Will Eduardo Nunez Learn to Catch and Throw the Baseball?

A previous post covered how horrible a fielder Eduardo Nunez has been in terms of catching and throwing the ball. Today's post examines baseball history for stats that may help gauge the likelihood of Nunez developing into at least a minimally competent shortstop. If nothing else, it is a good excuse to see some names from baseball history and to have some fun with graphs and statistics.
 
A couple of ways to look at this question are to:

• Check whether history suggests that an extremely poor  rookie fielding percentage such as Nunez' corresponds to the length of the player's career at shortstop
• Look at individual precedents.

Note: This post does not suggest that fielding percentage equals defensive ability. It merely measures one facet of defense - the fundamental ability to catch and throw.

To check this out, I looked at the rookie seasons for all shortstops since World War 2 who fielded at least 125 chances, to see if there is some correspondence between their rookie fielding percentages and the number of career chances they had after their rookie season.

The league average shortstop fielding percentage has risen in this period from .956 in 1946 to a high of .974 (in 1997, 2005, and 2008). To adjust for these differences across time, this analysis uses Fielding Percentage Plus (FP+), which is described in Diamond Mind Baseball's Explanation of Statistics page.

Notes: Diamond Mind Baseball calls this statistic "Fielding Average Plus"; for some reason, Diamond Mind Baseball uses the term "Fielding Average" instead of the traditional "Fielding Percentage". Also, since the MLB definition of "rookie" has changed across time, the older definition of rookie - 75 ABs - is used.

I posted a spreadsheet containing this data to Google Docs - see https://docs.google.com/spreadsheet/ccc?key=0AkMu391-e5t0dHJrY2tjQlVPd2NkVW1oNWNfbnFmTUE#gid=0. As you can see, the only rookie shortstops since World War 2 worse at catching and throwing are Juan Beniquez and former Yankee Erick Almonte.

What I wanted to see is whether rookies with extremely poor rookie fielding percentages tended to have shorter careers at the shortstop position; the goal is to find some basis of comparison for Nunez. To do this, I looked at the relation of rookie season FP+ and total career post-rookie chances at shortstop.

Since we are looking at career totals, I removed all shortstops who either had a rookie season after 2006 or is still listed as active in the Lahman database. Shortstops who started out the last few years would not have yet had an opportunity to have a full career, so including them could have skewed the data.

Here are some stats on the remaining 307 shortstops:

• Their median rookie FP+ was 99.3 - meaning the typical rookie shortstop had a fielding percentage around 99.3% of league average. 
• The vast majority of the rookie shortstops had FP+ scores ranging from 97-103. In other words, most of the rookie shortstops were in a range of 3% worse to 3% better than league average. 
• 21 of the remaining 22 shortstops had FP+ below 97; fielding percentages more than 3% worse than league average.  The following histogram shows the negatively skewed distribution:

To give some context for post-rookie career chances, the median shortstop in this group had 852 post-rookie chances, and only around 25% of them had more than 3000 chances.

Now we can finally look at the important numbers. I divided the  shortstops into quartiles based on their rookie season FP+, the shortstops with the lowest 25% of FP+ by far had the least career chances:

Quartile #	FP+	Median Career Chances
1	98.3	444
2	99.3	1097
3	100.0	828
4	< 100.0	1418.5

The lowest quartile includes some shortstops with extremely low FP+ scores, with the lowest being 93.1. If we break these guys into 2 groups, the 10% of the shortstops with the worst FP+ scores have 217 median post-rookie chances, while the shortstops in the 11th through 25th percentiles have a median 695.5 career such chances.

You might ask, what about MLB shortstops with rookie seasons prior to America's involvement in World War 2? I did the same analysis, and the results were similar. Throughout major league history, shortstops with extremely bad rookie fielding percentages are likely to have much shorter careers at the shortstop position.

