Gutierrez, one of the top two or three outfielders in baseball. Because the Indians had their “center fielder of the future” already in Sizemore. Same goes for the Pirates with McCutchen. It makes no sense to knock a player’s value down for being blocked, when we have a lot of evidence that he’s better than a corner outfielder.

    As for calling out stupidity, you’re looking a bit hypocritical when you make it quite clear that you either skimmed the article or you have a problem with reading comprehension. Dave explains how he to gave Dunn the benefit of the doubt and was conservative with Morgan’s projection, based on the noise in the data. But hey, don’t let it stop you.

    Dunn is only a little over half a win better? Yes, truly Morgan = Dunn.

    The headline was obviously a bit hyperbolic. Goodness.

    Comment by Teej — July 1, 2009 @ 5:05 pm

  63. Not to mention that Adam Dunn is the kind of player many common fans tend to dislike: He strikes out a ton, he can’t bunt, he can’t field, he can’t run, and there is a perception that he’s lazy and/or uninterested in playing hard. Hardly the kind of guy who puts butts in the seats.

    Comment by Teej — July 1, 2009 @ 5:07 pm

  64. Most of these comments are getting distracted by the minutiae instead of picking up on Mr. Cameron’s main point, best expressed in the last line: “It’s time we stopped evaluating players on their offensive worth alone.”

    As several have noted, Dunn has skills that are theoretically rarer, and he would possess considerably more value if he served as a designated hitter. But he’s not a designated hitter. He’s playing the field, and while we may debate exactly how much his deficiencies there are hurting his team, they clearly are to a significant degree. This isn’t about Dunn vs. Morgan so much as how the statistical community is evaluating and responding to different players.

    Many analytical sites, whether fantasy or otherwise, do suffer from substantial biases. We like to believe that we’re more enlightened than plain old scouts, because we use objective numerical elements to measure value, but how we employ them is often highly subjective. One of my favorite examples deals with how low walk batters are routinely dismissed before their careers even begin, with the occasional exception of someone like Howie Kendrick who was embraced in spite of violating a model they’ve used to dismiss so many other prospects. The existence of these biases is a good point by Mr. Cameron, and one the community needs to confront.

    Comment by J.D. Bolick — July 1, 2009 @ 5:17 pm

  65. being a +10 CF isn’t approaching the best CF ever; he would have been fifth overall last year. the +35 is the difference between a +10 CF with a +5 defensive adjustment and a -20 LF. Dunn is clearly a -15 to -20 LF and has been for the last four years. You could readily say that dunn is the worst defensive outfielder in the major leagues — he was outdone only by hawpe among OF last year in terms of UZR. if you’re going to see defensive extremes, adam dunn is it.

    as of today, morgan is valued at 1.9 wins above replacement so far this year. that’s better than dunn’s value in two of the last four years, and twice dunn’s current value. even taking into account some defensive regression, how is it ridiculous to suggest that dunn and morgan have similar value?

    and why would you note that his prior team put him in a corner position? why suggest that the pittsburgh pirates’ positional choices are somehow unimpugnable? they put nate mclouth in center instead of morgan, which by any reasonable measure was stupid.

    i think the nationals got a steal, not least because the worst qualifying defensive centerfielder last year was . . . lastings milledge, with a -20 UZR. lastings milledge has been around a long time to still be waiting on his potential to crop up. he’s 24, in his fourth ML season, and sporting a .202 wOBA.

    Comment by tom s. — July 1, 2009 @ 5:24 pm

  66. …and this is the column that makes me stop caring about what Dave “Homer” Cameron has to say. Quite the superiority complex.

    Comment by J.J. Hardy — July 1, 2009 @ 5:41 pm

  67. Dave Cameron, world’s biggest Nationals fan.

    Comment by Teej — July 1, 2009 @ 5:46 pm

  68. Yeah. Because I’m referring to his “love for the Nationals” and not the Mariners. +1!

    Comment by J.J. Hardy — July 1, 2009 @ 5:57 pm

  69. Being a true talent +10 CF is being one of the best CF ever. Likewise, pulling a regression out of his ass is not giving Dunn the benefit of the doubt on his defense. MGL has covered this ground, and a -20 UZR in LF is not really within the bounds of reason as a true talent level.

