Oh, the Humanities

As I tweeted a week or so ago, this was a good season for the part of me that is a Tigers' fan to miss. I have been dealing with a return of my wife's cancer (the outlook is not great but, as the last lines of the original theatrical release of Blade Runner go, "I didn't know how long we had together... Who does?"), in addition to moving house (and changing countries). However, I accumulated a few bookmarks and other ideas to work through, especially now we can only watch other teams in the post-season.

While I was busy, a very important blog post was made back in May. Phil Birnbaum, who is nothing if not insightful in writing about sabermetrics, announced that dWAR, a measure of fielding value, seemed to him to have a significant problem. Birnbaum proposed that dWAR inherently overvalued fielding. Birnbaum's argument is rooted in mathematical accuracy, so I don't feel confident trying to explain it. If you haven't read the post already, you should go to his blog to read how he explains it.

However, his explanation boils down to three key points, if we focus on the effects:

a) the runs allocated to the fielders under dWAR are too high, by an order of around fifty percent. (So a team dWAR of -40 is actually more like -20

b) The cause of this is that when one assumes "certain balls in play are the same" (as one has to do with older baseball statistics) then the math sends all the credit to the fielders.

c)

"Observations are a combination of talent and luck. If you want to divide the observed balls in play into observed pitching and observed fielding, you're also going to have to divide the luck properly."

Here, I think, we run into the problem of "All things being equal", or the distinction that the philosopher of history R.G. Collingwood made between meteorology and chemistry. It is an essential fact of human life that all things are NOT equal. People working in meteorology can collect observations of events, but cannot reproduce them at will, unlike people working in chemistry. By contrast, the historian can observe events, but they cannot create political or social crises at will, nor send qualified observers back into the past in order to collect the information needed to understand those events in the way scientists might send an expedition to view an eclipse or collect specimens. In scoring a baseball game, at best a sabermetrician can be a weatherman.

One can take issue with the statement "the assassination of Archduke Franz Ferdinand of Austria on 28 June 1914 triggered the First World War" as one of causality, but without doubt the shooting set off a diplomatic crisis that led to the war. More importantly, luck played a crucial role in the event because the Archduke's car came to a complete halt very close to where the "Yugoslav nationalist" Gavrilo Princip, had stationed himself. An earlier attempt to kill the Archduke in a moving car had failed. We have no idea whether Princip could have been successful if his targets had been in a moving car. So, what percentage of responsibility to the war do we assign to Princip, to the driver, to the governor of Bosnia at whose orders the driver stopped, to the Serbian officers who conspired to arm Princip, to the Archduke or to the general diplomatic situation? And any formula that did allocate "responsibility shares" to these people would be essentially an act of faith.

Birnbaum went on to add some further details to his understanding in a threat on the blog of Tom Tango, the tremendously influential pseudonymous saberist. In the comments section of Tango's thread on the post by Birnbaum, Birnbaum suggested in one reply that it was just not possible for a system like Defensive Runs Saved or Ultimate Zone Rating to make distinctions about balls in play that could tell us something about the skill of the fielder.But before that he stated that he wanted to assign the luck to the pitcher. However, reading the comments there is to venture into a world where something like the Responsibility Shares is thought to be possible. Possibly, with enough computing power, such things can be made for evaluating baseball players. But I can't help but think the effect will be small.

To reduce Birnbaum's position down, what he thinks is that about half of the dWAR effects at the team level need to be transferred from the fielder to the pitcher. Another way to think about it is that he wants a cap on the amount of Runs Allowed value distributed to the fielders. But this would also have effects on how we value players. A quick-and-dirty method would be to halve the UZR assigned to any player when calculating their WAR, although I suspect Birnbaum would object on the grounds that something true at the team level may not be true at the level of the individual player.

2014 Nationals' Fielding Review #6

Here is an update to last time's fielding numbers. My source for this is Fangraphs, which includes all the main metrics that interest me except for Michael Humphreys' Defensive Regression Analysis.* From Fangraphs, I've used Mitchel Lichtman's Ultimate Zone Rating, my personal first choice of the 'converted-to-runs-play-by-play' metrics, and my preferred measure of RZR. RZR is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached. I have included the MLB positional averages for RZR, to help give the players' numbers some context. Note that catchers do not have a Zone Rating. Instead, I have used the runs saved by framing, supplied by StatCorner.com

Player              UZR    Change    RZR   Change    LgAverage      DRA    Change   PFr
Lobaton (C)         n/a      --      n/a     ---         n/a        7.0     +0.6   -1.7
Ramos (C)           n/a      --      n/a     ---         n/a        5.7     +1.3   -2.0
Span (CF)          -0.4     +1.5    .934    +.016       .923        2.6     +2.9    n/a
Espinosa (2B)       0.1     -1.1    .788    +.004       .788        5.7     +2.1    n/a
LaRoche (1B)       -1.9     +2.4    .800    +.004       .807      - 8.0      ---    n/a
Harper (LF)        -1.3     +0.1    .872    +.023       .871      - 5.5     -2.1    n/a
Desmond (SS)       -2.3     +0.3    .803    +.005       .793      -10.8     -0.4    n/a
Werth (RF)         -4.5     -2.9    .933    +.004       .895      - 8.6     -3.5    n/a
Rendon (3B)         0.4     +0.7    .699    -.006       .708        2.4     +1.4    n/a
Zimmerman (LF)      0.4     ---     .905    +.003       .871        0.6     -1.6    n/a
Rendon (2B)         2.0     ---     .857     ---        .788      - 8.0     ---     n/a 
minimum 180 innings

By UZR Denard Span continues to improve, by exactly the same amount as last time. Danny Espinosa gave back all of his UZR improvement from the last review. Span and Anthony Rendon at third are the two players seen as improving both by the traditional statistics of DRA and by the play-by-play analysis of UZR. The only player losing ground under both systems was Jayson Werth. Rendon is now in the list as a secondbaseman as well as a thirdbaseman. Ryan Zimmerman at third just missed the cut.

Overall, both systems see the Nationals as seemingly having to found their level for the season. UZR adds up to a 0.7 increase, which is probably indicative of random variaton. DRA, including catchers, sees an identical improvement. UZR adds up the Nationals' regular fielders to -7.8, or costing the team about three-quarters of a win. DRA sees matters almost twice as bad, with -16.9 runs, or almost two wins. However, one probably ought to double the UZR total to get a better idea of the team's fielding over an entire season.

Wilson Ramos' pitch framing went down las time by 2.5 runs and this time by 1.4 runs. Jose Lobaton also slipped a bit, losing 0.4 runs, giving back some of his improvement last time. So, overall, the Nationals' catchers haven't been able to get the calls during this past two weeks or so.
____
* These are available at the Baseball Gauge of Seamheads.com. Humphreys wrote the excellent Wizardry, which is a way of looking at fielding using only the traditional statistics, and not the newfangled play-by-play metrics.