That being said, a small subset of such shortstops do go on to field a significant number of chances at shortstop. The following table lists the shortstops in this study who had horrific rookie fielding percentages - this is the group in which Nunez would fall:

Shortstop	Rookie Year	Rookie Age	Rookie Chances	Rookie FP%	Rookie SS MLB FP%	Rookie FP+	Post-Rookie Career Chances	Career FP+
Beniquez, Juan	1972	22	140	0.900	0.967	93.1	1	103.6
Almonte, Erick	2003	25	128	0.906	0.972	93.2	0	(NULL)
Mendez, Donaldo	2001	23	149	0.919	0.971	94.6	102	97.9
Richard, Lee	1971	22	323	0.920	0.967	95.1	79	97.4
Batiste, Kim	1992	24	167	0.922	0.969	95.1	206	98.3
Beltre, Esteban	1992	24	157	0.924	0.969	95.4	284	99.6
Olmedo, Ray	2003	22	195	0.928	0.972	95.5	114	101.0
Ramirez, Milt	1970	20	168	0.923	0.966	95.5	45	94.5
Templeton, Garry	1976	20	307	0.922	0.965	95.5	9511	99.6
Gonzalez, Julio	1977	24	252	0.921	0.963	95.6	217	98.9
Kunkel, Jeff	1984	22	218	0.922	0.964	95.6	543	97.7
Belloir, Rob	1975	26	153	0.922	0.961	95.9	32	97.3
Cedeno, Andujar	1991	21	257	0.930	0.969	96.0	2137	98.5
Sutherland, Gary	1967	22	207	0.928	0.963	96.4	262	97.7
Hernandez, Jackie	1968	27	341	0.927	0.962	96.4	1864	98.2
Spencer, Daryl	1953	24	234	0.927	0.962	96.4	2365	99.4
Nieves, Jose	1999	24	245	0.935	0.969	96.5	117	99.5
Pena, Roberto	1965	28	242	0.930	0.963	96.6	1527	100.3
Lillis, Bob	1961	31	224	0.924	0.957	96.6	1893	100.1
Berry, Neil	1948	26	201	0.930	0.962	96.7	527	99.6
Mantilla, Felix	1957	22	159	0.931	0.961	96.9	468	98.9
Meoli, Rudy	1973	22	401	0.933	0.962	97.0	174	100.9
Smith, Billy	1975	21	191	0.932	0.961	97.0	95	95.2
Jackson, Damian	1999	25	419	0.940	0.969	97.0	606	96.8
DeMars, Billy	1950	24	253	0.933	0.961	97.1	5	104.2
LeMaster, Johnnie	1976	22	174	0.937	0.965	97.1	4270	99.7
Snyder, Cory	1986	23	127	0.937	0.965	97.1	134	95.7
O'Brien, Eddie	1953	22	352	0.935	0.962	97.2	99	100.9
Lopez, Luis	1994	23	186	0.941	0.968	97.2	713	99.3
Clark, Ron	1968	25	173	0.936	0.962	97.3	148	100.1
Caruso, Mike	1998	21	629	0.944	0.970	97.3	569	98.9

As you can see, the bulk of these guys had insignificant careers at shortstop. However, there are a few exceptions. Garry Templeton clearly is the most significant; however, since Templeton was only 20 years old as a rookie, he's not an apt precedent for Nunez.

And then some of these guys went on to have careers at other positions such as second base and outfield - for example, Felix Mantilla, Gary Sutherland, and Juan Beniquez.

In this latter respect, it's worth mentioning a player who is active and thus not in the above table, Oriole second baseman Brian Roberts. Roberts' minor league stats indicate he was groomed by the Orioles as a shortstop. However, after his 2001 rookie season, in which Roberts FP+ at shortstop was 96.7, the Orioles apparently sent him to the minors to learn second base.

Let's circle back to Nunez. The historical precedents obviously suggest the odds are against him evolving into a shortstop with competent hands and arms. Of course statistics are just a generalization; it's not impossible that he will learn to concentrate better, or to have better mechanics in the field. The Yankees must feel it is worth a shot, since when they demoted him to the minors they announced he would play shortstop and not switch to a less demanding position such as second base. But the precedents do suggest that Nunez would be a very rare bird if he ends up as a regular major league shortstop.

Query To Select Rookies from the Lahman And Baseball-DataBank Databases

In working on the followup to the how bad a fielder Eduardo Nunez has been, I needed to query the Lahman database for players' rookie seasons. In googling, I noticed that some folks had posted questions regarding how to do this in the Lahman or Baseball-Databank databases; but I could not find a good response. So to help anyone else who might search for this, here I am posting the details of how to form such a query using MySQL.