    Comment by cpebbles — July 1, 2009 @ 6:36 pm

  70. It is pretty weird for a baseball writer to have a favorite team.

    Comment by Teej — July 1, 2009 @ 6:50 pm

  71. i guess you’re welcome to think what you like. there certainly were four players last year rated at +10 or better UZR at CF in 2008 — not sure how that makes him a “best ever” rather than a top 15% in any given year.

    but it seems like you’re simply saying that you don’t think that a) Dunn’s defense can be as extremely bad as -20 runs or that b) we can say anything about morgan’s defense. I don’t agree with either.

    regardless, this seems like a simple disagreement on the significance of defense between you and Dave. if you think everybody who disagrees with you is stupid, then i guess there’s not much point in talking.

    Comment by tom s. — July 1, 2009 @ 7:37 pm

  72. I agree that chicks dig the long ball, but as a guy who’s seen most of the major league games that Morgan has played, I’ve got to disagree that his “curb appeal” is that radically different, or radically less, than Dunn’s. Pirates fans really enjoyed Morgan, and that’s not just Stockholm syndrome; he has a lot of flair.

    Comment by KPatrick — July 1, 2009 @ 7:59 pm

  73. My only quibble with UZR here is that we only have BIS data and not STATS data too. Apparently, there are huge discrepancies between the results based on the different data:

    http://www.insidethebook.com/ee/index.php/site/article/suzr_v_buzr/

    Comment by Aaron B. — July 1, 2009 @ 8:11 pm

  74. And this is not saying that I’m ungrateful for it. “Only” bUZR is better than no UZR at all.

    Comment by Aaron B. — July 1, 2009 @ 8:12 pm

  75. Shouldn’t it be a +10 positional adjustment going from center (+2.5) to RF/LF (-7.5)?

    Comment by Gee — July 1, 2009 @ 8:28 pm

  76. While I disagree with David on the merits of the Nationals picking up Morgan for Milledge, some of you guys are really going out of your way to be nasty to David, which is completely uncalled for.

    Comment by Dan Szymborski — July 1, 2009 @ 8:33 pm

  77. When was the last time Dave’s Mariners fandom got in the way of his analysis on this site? Do you actually read his stuff?

    Comment by Mike I — July 1, 2009 @ 8:33 pm

  78. This is an EXCELLENT post. Everything that needed to be said was said. Well done LHamilton.

    Comment by Sekrah — July 1, 2009 @ 9:43 pm

  79. Hey, if he wants to go onto a popular website and be a writer and throw out idiotic claims like this one, he’s going to have to be able to take the heat.

    Comment by Sekrah — July 1, 2009 @ 9:50 pm

  80. It’s not just the claims, it’s him actually believing he knows better than the community at large.

    Comment by J.J. Hardy — July 1, 2009 @ 9:56 pm

  81. Negative six for this comment?

    Morgan’s a solid player, but no star. And guys who arrive late tend to leave early. The Pirates certainly took a risk here, but if Milledge figures it out, his ceiling is much higher and will be contributing further down the road. Hanrahan’s another risk, but has a much higher upside than Burnett.

    This was a *trade*, meaning the Nats had to give something up for Morgan (instead of $10M a year for Dunn.) It certainly wasn’t a landslide win for the Pirates, but probably a smart move for them.

    Comment by Sky — July 1, 2009 @ 9:57 pm

  82. I’m shocked anyone who reads his stuff both here and at USS-Mariner could say what you did and mean it.

    Comment by J.J. Hardy — July 1, 2009 @ 10:01 pm

  83. I agree with this.

    Comment by Nick — July 1, 2009 @ 10:02 pm

  84. He probably does.

    Comment by Nick — July 1, 2009 @ 10:03 pm

  85. Dave-

    While I completely agree with the assumptions of this article (Dunn ~ Morgan), I’m pretty sure that most people did not laud the Nationals for signing Dunn in the first place. Almost everyone in the sabermetric community released that Dunn was only a bit better than a league average player and he was being overpaid slightly be his contract. Combine that with the fact that A) the Nationals already had a glut of outfielders, and B) really bad teams should not pay market value per win; and I seriously doubt that anyone actually considered this a good signing in the first place.