2014 Tigers' Fielding Review #3

Here is an update to my regular survey of the Tigers' fielding. My source for this is Fangraphs, which includes all the main metrics that interest me except for Michael Humphreys' Defensive Regression Analysis.* From Fangraphs, I've used Mitchel Lichtman's Ultimate Zone Rating, my personal first choice of the 'converted-to-runs-play-by-play' metrics, and my preferred measure of RZR. The last is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached. I have included the league positional averages for RZR, to help give the players' numbers some context. DRA is added to these two, while I have dropped Dewan's Defensive Runs Saved. Note that catchers do not have a Zone Rating. Instead, I have used the runs saved by framing, supplied by StatCorner.com

Player              UZR    Change    RZR   Change    ALAverage     DRA    Change   PFr
Avila   (C)         n/a      --      n/a     ---         n/a      - 3.5   -2.2     -8.1
Holaday (C)         n/a      --      n/a     ---         n/a      - 1.1    0       -4.2
Kinsler (2B)        5.2     -1.2    .826   +.004        .799      - 7.4   -4.3      n/a
Jackson (CF)       -4.6     -0.7    .920   +.004        .909      - 2.2   +1.9      n/a
Cabrera (1B)        6.3     +0.1    .825   +.013        .800       18.1   +4.3      n/a
JD Martinez (LF)   -2.4     +0.3    .875   +.014        .881      - 3.2   +0.9      n/a
Rajai  (LF)        -3.6     +3.3    .856   +.008        .881      - 0.1   -2.2      n/a
Romine (SS)        -2.1     +0.6    .722   +.011        .752      - 4.2   +1.0      n/a
Hunter (RF)       -12.0     +1.9    .863   +.022        .888      - 9.0   +2.0      n/a
Castellanos (3B)   -9.1     -2.7    .616   +.005        .705      - 2.7   +2.1      n/a
Suarez (SS)         0.3     +0.1    .808   +.119        .752        2.8   +1.8      n/a
minimum 170 innings

RZR continues its positive views on the Tigers' fielding, seeing improvement across the board. Corner outfield and third base still are below the league average, but everywhere else the Tigers are getting to more balls in the zone than the average player at the position.

UZR sees an overall improvement, up by 1.7 runs. The extreme results see Rajai Davis playing better, while Nick Castellanos plays worse. DRA sees even more improvement, with the team overall up by 5.3 runs, or half a win. However, we ought to bear in mind that the last review showed a sharp fall in Tigers' fielding, and they have only made up part of that ground. A good part of the gains come from an improved showing by Torii Hunter, who is up by sizeable amounts across the board. I can't help but look at DRA and think that some kind of accounting trick has transferred some of Ian Kinsler's fielding to Miguel Cabrera. We might want to keep an eye on that.

The sharpest divergence comes at left field and third base, with Rajai Davis' UZR way up, while his DRA rating is down. Castellanos and, to a lesser extent, Austin Jackson reverse that.

On pitch framing, Alex Avila's previous improvement was largely undone, losing -1.1 runs. Brian Holaday also fell by 0.8.
____
* These are available at the Baseball Gauge of Seamheads.com. Humphreys wrote the excellent Wizardry, which is a way of looking at fielding using only the traditional statistics, and not the newfangled play-by-play metrics.

2014 Nationals Fielding Review #5

Here is an update to last time's fielding numbers. My source for this is Fangraphs, which includes all the main metrics that interest me except for Michael Humphreys' Defensive Regression Analysis.* From Fangraphs, I've used Mitchel Lichtman's Ultimate Zone Rating, my personal first choice of the 'converted-to-runs-play-by-play' metrics, and my preferred measure of RZR. RZR is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached. I have included the MLB positional averages for RZR, to help give the players' numbers some context. Note that catchers do not have a Zone Rating. Instead, I have used the runs saved by framing, supplied by StatCorner.com

Player              UZR    Change    RZR   Change    LgAverage      DRA    Change   PFr
Lobaton (C)         n/a      --      n/a     ---         n/a        6.4     -0.2   -1.3
Ramos (C)           n/a      --      n/a     ---         n/a        4.4     -0.6   -0.6
Span (CF)          -1.9     +1.5    .918    -.001       .923      - 0.3     +3.6    n/a
Espinosa (2B)       1.2     +1.1    .784    -.006       .788        3.6     +2.5    n/a
LaRoche (1B)       -4.3     -2.9    .796    -.008       .807      - 8.0     -2.2    n/a
McLouth (LF)       -0.6     +0.1    .880     ---        .871      - 1.2     +0.1    n/a
Harper (LF)        -1.4     +0.2    .849    -.029       .871      - 3.4     -0.5    n/a
Desmond (SS)       -2.6     +1.3    .798    -.001       .793      -10.4     -1.7    n/a
Werth (RF)         -1.6     +0.7    .929    -.004       .895      - 5.1     -1.1    n/a
Rendon (3B)        -0.3     -0.8    .705    -.008       .708        1.0     +0.6    n/a
Zimmerman (LF)      0.4     -0.3    .902    -.069       .871        2.2     -1.9    n/a
minimum 170 innings

RZR thinks the Nationals' fielding has decline across the board, although as usual one must also take note of the league averages when using RZR. Comparing individual players with their league peers suggests that overall the Nationals fielding stands comparison, although decline in RZR by Adam LaRoche continues.

UZR and DRA diverge in their opinions of Bryce Harper (UZR up, DRA down), Ian Desmond (ditto), Jaysotops the list of those rising n Werth (ditto again), and Anthony Rendon (UZR down, DRA up). LaRoche leads the declines in both of these systems, while Denard Span is tops on the list of improvers. UZR thinks that overall the Nationals' regulars have improved again, up by 0.9. DRA sees a slight improvement, up by 1.1, but that still leaves the bulk of the Nationals' sharp fall last time weighing them down.

Ramos' pitch framing went down quite a bit, by 2.5 runs. Lobaton improved slightly, by 1.4. So, overall, the Nationals' catchers haven't been able to get the calls during this past two weeks or so.
____
* These are available at the Baseball Gauge of Seamheads.com. Humphreys wrote the excellent Wizardry, which is a way of looking at fielding using only the traditional statistics, and not the newfangled play-by-play metrics.

2014 Tigers' Fielding Review #2

Here, a couple of days late, is an update to my regular survey of the Tigers' fielding. My source for this is Fangraphs, which includes all the main metrics that interest me except for Michael Humphreys' Defensive Regression Analysis.* From Fangraphs, I've used Mitchel Lichtman's Ultimate Zone Rating, my personal first choice of the 'converted-to-runs-play-by-play' metrics, and my preferred measure of RZR. The last is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached. I have included the league positional averages for RZR, to help give the players' numbers some context. DRA is added to these two, while I have dropped Dewan's Defensive Runs Saved. Note that catchers do not have a Zone Rating. Instead, I have used the runs saved by framing, supplied by StatCorner.com

Player              UZR    Change    RZR   Change    ALAverage     DRA    Change   PFr
Avila   (C)         n/a      --      n/a     ---         n/a      - 1.3   -2.3     -7.0
Holaday (C)         n/a      --      n/a     ---         n/a      - 1.1   -1.5     -3.4
Kinsler (2B)        6.4     +2.2    .822   +.014        .799      - 3.1   -1.9      n/a
Jackson (CF)       -3.9     -4.7    .916   -.001        .909      - 4.1   -1.9      n/a
Cabrera (1B)        6.3     +2.5    .812   -.011        .800       13.8   +4.5      n/a
JD Martinez (LF)   -2.4     -0.8    .861   -.039        .881      - 4.1   -4.5      n/a
Davis  (LF)        -6.9     +0.6    .848   +.010        .881        2.1   +1.8      n/a
Romine (SS)        -2.7     -0.2    .711   +.002        .752      - 5.2   -0.8      n/a
Hunter (RF)       -13.9     +0.1    .841   +.016        .888      -11.0   -1.6      n/a
Castellanos (3B)   -6.4     -5.2    .611   -.004        .705      - 4.8   -1.4      n/a
Suarez (SS)         0.2      --     .769     ---        .752        1.0     --      n/a
minimum 160 innings