    Also, while the Nationals got a pretty good player in Morgan, he certainly doesn’t help them much at all. They are still an awful team, and given that he will be a free agent by the time they are good (that 2011 rotation with Strasburg and Zimmerman could be pretty sick), he really just doesn’t add to their team much. I would argue that even though Milledge is an inferior player now, he also has the potential to be better than Morgan, and he is under team control longer.

    Also, swapping Hanrahan for Burnett is probably a net loss of .5 wins in itself, although I’m not sure the deal could have been done without that.

    Comment by Nick — July 1, 2009 @ 10:12 pm

  86. It’s not a fallacy when you’re playing in the AL. A guy who can hit 40 HR with a .400 OBP with terrible defense is destined to fit the skill set of designated hitter.

    Comment by Will — July 1, 2009 @ 10:52 pm

  87. I agree with your point, but with the $4.5M / win rate that free agents were going the previous season, his “slightly above average” player status pretty much fits the going rate (average players worth $9M a year).

    Comment by Michael — July 1, 2009 @ 10:58 pm

  88. Dave has spent the past three or four years criticizing the hell out of the Mariners, until this offseason. A homer is probably the last thing I’d accuse Dave of being.

    Comment by Teej — July 1, 2009 @ 11:03 pm

  89. Wrong. The Nationals have a surplus of corner outfielders (Willingham, Dukes, Dunn, Kearns) but only one CF (Willie Harris, who the club value more as a superutility guy). Langerhans, through several hundred innings at CF, has proven to be below average. His real value lies in RF and LF, where his defense about a win above average. Langerhans does not equal Morgan. They’re two completely different players.

    Comment by Will — July 1, 2009 @ 11:04 pm

  90. Agreed. People seemed to be arguing against a Dunn signing on the assumption that he was going to be grossly overpaid. That never happened, and Dunn got about what he was worth as a free agent, so the actual signing didn’t end up drawing the criticism that many, including myself, thought it would.

    Comment by Teej — July 1, 2009 @ 11:07 pm

  91. There is an aspect to this trade that no one has mentioned.

    I am one of the few thousand souls masochistic enough to watch the Nationals on TV most nites. This is quite possibly the very worst team I have followed in the decades I’ve been following baseball.

    This team has no legacy to draw on, such as an inter-generational fan base. The ownership of the Nationals I suspect has come around to realize that they are perilously close to vaporizing what is left of their fledgling fan base by putting an absolutely EMBARRASSING product on the field. Lastings Milledge was a passable CORNER outfielder, but an absolute train wreck in Center Field. Dunn has also been hideous to watch, lumbering after ball after ball other players would get to.

    The Nats had to do something NOW, to put something resembling quality defense on the field NOW– especially with a young pitching staff– to avoid even more long-term permanent damage. I concur with previous posts about the value in Dunn and Willingham no longer having to cover as much ground.

    Yes, in a sense, it’s foolish to trade POTENTIAL upside in a young player for a player at his peak when the team has no chance of winning it all this year or next– but there are other long-term contingencies to success of the franchise as a whole to think about.

    Milledge’s horrible plate discipline, bad (uncoachable) attitude, and realistic projection to a corner outfielder (but without the power), makes me far far less sanguine about his prospects for becoming a truly elite player. At best, he is a lottery ticket whose odds of payoff have weakened considerably by his play this year before he got hurt.

    I’d say it was about a dead-even trade, with critical upside for the Nats in the PR department.