During the past two weeks, the metrics that convert fielding data to runs show the Tigers' fielding has mostly deteriorated quite sharply. UZR sees the Tigers' numbers decline by 5.5, or half a win. DRA paints an even worse picture, with a decline of 9.6 runs, or almost an entire win. The same broad outlines discussed last time still persist. The Tigers' outfielders are not very good defenders, UZR evaluating them at -27.1 and DRA at -17.1. The left side of the infield is also a point of weakness (UZR -8.9, DRA -9), while the right side is the Tigers' greatest strength in the field (UZR +8.5, DRA +10.7). Mostly the systems agree with each other with the exception of Ian Kinsler, whom UZR sees as a positive fielder, but DRA thinks is a negative one. The opposite applies to Rajai Davis.

However, RZR paints a more positive picture. Kinsler, Austin Jackson and Miguel Cabrera have been all above-average fielders at their position. The systems divide over Jackson, who is an asset under RZR, and a negative factor under the other two. Looking at the UZR for all AL centrefielders, it seems that while UZR thinks Jackson has good range, his arm is below-average and he makes too many errors.

A brief word about Eugenio Suarez: so far all the systems see him as above average. Suarez Runs Created are also higher than other shortstops the Tigers have tried out as regulars. Perhaps a lot of the credit for the Tigers' recovery from a late-spring swoon is down to him.

On pitch framing, Alex Avila has improved, but Bryan Holaday has got a lot worse.
____
* These are available at the Baseball Gauge of Seamheads.com. Humphreys wrote the excellent Wizardry, which is a way of looking at fielding using only the traditional statistics, and not the newfangled play-by-play metrics.

2014 Tigers Fielding Review #1

I am adding the Tigers to my survey of fielding numbers, which I have been meaning to do since the start of providing these. So I will alternate the Tigers and the Nationals, with updates every other week. My source for this is Fangraphs, which includes all the main metrics that interest me except for Michael Humphreys' Defensive Regression Analysis.* From Fangraphs, I've used Mitchel Lichtman's Ultimate Zone Rating, my personal first choice of the 'converted-to-runs-play-by-play' metrics, and my preferred measure of RZR. The last is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached. I have included the MLB positional averages for RZR, to help give the players' numbers some context. DRA is added to these two, while I have dropped Dewan's Defensive Runs Saved. Note that catchers do not have a Zone Rating. Instead, I have used the runs saved by framing, supplied by StatCorner.com

Player              UZR    Change    RZR   Change    MLBaverage     DRA    Change   PFr
Avila   (C)         n/a      --      n/a     ---         n/a        1.0     --     -8.8
Holaday (C)         n/a      --      n/a     ---         n/a        0.4     --     -2.0
Kinsler (2B)        5.2      --     .822     ---        .789      - 1.2     --      n/a
Jackson (CF)       -3.9      --     .917     ---        .924      - 2.2     --      n/a
Cabrera (1B)        3.8      --     .823     ---        .812        9.3     --      n/a
JD Martinez (LF)   -1.6      --     .900     ---        .872        0.4     --      n/a
Davis  (LF)        -4.5      --     .838     ---        .872        0.3     --      n/a
Romine (SS)        -2.7      --     .709     ---        .775      - 4.4     --      n/a
Hunter (RF)       -14.0      --     .825     ---        .895      - 9.4     --      n/a
Castellanos (3B)   -1.2      --     .615     ---        .709      - 3.4     --      n/a
minimum 130 innings

Without the same amount of information monitored as this blog has for the Nationals, it is impossible to ascertain whether the Tigers' fall-off in performance is associated with bad fielding. However, there are several obvious pieces of information to draw from these numbers. One is that the Tigers' outfielders aren't doing very well. On all measures we see negative numbers except for J.D. Martinez' RZR and DRA, and Rajai Davis' DRA. But the DRA numbers aren't much above average.

The infield pffers a slightly better picture, with Ian Kinsler and Miguel Cabrera doing well under UZR and DRA is really enthusiastic about Cabrera. However, RZR sees the left side of the infield as quite a problem, with both Andrew Romine and Nick Castellanos well under the league average.

I do have data for overall team fielding going back some weeks. The Tigers' team UZR has fallen quite sharply over the last ten days to two weeks, from average-ish to moderately bad. (UZR thinks the worse teams are the Indians and Astros). The Tigers are probably at the fringe of what is tolerable. Defensive Runs Saved sees the picture a bit worse, with the Tigers being the fourth-worst team in MLB. RZR shares UZR's ranking.

On pitch framing, Alex Avila is one of the worst in baseball, with only three catchers with lower scores.
____
* These are available at the Baseball Gauge of Seamheads.com. Humphreys wrote the excellent Wizardry, which is a way of looking at fielding using only the traditional statistics, and not the newfangled play-by-play metrics.

2014 Nationals Fielding Review #3

Here is an update to last time's fielding numbers. My source for this is Fangraphs, which includes all the main metrics that interest me except for Michael Humphreys' Defensive Regression Analysis.* From Fangraphs, I've used Mitchel Lichtman's Ultimate Zone Rating, my personal first choice of the 'converted-to-runs-play-by-play' metrics, and my preferred measure of RZR. The last is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached. I have included the MLB positional averages for RZR, to help give the players' numbers some context. DRA is added to these two, while I have dropped Dewan's Defensive Runs Saved. Note that catchers do not have a Zone Rating. Instead, I have used the runs saved by framing, supplied by StatCorner.com

Player              UZR    Change    RZR   Change    MLBaverage     DRA    Change   PFr
Lobaton (C)         n/a      --      n/a     ---         n/a        6.3     +0.2   -1.2
Ramos (C)           n/a      --      n/a     ---         n/a        2.7     --     -0.3
Span (CF)          -1.4     +1.2    .929    -.003       .914      - 3.6     +0.4    n/a
Espinosa (2B)       0.0     +0.3    .779    +.013       .789        2.3     -2.2    n/a
LaRoche (1B)       -2.3     -2.1    .868    +.022       .812      - 1.1     +2.9    n/a
McLouth (LF)       -1.0      --     .880     ---        .872      - 0.8     --      n/a
Harper (LF)        -1.7     -0.2    .875     ---        .872      - 2.6     +0.8    n/a
Desmond (SS)       -4.6     -0.1    .795    +.033       .775      -10.7     +0.8    n/a
Werth (RF)         -1.3     -3.5    .932    -.023       .895      - 0.8     +2.0    n/a
Rendon (3B)        -1.2     +2.8    .707    +.009       .709        1.0     +1.5    n/a
minimum 130 innings

The abysmal Nationals' fielding of the early season now seems to have done a little bit of 'regression to the mean, no matter what particular statistic one looks at. We see some extreme results, with DRA, the method using traditional fielding statistics, seeing improvement from Adam LaRoche and Jayson Werth, while UZR takes a remarkably opposed view. RZR splits the difference, with one up and one down. Interestingly, RZR now sees Ian Desmond as an above-average shortstop, while DRA and RZR think he is improved but still the worst regular on the team. Anthony Rendon has improved quite dramatically with more time at third base.

If one sums the UZR changes, the Nationals' regulars have deteriorated by about -1.6 runs, and locates the problem on the right side of the field, with Werth and LaRoche dampening the improvement elsewhere. DRA sees a 6.1-run improvement, which is half a win. The Nationals are winning more, too. Once again, my own impression aligns more with DRA.

As a side note, the pitch-framing data for the catchers shows some improvement for Lobaton, who scored -2.1 last time.
____
* These are available at the Baseball Gauge of Seamheads.com. Humphreys wrote the excellent Wizardry, which is a way of looking at fielding using only the traditional statistics, and not the newfangled play-by-play metrics.

2014 Nationals' Fielding Review #1

With a month of the season completed, it is time to start reviewing the fielding data periodically. My source for this is Fangraphs, which includes all the main metrics that interest me. I've used John Dewan's Defensive Runs Saved, Mitchel Lichtman's Ultimate Zone Rating, my personal first choice of the 'converted-to-runs-play-by-play' metrics, and my preferred measure of RZR. The last is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached. I have included the MLB positional averages for RZR, to help give the players' numbers some context. Note that catchers do not have a Zone Rating. I'll do something about pitch-framing runs saved another time, perhaps.

Player              DRS   Change  UZR      Change  RZR   Change   MLBaverage   DRA
Lobaton (C)         -1    --       n/a      --     n/a     ---         n/a      5.9
Span (CF)            1    --      -0.4      --    .922     ---        .912     -2.9  
Espinosa (2B)       -3    --      -1.0      --    .757     ---        .785      4.4
LaRoche (1B)         0    --      -0.2      --    .846     ---        .814     -3.2
Harper (LF)          1    --      -1.8      --    .875     ---        .865     -2.9
Desmond (SS)        -4    --      -4.6      --    .740     ---        .768     -8.7
Werth (RF)          -7    --      -1.9      --    .959     ---        .886     -1.2
Rendon (3B)         -3    --      -5.0      --    .642     ---        .721     -2.4
minimum 120 innings

I was a bit shocked that Ryan Zimmerman was absent even with a 70-inning minimum. Has he really been gone that long?

What the RZR data highlights is how poor the Nationals' infield defence is, relative to the rest of the league. Anthony Rendon in particular is hurting the Nationals' at third base. Jayson Werth doesn't look so bad in RZR, but his DRS number is a bit disturbing.

UPDATE: I have added the one-year numbers of Michael Humphrey's Defensive Regression Analysis system, available at the Baseball Gauge of Seamheads.com, Humphreys wrote the excellent Wizardry, which is a way of looking at fielding using only the traditional statistics, and not the newfangled play-by-play metrics.

Nationals' Fielding Review #2

Being five hours ahead of North America is making it difficult for me to keep up with the games, although that's partly because I have to help with homework and read bedtime stories, as well as going up to the University of London during the day.

Anyway, here is some fielding data. I've replaced the in-season cumulative UZR with Chris Dial's Defensive Runs Saved. You can find an explanation of what he's up to here. UZR/150 is the current UZR extrapolated to 150 games. RZR is Revised Zone Rating, which is like a fielding average but counts balls hit into a zone, rather than those the fielder actually reached.

Player              DRS   Change  UZR/150  Change  RZR   Change   MLBaverage
Zimmerman (3B)       4    +2      24.7     +3.0   .755    +.008       .704
Desmond (SS)         3    -2      14.8     +0.2   .833    -.034       .811
Gonzalez (3b)        3    -1      44.0     -1.9   .882    +.007       .704
Maxwell (RF)         2     -      21.6     -6.1   .857    +.018       .885
Harris (RF)          0     -      13.3     -0.7   .852    +.044       .885
Kennedy (2b)         0     -     - 4.8    -12.2   .800    -.039       .819
Guzmán (SS)          1    +1       8.9     +1.8   .750    -.017       .811
Willingham (LF)     -2     -       5.8     +4.9   .880    +.021       .871
Guzmán (2B)          3    +3     - 2.6     +0.5   .839    +.022       .819
Dunn (1B)           -2    +1     - 2.7     +2.4   .776    -.006       .772
Bernadina (RF)       1    +1     -11.3     +3.0   .915    +.015       .885
Morgan (CF)          0    -1     -11.8     +0.3   .883    +.009       .916
minimum 73 innings

Harper commented after the last fielding review on how we need more like three years' data to get a sense of a player's talent as a fielder. This is true, but one of my old-fashioned sabermetric principles is that we should be more interested, as fans, in what happened rather than what we can expect to happen. Over three years, Dunn might turn out to be the worst first-baseman in the league. However, for the moment he's been doing better than I thought we had any right to expect. I'll keep singing praises for his effort and dedication so long as he keeps that up — and even if he starts having difficulties, because he's clearly trying his best. Jolly good show!

2009 Nationals' Postmortem #1 - Fielding Overview

Sorry about not keeping up with the Fielding Weekly for the last seven weeks of the season. The trip to England messed up my routine too much, and then returning my roots to the groves of academe didn't help either. I didn't pay much attention to baseball until the last week of the season.

With the mea culpa out the way, let's talk leather. No, not Jim Bowden's supposed trouser preferences, but the gloves of our ballplayers. Actually, maybe you wouldn't want me to draw attention to this. I've seen the figures already.

To recapitulate, the columns below show the Nationals' fielding as measured by Ultimate Zone Rating (UZR), a fielding metric devised by Mitchell Lichtman of The Book fame. It's widely accepted as the 'gold standard' of fielding. The first column is 'raw UZR', which measures the total effect of a player's fielding at a given position, in terms of runs above or below a notional 'average fielder' over the innings he played this past season. The second column is the gain or loss since the last time I did this exercise. The third column is UZR/150, or the total UZR score normalized to what it would be over 150 games. It has a comparative value—you immediately can see how one player at a position compares to another. The fourth column is the gain or loss in UZR/150 since last time. Next we have Revised Zone Rating RZR, which measures how many times a player caught the ball, given the chances to catch the ball. Then the gain or loss, etc. Finally, the average RZR at that player's position in MLB. The order is total innings played at that position. The guys in bold are the biggest gainers in each of the three columns. The cut-off point is 120 innings.