    Comment by Jim — July 1, 2009 @ 11:07 pm

  92. Feel free to refute him.

    Comment by Teej — July 1, 2009 @ 11:19 pm

  93. Sorry, but that’s not Dave’s main point. Dave’s thesis is that Nyjer Morgan is approximately as valuable as Adam Dunn. This is not an article about the general over-valuing of offense, even if that theme is touched on. I hardly think a bunch of replies asking if Nyjer Morgan is indeed better than Adam Dunn qualify as some sort of navel-gazing missing-the-forest-for-the-trees exercise.

    Comment by battlekow — July 1, 2009 @ 11:35 pm

  94. Teej, are you suggesting that calling Dave a “homer” isn’t actually a viable rebuttal?

    Comment by twinsfan — July 2, 2009 @ 1:31 am

  95. You guys are right about the contract value, it looks to be pretty much right on. Still, the deal wasn’t praised by most people; it just wasn’t lambasted.

    Comment by Nick — July 2, 2009 @ 2:16 am

  96. Doesn’t value necessarily involve money in this context? Look at the salaries- Dunn gets $10 million a year; Morgan gets $400k. You can buy a decent amount of talent for $9.6 million.

    Dave’s a good writer, and the stick he gets from some of his readers is absolutely uncalled-for. I’m not sure of my feelings on this deal. I live in DC and watch the Nationals, and this is a team in transition that needs to rebuild not just at the major league level but in the front office and overseas. Morgan will be a productive player for a few years yet, at league minimum salary no less, and that’s no small concern, I think, for an organization that has no hope of making the post-season and needs to recover from Jim Bowden’s tenure. On the other hand, Milledge has a chance to be a special player. I’m on the fence, but no matter how strongly you feel, things like:

    “Hey, if he wants to go onto a popular website and be a writer and throw out idiotic claims like this one, he’s going to have to be able to take the heat.”

    are just beneath the standards this website ought to have.

    Comment by MCR — July 2, 2009 @ 3:31 am

  97. Exactly. Hitf/x.

    Well, when they really get it worked out, that is. Once they get it to the point where we are given the velocity off the bat, the angle on the field, and the trajectory, we should have everything we need. Given time, a database could be built up that assesses plays for each player based on the frequency which they are turned into outs by all players at that position.

    Should be a way to truly evaluate defense in a way that people would buy into and get behind.

    Comment by Nathaniel Dawson — July 2, 2009 @ 4:00 am

  98. What a riveting debate you have started here Dave. I have been studying sabermetrics over the past couple of months and Defensive evaluations to me are the most interesting and perplexing aspects of Sabermetrics. I no longer view players strictly based on their offensive abilities.

    I really think you are on to something with this article; every talent evaluator or stat-head needs to bring defensive value into their line of sight. (I enjoyed your blind-spot reference).

    I truly hope that UZR can be refined to produce accurate representations of players run/win values. I do commend you on writing this article, because I am sure you knew how harsh some Saber’s would treat you afterwards.

    Comment by Matt B — July 2, 2009 @ 4:13 am

  99. “Dave — given that Morgan is such an elite defensive outfielder, would the Nationals be better served trading Dunn and also employing a Morgan-esque Juan Pierre in left field? While the Dodgers were ripped for putting Pierre’s awful bat in left, I would have to think that either Pierre or Morgan would go from a great defender in center to an elite one in left field, no?”

    My thought on this is that, unlike with batters, you eventually reach a point of diminishing returns when playing two elite fielders next to each other, particularly in the outfield. This was relatively clear in watching the Orioles early this season, with Pie in left field and Jones in center.

    In a batting lineup, if you acquire an elite hitter to hit behind a similarly elite hitter who was already on the team, there is no substantial effect on the incumbent player. Since an at-bat is a (mostly) independent event, the presence of one elite hitter will not affect the amount of contribution that can be made by another.

    In the field, though, plays are not independent events, since defensive players work simultaneously as a unit, especially in the outfield. It seems likely to me that the presence of an elite fielder in, say, center field would limit the amount of value an elite left fielder could provide. There are so many balls that can be caught, after all, and if one player is successful at catching the vast majority of balls in his zone and many balls OUT of his zone, how much additional positive impact can the players next to him really make?