Player              UZR   Change  UZR/150  Change  RZR   Change   MLBaverage
Zimmerman (3B)     17.9   + 5.2    14.1    - 2.8  .748   -.021     .712
Guzmán (SS)       - 2.5   + 1.9   - 1.8    + 1.2  .785   +.016     .801
Johnson (1B)      - 5.8   + 0.2   - 5.9    + 0.9  .776   +.006     .782
Willingham (LF)   - 4.7   - 2.3   - 7.1    - 1.1  .925   +.006     .894
Dukes (RF)          3.6   - 3.6   - 5.8    - 4.3  .893   -.005     .908
Dunn (1b)         -13.9   - 6.6   -18.7    +16.7  .651   +.115     .782
Dunn (LF)         -14.4   - 2.0   -28.4    - 4.1  .832   n.c.      .894
Hernandez (2B)      2.0   + 0.8     3.7    + 1.4  .859   +.005     .813
Harris (CF)       - 6.1   - 1.0   -13.6    + 2.4  .935   -.003     .931
Morgan (CF)        13.0   - 0.2    31.7    - 4.2  .960   -.004     .931
Kearns (RF)         2.1   + 0.4    11.1    + 5.8  .893   n.c.      .908
Gonzalez (2b)     - 3.0   + 0.8   - 5.8    +10.5  .775   +.018     .813
Belliard (2B)       2.2   - 1.1     5.0    - 6.3  .871   -.022     .813
Gonzalez (SS)     - 3.6   + 2.2   -26.2    +13.8  .691   +.043     .801
Dukes (CF)        - 4.8   + 0.3   -15.1    + 3.5  .898   n.c.      .931
Willingham (RF)   - 1.0   - 0.3    -5.3    - 3.0  .966   +.004     .905
Maxwell (CF)        5.0   n/a      26.2     n/a   .932   n/a       .931
Dunn (RF)         - 8.1   + 0.2   -33.0    + 5.8  .761   n.c.      .908
Harris (LF)         1.9   n/a      12.6     n/a   .846   n/a       .894
Desmond (ss)        0.3   n/a       0.9     n/a   .738   n/a       .801
Orr (2B)            3.9   n/a      27.4     n/a   .891   n/a       .813

Trying to estimate the impact of fielding, of course, is an inexact business. UZR doesn't think Willingham is as good a fielder as RZR does, for example. Nonetheless, we can make some assessments.

1) Going by RZR, which for various reasons I prefer, the Nationals' fielding wasn't as bad as one might think, but it's really not at all good.
2) Adam Dunn dramatically improved as a firstbaseman; but he still was the worst in the league.
3) Alberto Gonzalez was not an improvement over Anderson Hernandez.
4) Desmond probably isn't a significant improvement over Guzmán at shortstop. It may be that Guzmán still ought to be the starting shortstop next season. Unless he's traded.
5) Zimmerman might be a Gold Glove thirdbaseman, but his statistics are not conclusive evidence. Nonetheless, he is a singular bright spot on the infield.
6) Dukes was an average fielder for a while, but by season's end he wasn't. Combine that with his batting problems, and one has to wonder if his off-field baggage really renders him a marginal major-leaguer.
7) Maxwell and Morgan are pretty good centrefielders.
8) Willingham is an enigma. Is he average or not?
9) Rizzo basically traded away his best infield options, where he had a choice, suggesting he doesn't rate fielding particularly highly.

I'll return to the fielding again, but my thoughts are currently moving toward the lineup.

Catching Up #2 - Nationals' Fielding 'Weekly' #8

Here's the latest fielding statistics for Ultimate Zone Rating (UZR) and Revised Zone Rating (RZR), as I've been keeping them throughout the season. There's loads of stuff I want to say, but if I say too much I'll steal some of my thunder from other 'Catching Up' posts I have planned. So let's try and stick to players still with the Nationals.

Adam Dunn is doing as well as I expected. It's just a terrible mistake to expect him to be a tolerable fielder at 1B. Gonzalez has defied confidence in his glove all year. I'm told Belliard is deceiving me with his good numbers. But, here we are, and there's little alternative to them.

When you think about what the Nationals have now, you'd have to say that overall their fielding is potentially worse than it was in the first half of the season, when it was bad enough. They consequences of various moves has been to upgrade the outfield, but to degrade the infield. Since the outfield was way below average, and the infield only a little bit below, I guess if you were paying the price in one place for an upgrade you'd do it like that. But they didn't. They dealt away better infielders and replaced them with weaker ones. The upgrade in the outfield came independently of the degraded infield.

Guzmán's improvement is their only sign that they might be saved from the potential disaster in the infield. Is he regressing to his true talent level? Or is this a misleading glimpse of sun during a perfect storm at sea? I'm putting my life jacket on, just to be safe.

Player              UZR   Change  UZR/150  Change  RZR   Change   MLBaverage
Zimmerman          12.7   + 3.3    16.9     +1.6  .748   +.008     .715
Johnson           - 6.0   + 0.4   - 6.8     +0.3  .770   +.006     .785
Guzman            - 4.4   + 1.9   - 3.0     +3.2  .769   +.010     .802
Dunn (LF)         -12.4   n.c.    -24.3     +0.2  .832   n.c.      .889
Hernandez           1.2   + 0.2     2.3     +0.2  .854   +.002     .814
Willingham (LF)   - 2.4   - 0.7   - 6.0     -0.8  .919   +.009     .889
Kearns (RF)         1.7   + 0.2     5.3     +1.9  .893   n.c.      .905
Morgan (CF)        13.2   + 6.0    35.9     +5.5  .964   +.007     .933
Harris (CF)       - 5.1   - 0.1   -16.0     -0.6  .938   n.c.      .933
Dukes (CF)        - 5.1   + 0.5   -18.6     +1.7  .898   n.c.      .933
Belliard (2B)       3.3   n.c.     11.3     -2.3  .893   +.003     .814
Dukes (RF)          0.0   - 0.5   - 1.5     -4.7  .898   -.078     .905
Gonzalez (SS)     - 5.8   - 0.6   -26.2     -1.5  .691   -.001     .802
Willingham (RF)   - 0.7   - 0.1    -2.3     -1.3  .962   +.002     .905
Dunn (1b)         - 7.3   - 3.8   -35.4     -3.6  .536   +.007     .785
Gonzalez (2b)     - 3.8   - 2.6   -16.3     -7.1  .757   -.009     .814
Dunn (RF)         - 8.3   - 0.9   -38.8     -4.2  .761   n.c.      .905

The table is in order of innings played at the position, with the cut-off at 180 innings. The biggest moves, either up or down, in each of the three measures are in bold.

UZR converts fielding statistics into a +/- rating reflecting the number of runs saved or given away. There is a basic version, which is based on the number of innings played by the player, and a normalized version that assumes everybody plays 150 games. It also is forced to sum to zero, so a player can have his UZR adjusted even when he doesn't play.
RZR is like a fielding 'batting average', dividing the number of plays made by the number of opportunities. Unlike old Zone Rating, it excludes plays made 'out of zone'. (Zone Rating divides the field into zones, and assigns responsibility among the players.) For comparative purposes, I've put at the end of each player's line the MLB average RZR for position.

Nationals 'Fielding Weekly' #7

This covers more than a week, owing to a delay caused by the SABR annual convention in Washington, D.C. The data is drawn from Fangraphs' Ultimate Zone Rating (UZR) data and Revised Zone Rating (RZR) at The Hardball Times. I've put the biggest movers in each category in bold.

Not such a strong sign of a 'Riggleman effect' on the fielding after two weeks. For some reason, Dunn looks good in UZR but not so good in RZR. This probably has some connection with different structures to the zones in each. I still believe that it is a mistake to consider Dunn an acceptable Major-League 1B. He does less damage in LF. However, the Nationals are committed to finding out if I'm wrong. Belliard puts up some good numbers at 2B, but that belies the evidence of last year.

Let's assess the UZR effect of trading Johnson and moving Dunn to 1B

Johnson 1B, Hernandez 2B, Dunn LF, Willingham RF  +10.6
Dunn 1B, Hernandez 2B, Willingham LF, Dukes RF    + 5.8
Dunn 1B, Belliard 2B, Willingham LF, Dukes RF     +17.3

As things stand, the Nationals are liable to be giving away half a win over 150 games with the new alignment, and Hernandez at 2B, or gaining over half a win with Belliard at 2B. Hernandez, however, appears to be the better hitter.

Player              UZR   Change  UZR/150  Change  RZR   Change   MLBaverage
Zimmerman          12.7   - 0.6    16.9     -2.5  .740   -.013     .717
Johnson           - 6.4   + 0.4   - 7.1     +0.7  .764   +.004     .785
Guzman            - 4.4   - 0.4   - 6.2     n.c.  .759   -.008     .803
Dunn (LF)         -12.4   - 0.4   -24.5     -1.4  .832   -.017     .890
Hernandez           1.0   n.c.      2.1     +0.1  .852   +.004     .815
Kearns (RF)         1.5   + 1.1     3.4     +4.5  .893   n.c.      .907
Willingham (LF)   - 1.7   - 0.4   - 5.2     -1.2  .910   +.002     .890
Harris (CF)       - 5.0   n.c.    -15.4     -0.4  .938   n.c.      .933
Dukes (CF)        - 5.6   n.c.    -20.3     +0.8  .898   n.c.      .933
Morgan (cf)         7.2   + 2.3    30.4     +3.9  .957   -.012     .933
Gonzalez (ss)     - 5.2   - 0.7   -24.7     -2.2  .692   +.004     .803
Belliard (2b)       3.3   + 1.8    13.6     -6.8  .890   +.001     .815
Willingham (RF)   - 0.6   + 0.7    -1.0     +3.4  .961   -.016     .907
Dunn (RF)         - 7.4   - 0.7   -34.6     -1.7  .761   +.005     .907
Dukes (RF)          0.5   + 0.7     3.2     +6.6  .960   +.008     .907
Gonzalez (2b)     - 1.2   + 0.7   - 9.2     +5.3  .766   -.028     .815
Dunn (1b)         - 3.5   + 0.1   -35.4     +9.1  .529   -.016     .785

The table is in order of innings played at the position, with the cut-off at 100 innings.

UZR converts fielding statistics into a +/- rating reflecting the number of runs saved or given away. There is a basic version, which is based on the number of innings played by the player, and a normalized version that assumes everybody plays 150 games. It also is forced to sum to zero, so a player can have his UZR adjusted even when he doesn't play.
RZR is like a fielding 'batting average', dividing the number of plays made by the number of opportunities. Unlike old Zone Rating, it excludes plays made 'out of zone'. (Zone Rating divides the field into zones, and assigns responsibility among the players.) For comparative purposes, I've put at the end of each player's line the MLB average RZR for position.

Nationals' Fielding 'Weekly' #6

I took a break in posting the Fielding Weekly for the Washington Nationals last week because there had only been four games after the All-Star Break. (Also, I had a lot of work on last week.) This week's figures could be read as giving some support to the notion that Mr Riggleman's new routine is having some effect, nothwithstanding this horror story. The problem with taking that too far is that Ultimate Zone Rating (UZR) is not so enthusiastic as Revised Zone Rating (RZR). See an explanation for these stats after the table.

The big movers are Dunn, Dunn and Morgan, which sounds like an old-fashioned firm of stockbrokers. Morgan's really made a big difference, and I wonder if he's responsible for some of Dunn's improvement in leftfield this week. Anyone who tells you to trade Nick Johnson and move Dunn to first base should be shown Dunn's RZR. It makes this grown man cry.

Player              UZR   Change  UZR/150  Change  RZR   Change   MLBaverage
Zimmerman          13.3   + 3.7    19.4     +2.6  .753   +.014     .717
Johnson           - 6.4   - 1.8   - 7.8     -1.0  .760   +.008     .783
Guzman            - 4.0   - 0.1   - 6.2     +0.5  .767   +.010     .803
Dunn (LF)         -12.0   - 0.3   -23.1     +4.3  .849   +.036     .890
Hernandez           1.0   - 0.2     2.0     -0.6  .848   +.004     .816
Kearns (RF)         0.4   + 0.3   - 0.9     +2.3  .887   +.006     .907
Willingham (LF)   - 1.3   + 0.5   - 4.0     +1.3  .908   +.003     .890
Harris (CF)       - 5.0   - 0.3   -15.0     -0.6  .938   +.002     .933
Dukes (CF)        - 5.6   + 0.3   -21.1     +0.7  .898   n.c.      .933
Gonzalez (ss)     - 4.5   - 1.3   -22.5     -0.1  .689   +.007     .803
Belliard (2b)       1.5   + 2.6     6.8     -0.2  .889   +.020     .816
Willingham (RF)   - 1.3   - 1.5    -4.4     -5.9  .977   -.023     .907
Dunn (RF)         - 6.7   + 0.1   -33.9     -0.2  .756   n.c.      .907
Morgan (cf)         4.9   + 4.2    26.5    +22.0  .969   +.027     .933
Dukes (RF)        - 0.2    n.c.   - 3.4     +0.1  .952   n.c.      .907
Gonzalez (2b)     - 1.9   - 1.0   -14.5     -1.5  .794   +.021     .816
Harris (2b)       - 2.4    n/a    -27.6     n/a   .739   n.c.      .816
Dunn (1b)         - 3.6   - 0.5   -44.5     -1.7  .545   -.055     .783

The table is in order of innings played at the position, with the cut-off at 75 innings.

UZR converts fielding statistics into a +/- rating reflecting the number of runs saved or given away. There is a basic version, which is based on the number of innings played by the player, and a normalized version that assumes everybody plays 150 games. It also is forced to sum to zero, so a player can have his UZR adjusted even when he doesn't play.
RZR is like a fielding 'batting average', dividing the number of plays made by the number of opportunities. Unlike old Zone Rating, it excludes plays made 'out of zone'. (Zone Rating divides the field into zones, and assigns responsibility among the players.) For comparative purposes, I've put at the end of each player's line the MLB average RZR for position.

Nationals Fielding Weekly #4

There's no real pattern I can see to the figures this week. Using Ultimate Zone Rating, a method for converting fielding chances into a +/- runs figure, which is complex and is adjusted according to how other players perform as well as to the peculiarities of the park, the non-Zimmerman infield improved sharply. The same sort of applies to Revised Zone Rating, which gives an average like batting average. However, UZR suggests the outfield play deteriorated. As far as I can tell, which isn't very far, these adjustments relate to some of Harris's play, some of Dunn's play and, on the infield, some of Zimmerman's play.*

'Median' refers to the zone ratings at the midpoint of the National League RZR chart at that position. I've put those players who are fielding above this level in bold.