    Comment by Eric — July 2, 2009 @ 10:55 am

  100. As a local, I have to agree with this post. Taking Hanrahan/Burnett first. The Nats absolutely HAD to get rid of Hanrahan. He was destroying fan morale, and probably that of the young starting pitcher rotation. Even if the save/closer potential is higher with Hanrahan, he wasn’t making it this year.

    As far as Morgan, again, you have to put some wins on the table–THIS YEAR. Not 5 years from now, NOW. They don’t even have to play .500 ball, but just so that you don’t have to buy a Powerball lottery ticket with a better chance of winning than going to a game.

    Otherwise, the club is going to lose the fan base, and end up with average 10K attendance. So you do have to look at some short-term moves.

    Comment by NatsLady — July 2, 2009 @ 10:56 am

  101. This comment of Sekrah’s is the sort of blunt-force stupidity that is ruining our ability to have even a semi-rational conversation in this country. (“He published something I don’t like, therefore I have the right to be irrationally vicious to him!” Ehh, go soak your head.)

    Comment by The Ancient Mariner — July 2, 2009 @ 11:45 am

  102. I’m shocked that anyone would think that Dave rooting for the M’s would have any logical influence on his evaluation of a deal between two NL teams on the opposite coast that does not in any way affect anything that has any significant relationship to the Seattle Mariners. Pretzel “causality” at its finest: the paranoid style in sports-blog-post-commenting at work.

    Comment by The Ancient Mariner — July 2, 2009 @ 11:47 am

  103. “I really think you are on to something with this article; every talent evaluator or stat-head needs to bring defensive value into their line of sight. (I enjoyed your blind-spot reference).”

    What he’s onto is a straw man, and he’s all over it. If this article came out 8 years ago, he’d still demonstrate a glaring misunderstanding of the range of true-talent defensive performance, but he’d have a point.

    It’s been years and years since the sabermetric community-at-large overrated Dunn, or systematically underrated defensive specialists. I hope when Cameron tears us all to shreds for our attachment to range factor in his next column, he at least demonstrates an understanding of how it should be regressed.

    Comment by cpebbles — July 2, 2009 @ 1:24 pm

  104. Dave,

    I tend to agree about the undervalued-ness of defensive metrics, but I think you may lean too far the other way here. I tend to agree with the commenter who said that the confidence in and proven utility of the metrics should factor in to how you weight them. The offensive metrics have many more years of proven utility and proven predictability. Defensive metrics still suffer from lots of noise. Regressing Morgan’s defensive metrics for sample size is necessary, but I think the noise in the metrics may also argue against using a direct offense v. defense comparison in this way.

    Moreover, this also brings up one of my main methodological misgivings about both the offensive and defensive metrics. I know both are denominated in “runs” but they sort of measure different things – the offensive metrics measure runs created, and so can be “normed” to the overall level of runs scored in a given year – the size and scope of the contributions can be kept within the overall context. But there’s really no way to measure, in the aggregate, how many runs are prevented each year – so there’s no context for whether a run prevented is really the equivalent of a run created offensively.

    So, that calls into question whether Morgan’s +35 defense is really only ten runs worse than Dunn’s +45 offense. 22,585 runs scored last year. How many were prevented? Without that comparison, we can’t really compare offense to defense this way – the underlying assumption that a scored run = a prevented run may or may not hold. It may actually be that Morgan’s +35 > Dunn’s +45, or that Dunn’s +45 is worth +90 runs defensively. The comparisons are always runs above average, but that average tends towards the mythical, since it’s not tied down in the reality of actual production, leaving open to question the comprability of the numbers across players and offense to defense.

    The difficulty, as Dave says, is in comparing players at the extremes. A team of all Dunns would undoubtedly put up great offensive numbers, but be awful in the field. A team of all Morgans would be equally ineffective the other way. I think the point that sabermetrics tends to value the more easily measurable (offense) and discount the more difficult, less confident more controversial (defense) is a good one, but I think you may have leaned too far to try to prove a unlikely supposition.