Player           UZR   Change     UZR/150   Change   RZR    Change    Median
Zimmerman (3b)   7.6    - 1.2      17.8    - 3.9    .757    -.018     .709-.701
Johnson (1b)    -6.4    + 0.5     -10.4    + 1.5    .729    +.011     .781-.776
Guzman (ss)     -3.8    - 0.9     - 7.4    + 6.3    .754    +.011     .807-.793
Hernandez (2b)   1.5    + 0.6       3.2    + 1.2    .846    -.010     .820-.806
Dunn (lf)       -9.5    - 3.8     -23.4    - 9.7    .812    +.006       .873
Kearns (rf)     -0.5    - 0.3     - 5.6    - 0.4    .877    +.002       .912
Willingham (lf) -2.2     n.c.     - 6.5    - 0.1    .902    +.002       .873
Dukes (cf)      -6.0     n.c.     -23.3    + 0.4    .898    +.001     .944-.942
Harris (cf)     -4.9    + 0.9     -15.4    + 8.4    .935    +.010     .944-.942
Dunn (rf)       -6.7    - 0.2     -35.1    - 1.4    .756     n.c.        .912
Belliard (2b)    1.4    - 0.2       9.0    - 0.8    .881    -.012     .820-.806    
Gonzalez (ss)   -2.0     n.c.     -15.1    + 0.4    .718     n.c.     .807-.793
Dukes (rf)       0.9    + 1.1     - 3.5    + 9.1    .952    +.005        .912
Willingham (rf)  0.4      n/a       5.7      n/a   1.000     n/a         .912
Dunn (1b)       -2.0    - 0.1     -37.4    - 0.6    .714     n.c.     .781-.776
Maxwell (cf)     1.9    + 0.1      36.9    + 1.7    .833     n.c.     .944-.942

__________
* As I've said before, I don't think UZR is really designed to be used in this way. It's more of an end-of-season thing. The best it can do in-season is give a guide as to who is the most valuable fielder on a team, and how he is doing relative to other fielders at his postion.

Nationals' Fielding Weekly #3

This week's chart is less exciting than last week's and just shows how Ultimate Zone Rating is rather opaque on a week-to-week basis. (I don't think it's designer, The Book author Mitchel Lichtman, intended it to be used the way I am doing.) Revised Zone Rating, meanwhile, shows that the good play in the outfield has continued, although the gains are not as dramatic.*

The story of the week is really the flubs on the infield. The only infielder to post a gain this week was Adam Dunn, who got a few innings at first base after Nick the Sick was hit by a pitch. I've already highlighted an error by Zimmerman in one of my Goats of the Day presentations as having played a significant part in a defeat. That, to me, has taken on symbolic status of how the significance of fielding remains a subject for debate. The average sabermetrically inclined person (i.e., me) can wonder whether the real fault lay in the inability of the Nationals to score runs up to that point? Or was Zimmerman's fielding muff the real culprit in a collapse in morale? Or was it Jesus Colome? Inquiring minds, blah, blah, blah.

Player           UZR   Change     UZR/150   Change   RZR    Change
Zimmerman (3b)   8.8    + 0.4      17.8    - 1.2    .757    -.010
Johnson (1b)    -6.9    - 1.0     -11.9    - 0.6    .729    -.004
Guzman (ss)     -2.9    - 2.0     -13.7    - 2.1    .743    -.004
Hernandez (2b)   0.9    - 0.7       2.1    - 2.2    .856    -.006
Dunn (lf)       -5.7    - 0.8     -13.7    + 1.2    .806    +.006
Kearns (rf)     -0.2    - 0.5     - 5.2    - 0.9    .875     n.c.
Willingham (lf) -2.2    - 2.1     - 6.4    - 7.2    .900    +.020
Dukes (cf)      -6.0     n.c.     -23.7    - 4.5    .897     n.c.
Harris (cf)     -5.8    + 0.1     -23.8    + 4.6    .925    +.013
Dunn (rf)       -6.5    - 0.2     -33.7     n.c.    .756     n.c.
Belliard (2b)    1.6    - 0.5       9.8    + 3.4    .893    -.011    
Gonzalez (ss)   -2.3    - 1.0     -19.1    - 7.2    .718    -.012
Dukes (rf)      -0.2    - 1.5     - 3.5    -14.1    .947    +.009
Dunn (1b)       -2.0    - 0.1     -36.8    +10.7    .714    +.114
Maxwell (cf)     1.8    - 0.1      36.9    + 0.5    .833     n.c.

________________
* I really ought to introduce a table showing an average RZR score at each position. I've got acres of space on the sidebars I'm not using.

Nationals' Fielding Weekly #2

Everybody asserted that the Nationals' broke their losing streak because they made more of the plays in the field they should have done. I'm not going to report that this is wrong; it's very right. Here's some numbers, showing Ultimate Zone Rating, Ultimate Zone Rating Adjusted to 150 Games (both from Fangraphs), and Revised Zone Rating (from The Hardball Times). I've also put in columns showing the changes from last week.

Player           UZR   Change     UZR/150   Change   RZR    Change
Zimmerman (3b)   8.4    + 0.4      19.0    - 0.9    .767    -.004
Johnson (1b)    -5.9    - 0.5     -11.3    + 0.9    .733    +.002
Guzman (ss)     -4.9    + 0.1     -11.6    + 1.0    .747    +.006
Hernandez (2b)   1.6    + 1.2       4.3    + 0.1    .862    -.006
Dunn (lf)       -4.9    - 0.3     -14.9    + 1.6    .800    +.031
Kearns (rf)      0.3    + 0.3     - 4.3    + 1.6    .875    -.004
Dukes (cf)      -6.0    - 1.6     -23.7    - 4.5    .897    -.014
Willingham (lf) -0.1    - 0.1       0.8    - 0.7    .880     n.c.
Harris (cf)     -5.9    + 1.3     -28.4    +14.5    .912    +.018
Dunn (rf)       -6.3      n.c.    -33.7    + 0.1    .756     n.c.
Belliard (2b)    2.1    + 0.1      13.2    + 0.2    .904     n.c.    
Gonzalez (ss)   -1.3    - 0.2     -11.9     n.c.    .730    +.008
Dukes (rf)       1.3    + 0.5      10.6    + 6.9    .938    +.038
Maxwell (cf)     1.9     n.c.      36.4    + 0.5    .833     n.c.
Dunn (1b)       -1.9     n.c.     -47.5    + 0.1    .600     n.c.
Harris (2b)     -1.7     n.c.     -36.8    + 0.1    .733     n.c.
Milledge (cf)   -0.5    - 0.1     - 9.4    - 1.0    .727     n.c.

Some notes.
1) UZR adjusts for other players and shows a +/- in terms of runs; RZR does not and works like batting average. That's why people who didn't actually play still get adjustments. It would help to know average RZR at each position, but instead you'll have to look at UZR to get an idea of how good they are relative to their peers.
2) In spite of all the noise in certain broadcasting quarters, the Nationals' problem with the leather was not on the infield, but the outfield. The top three RZR gains are all from outfielders. Result = 4 wins.
3) Dunn, Harris, Dukes, reading left to right, is probably the first-choice outfield at the moment. KEEP DUKES OUT OF CENTRE IF AT ALL POSSIBLE.
4) Gonzalez, given more playing time, hasn't risen to the occasion. I'd be inclined to abandon my 'groundball pitcher = Gonzalez at short' theory.

Nationals' Fielding Weekly #1

In two earlier posts, I outlined the sorry state of fielding by the Nationals. Some of the problems in holding a lead are related to the difficulty the fielders have in preventing hits and their propensity to make errors. Although media attention has focused on the infield, the real problems appear to lay in the outfield. The infield is hovering close enough to the league average to make it tolerable. (An exception could be made for Cristian Guzman, who is poor; anyone for shortstop practice?)

It might be a good idea, since this blog is turning gradually into a Nationals' blog, to look at this more systematically, rather than just when some mediot starts yapping. In that spirit, I'm going to start monitoring fielding on hopefully a weekly basis, in order to spot any trends. There are two measures at the moment that are easily available and make use of play-by-play data. Ultimate Zone Rating (UZR), published at Fangraphs, and Revised Zone Rating (RZR), which can be found at the Hardball Times. For this exercise, I'm going to make use of both UZR and UZR/150. (The latter is an extrapolation of existing data for 150 games.) UZR/150 is good for getting a sense of 'true talent'.

So, here's the ratings for all players who have played at least 40 innings on defence at a given position. They are ranked in order of innings at position.

Player          UZR       UZR/150      RZR
Zimmerman (3b)   8.0       19.9        .771
Johnson (1b)    -5.4      -12.2        .731
Guzman (ss)     -5.0      -12.6        .741
Hernandez (2b)   0.4        1.5        .856
Kearns (rf)      0.0      - 5.9        .879
Dunn (lf)       -4.6      -16.5        .769
Willingham (lf)  0.0        1.5        .880
Dukes (cf)      -4.4      -19.2        .911
Dunn (rf)       -6.3      -33.8        .756
Harris (cf)     -7.7      -42.9        .894
Belliard (2b)    2.0       13.0        .904
Gonzalez (ss)   -1.1      -11.9        .722
Dukes (rf)       0.8        3.7        .900
Maxwell (cf)     1.9       35.9        .833
Dunn (1b)       -1.9      -47.6        .600
Milledge (cf)   -0.4      - 8.4        .727
Harris (2b)     -1.7      -36.9        .733

Why 40 innings? Because I wanted to demonstrate that trading Johnson and moving Dunn to first-base is arguably not a good idea. (Dunn has played 44 innings at first.) Johnson is enough of a handicap at first. Why make it worse?

It's harder to make sense of RZR unless you know what the league average is at each position. Maxwell's relatively low RZR yet excellent UZR reflects his ability to get to balls out of his zone. The same also applies to Gonzalez at shortstop.

As far as fielding goes, this is the lineup I'd want to see based on current personnel in the system:

1b Johnson
2b Hernandez
3b Zimmerman
ss Gonzalez with a groundball pitcher, otherwise Guzman
lf Dunn
cf Maxwell
rf Dukes

As you can see, the odd-men-out here are Willingham and Kearns. If either can play some first, that would be a big help because it would then make more sense to move Johnson in a trade. Kearns is probably worthless on the market given his current problems hitting. Putting a bat like that at first isn't really worthwhile, either. He's a fifth outfielder as things stand. Willingham might have some value, but I don't know if it's all that much. Could he play first? His bat won't give away quite as much as Kearns's would.

Defensive Breakthroughs?

It really started here. But that was a blogpost about this. And it was soon followed by something we perhaps didn't know we needed.

While there are many fine new fielding metrics around using PBP (play-by-play data), to me the really important breakthrough is going to come when we get a reliable metric for fielding in the pre-PBP days. That's going to involve some kind of assumption about a piece of information we'll never have for most of baseball history - how many hits was a fielder responsible for.

I suspect it's quite important to start from the team level when calculating the value of fielding. (See Clay Davenport's article in the 2002 Baseball Prospectus for an example.) You then have a choice - either you can calculate the total of opportunities from the assumption that all balls will fall for hits, and thus the caught balls represent runs saved. Or you can look at fielding in terms of runs allowed, by thinking in terms of the number of mistakes. I suspect for a decent 'all of baseball history' metric, we need to think more along the lines of the latter. We need to establish 'expected runs' baseline against which to measure fielding.

I also came to a key realization after reading all this. Looking at the statistics when I was developing my Defensive Winning Percentage, I was surprised at how narrow the band is of fielding effectiveness. This is, I think, a reflection that there's both a floor and a ceiling to the effectiveness of fielders. Some of the credit is all down to the pitcher. All home runs except those inside-the-park, for example, are the pitcher's fault. Only some of the runs scored by those who walk, however, are the pitcher's fault. So there's some fielding input on that. But a walk and home run equals all the pitcher's fault.

Probably, the Fielding Experts will say I'm not being very original here. But I hadn't thought of it that way before.

More Field of Goats

I didn't realize, but UZR (a/k/a Ultimate Zone Rating), another method of rating fielding, is now available through Fangraphs. It's a daunting set of numbers, but the key one to look at is the last column, which measures the number of runs that a given fielder will 'save' over 150 games.

So let's run that ranking I did yesterday again. This ranked each player's fielding rating against those of qualified players in the NL. In most cases, that's about twelve players per position.

1b: Johnson, last
2b: Belliard, 4th; Hernandez, 9th.
3b: Zimmerman, 1st.
ss: Guzman, 11th.
lf: Willingham, 7th; Dunn, last.
cf: Dukes, last; Harris, last.
rf: Kearns, 7th; Dunn, last.

They more or less agree. Zimmerman comes out a lot higher in UZR, Willingham a bit lower. UZR does suggest that Dukes would make a good rightfielder, and Maxwell a GREAT centrefielder. If Mr Acta was channelling the Ghost of Bill McKechnie, he'd put Maxwell in centre, regardless of how he hit. In the circumstances, I think I would, too.

Field of Goats

I haven't decided whether to do a Goat of the Day for the loss today against the Phillies. I have, however, finally done something I've been meaning to do for a while.

Nationals fans don't need to be told that their fielding is poor. The real question is 'how poor?'. Your typical baseball fan looks at the players objectively, in a kind of philosophical way. Babe Ruth or Ted Williams is the 'ideal' major-league ballplayer. Someone like Willie Horton has certain ballplayer qualities, but is not so perfect. Then you have the Sandy Valdespinos of this world, who have been seen with major-league ballplayers, but few would mistake them for one.

However, in practical terms, a baseball 'manager' needs to look at a team relatively, not objectively. Relative to what? Why, to their peers, of course. If your team's players are universally in the lower regions of a skill set, you've got a problem that needs fixing. Compare all your problems, and the worst ones are the ones that most need attention.

Instead of trotting out some numbers, and pretending to be all scientific, I'm going to be wholly empirical. Let me take the RZR numbers from The Hardball Times' fielding statistics, and rank all Nationals' fielders who have played 96 innings against the National League qualifiers at each position. (Catchers don't get an RZR, so we'll ignore them.) Please note that at no position do we have sixteen qualifiers, but usually about twelve or so.

1b: Johnson, 9th.
2b: Hernandez, 8th; Belliard, 1st.
3b: Zimmerman, 5th.
ss: Guzman, last; Gonzalez, 12th (or one ahead of Guzman).
lf: Willingham, 4th; Dunn, last.
cf: Dukes, last; Harris, last.
rf: Kearns, 12th; Dunn, last.

So, any suggestions?

I have two. Use Belliard as a late-inning defensive replacement for Hernandez. He's doing quite well with the leather. Make Willingham the regular left-fielder. On the evidence so far he's a tolerable glove out there. Most everything else is a sinkhole, defensively, except for Zimmerman, and even he's not going to make anyone forget Brooks Robinson in a hurry. A centerfielder is probably the most urgent need, since none of the options tried so far (except possibly for Kearns, whose sample size is as yet too small) finish anything but last.