    Comment by Marc — July 2, 2009 @ 1:47 pm

  105. Hi.

    Comment by Johnny Damon — July 2, 2009 @ 3:08 pm

  106. Did you really compare Adam Dunn to Frank Thomas in his prime? Let me put it this way. Adam Dunn is just 29 years old, so he is still in the prime of his career. His career wOBA is .384. Frank Thomas’ career wOBA, which includes 8 seasons marred by injury and age, is .416. His numbers for the first 10.5 seasons of his career absolutely decimate Dunn’s career line. And Thomas played a far more passable first base than Dunn ever has. Thomas won a batting title and contended for even more, for crying out loud.

    Dunn’s consistent lack of contact combined with patience and power make him an above-average offensive player. His abject inability to play passable defense at any position undercuts his overall value severely. Frank Thomas during the first 10.5 seasons of his career, a point which Dunn is fast approaching, was as good as Albert Pujols. Think about that before you make such a ridiculous comparison again.

    Comment by Alireza — July 3, 2009 @ 6:22 am

  107. Something small to consider, just as a thought I had:

    While the metrics used here suggest that they have equal value within the Nationals’ system, could there also be a positive effect on Dunn’s defensive value if he has an elite fielder in center field to cover some of his poor defensive ability?

    Even if it is a small difference, Dunn may be relied upon less to make plays in the field. This, in turn, could make up for at least some of his negative impact on defense. While in turn, it’s possible that Morgan’s defensive measures will decrease with his added responsibility on the field. Morgan had the benefit of playing alongside an amazing defensive CF for a few weeks in Andrew McCutchen in Pittsburgh, and while Nate McClouth was measuring pretty poor defensively himself, he at least had the speed to cover ground in CF as well. The downgrade in the defensive ability from McCutchen/McLouth to Dunn is significant.

    I’m not an active follower of sabermetrics, and I suppose this is more of an intangible argument than something that would be covered by sabermetrics, but I believe it is something to consider.

    Comment by Josh — July 3, 2009 @ 9:21 am

  108. Well, I wouldn’t say horrifically irresponsible, except for fact Dave said the Rockies’ managers should be embarrassed by how they didn’t start Seth Smith over Spillborghs based on 150 at bats.

    Comment by wobatus — July 5, 2009 @ 10:55 pm

  109. Garrett– UZR does not evaluate an outfielder’s arm. Check out Hardball Times’s stats page for an attempt to value OF arms.

    It can make a significant difference in a player’s value. Michael Cuddyer has a great arm and awful range, so just using range metrics makes him look like one of the worst OF in the game (instead of “slightly below average”).

    Comment by Paul Thomas — July 6, 2009 @ 1:10 pm

  110. I am a big Nats fan and I would rather have Morgan than Dunn anyday. We win more often when Dunn does not play than when he does play. His defense is the worst I have ever seen, and unfortunately he is clearly trying as hard as he can on defense. He is just really really baaaad. We won 2 out of three out of the Yankees in NY and 2 of three from the Blue Jays in Toronto because he was our DH. but the second he plays in the outfield we give up run after runon routine plays to the outfield. Dunn is a worse fielder than any other DH in the game. And yes he is no more than a DH. I would play the big hurt in LF before him.

    Oh and by the way Millage has got to be nearly as bad as Dunn. All 7 of our first games this season Millage lost the game with a blown play in the outfield at least once. My memories of him are Millage running in semi-circles while the ball flew over his head repeatedly. And he always grounded out on the first pitch if someone was on base. Pitsburg could not have been dumber with this trade!!!!!

    Comment by PhDBrian — July 6, 2009 @ 3:16 pm

  111. This comparison isn’t looking so “equal” anymore.

    Comment by WY — September 2, 2010 @ 5:42 pm

  112. This article is amazing in hindsight.

    Comment by Joel — March 27, 2011 @ 1:01 pm

  113. Bump

    Comment by Nailed It — August 1, 2011 @ 9:06 am

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>


Current day month ye@r *

Close this window.

0.143 Powered by WordPress