Evidence that Brian McNamee Made Up Roger Clemens Steroid Accusation

Back in 2008, after reading a lot of the documents related to Brian McNamee's claim that Roger Clemenns took steroids, I decided that it was not reasonable to believe McNamee. The whole case against Clemens really amounts to what has come out of Brian McNamee's mouth; in my reading, I found far too many contradictions, etc., to take McNamee seriously. Here is a case in point.

In pp.15-16 of Brian McNamee's 2/7/2008 deposition to Congress, McNamee declared he had injected Congress in a storage closet in Tampa Bay, after a day game on a getaway day. On pp. 127-128 of the deposition, after questioning McNamee says this occurred in a June 15-17 series Jays-Devil Rays series.

McNamee provides a lot of details in his story. Some of the details cannot be fact-checked, some can. In this case one can check Retrosheet to see whether a getaway day game occurred when McNamee claimed. According to the 1998 Blue Jays' game log, the June 17 getaway game was a night game, not a day game.

A-Rod's HR Stats

A-Rod's 2012 HR performance to date has begun to generate the usual ignorant bile that, unfortunately, characterizes the bulk of the NY baseball media.  Not all of the media reaction has been as hysterical and nasty as Andrew Marchand's nonsense, where he actually advocates for putting Raul Ibanez ahead of A-Rod in the batting order. I didn't know much about Marchand; but this article taught me to never again waste valuable time reading anything with a byline saying "Andrew Marchand".

Now to be fair, whether A-Rod's power is in a serious decline is a legitimate question - he is now 36, and coming off a season in which his power was curtailed by a serious knee injury. However, one cannot take seriously all the pundits who regurgitate A-Rod current stats without context. For example, when Marchand mentions the following:

   "But Ibanez's nine homers and 27 RBIs are almost double A-Rod's production (five homers, 15 RBIs), in 38 fewer at-bats."

Yes, those are their stats at the moment. But what do those numbers mean? Specifically, what does A-Rod having only 5 HRs signify in terms of gauging what to expect from A-Rod going forward?

One way of looking at this is to see if A-Rod's current stats - 5 HRs in 167 Plate Appearances (PAs) - is unprecedented or aberrational for him. I've a program that answers this type of question, and what the program shows is that in A-Rod's career through 2010 there have been 256 times where he has had streaks of 167 PAs in which he had 5 or less HRs. This is out of 7508 appearances in which he was in the midst of consecutive appearances of 167 PAs within a season; in other words, A-Rod has experienced such HR slumps approximately 3.4% of the time.

(Note: I excluded 2011 because we essentially are trying to determine whether or not A-Rod's  performance suggests he will hit with the power he demonstrated prior to 2011.)

These HR slumps occurred in 1996, 1997, 1998, 2004, 2009, and 2010. In only one of those seasons - 1997, at age 21 - did A-Rod hit less than 30 HRs; and then after 1997 he hit 30+ HRs for 13 straight seasons. The point being that his stats to date are not aberrations for him - Joe Girardi should not have to endure idiotic questions over dropping A-Rod in the batting order because of these 167 PAs..

Since we have only a 1/4-season's worth of 2012 data, another perspective to consider is that A-Rod's performance is just due to a small sample size. Statistics provides binomial distribution formulas that can estimate this.  Let's say we're trying to gauge whether A-Rod still has the power he had in 2009-2010, in which he hit HRs in 6.2% of his ABs (60 HRs in 966 ABs).  A-Rod to date has 144 ABs. Using this method, and assuming we are looking for numbers that provide a 95% degree of confidence, in 144 ABs we would expect a hitter with a true talent level of hitting a HR 6.2% of the time to be in the range of 4-14 HRs.

Since A-Rod has 5 HRs, he's well within that range, which means any wailing and gnashing of teeth over his current stats is ridiculous.

Note: For reference, I calculated this using the following R formula: qbinom(c(.95, .05), 144, .062)

Curt Schilling Bounces $1.1 Million Check, Takes Corporate Welfare

Curt Schilling's video game company venture has devolved into a mini-Enron of sorts, and there are several aspects of this that are comical. Though I should warn you that if you are a Rhode Island taxpayer, or one of Schilling's employees who are being stiffed because his company cannot meet payroll, you may not appreciate the ironies:

• Schilling's company was lured to Rhode Island by $75 million in taxpayer-guaranteed loans. One might joke that whomever was governing Rhode Island at the time was smoking crack; however, this is unfair to crackheads, who would never guarantee so much money to a novice company venturing into the risky video game market.
• Schilling's company, "38 Studios", bounced a $1.1 million check to the state of Rhode Island. The details are funny: apparently 38 Studios first was going to wire the money to the state. It then hand-delivered the check. But after the check was delivered, the company's CFO then told the state the check would bounce.
• A spokeswoman for Rhode Island Governor Lincoln Chaffee provided a classic example of understatement: "Once the company notified the state that they were going to wire the money, that seemed to be a positive sign. I don’t think I’m going out on a limb too far to say that presenting a check without sufficient funds does not have a positive feel."
• Schilling has been known to stump for GOP politicians such as George W. Bush and John McCain. Having the government subsidize a high-risk video game start up company run by a celebrity is the polar opposite of boilerplate Republican rhetoric about fiscal responsibility. But of course regardless of one's political orientations, what Bush and McCain did together - cutting taxes while simultaneously initiating a trillion dollar war in Iraq, etc. - was very irresponsible from a fiscal perspective. What I think this means is that being a member of the Corporate Dole makes Schilling a perfect GOP presidential candidate in 2016.

CASHIERS: Do not accept checks from either of these guys.

(Photo source: http://i.a.cnn.net/si/2004/writers/phil_taylor/11/01/schilling.bush/p1_schilling_ap.jpg)

How Bad a Fielder Has Eduardo Nunez Been?

Eduardo Nunez is without doubt the most frightening Yankee fielder since the days when Chuck Knoblauch was hitting fans in the stands. In his brief career, Nunez has made 23 errors in about 2/3 of a season's worth of defensive innings at 2B, SS, and 3B. He has been brutal.

Before proceeding, I should acknowledge that of course errors and fielding percentage are hardly the most important stats to measure defense. However, a basic ability to catch and throw the baseball is necessary - it doesn't matter how far a player ranges if he handles the baseball the way that Ozzie Guillen handles public relations.

Also, let's keep in mind that there's a lot to like about Nunez: he has a live bat for an infielder, and is an excellent base-stealer. Clearly the Yankees like his potential, and Nunez will be back in the majors if he can just learn to catch and throw the @$%&*#@ ball. Do his fielding statistics shed any light on whether he can improve?

Today let's get some perspective on  Nunez' defensive performance by looking at last year's rookie season. In 2011, Nunez had 161 chances as a shortstop and made 14 errors, for a .913 fielding percentage (FP). How bad was this? The average shortstop in 2011 had a .972 FP. In other words, the average shortstop would make 4.5 errors in 161 chances, 9.5 errors less than Nunez made.

He was pretty lousy at third base too, and has been lousy in 2012 too; Nunez inability to catch and throw unfortunately has not been confined to one position or one season. I am focusing on shortstop only because that is the position he played the most in his rookie season, which thus allows for some comparisons with other shortstops' seasons.

How bad has Nunez been? Since 1986 major league shortstops have averaged about a .911 fielding percentage, with annual fielding percentages ranging from .965 to .974. For this period,  the following table lists the shortstops with the 10 worst fielding percentages in seasons where the shortstop had at least 150 chances. Nunez' 2011 season tops out the list:

Player	Season	Age	C	E	FP
Nunez, Eduardo	2011	24	161	14	0.913
Stillwell, Kurt	1987	22	209	18	0.914
Polcovich, Kevin	1998	28	238	20	0.916
Batiste, Kim	1992	24	167	13	0.922
Beltre, Esteban	1992	24	157	12	0.924
Gutierrez, Ricky	1994	24	293	22	0.925
Blauser, Jeff	1996	30	312	23	0.926
Olmedo, Ray	2003	22	195	14	0.928
Lopez, Felipe	2003	23	208	15	0.928
Cedeno, Andujar	1991	21	257	18	0.930

You'll note this list consists of players who had about 1/4 to 1/2 a season's worth of chances. The 10 worst fielding percentages of shortstops who had at least 600 chances in a season ranged from .935 to .955 - a .935 fielding percentage seems to be about the lowest a major league team will put up with for an entire season (think Jose Offerman and Jose Valentin). Regarding the players in the above table,we do have to keep the relatively small sample size in mind - it's likely that some of them would have been able to catch and throw at that minimal level if given more chances that year, while others were just horrifically bad shortstops.

This list does not exactly provide a promising set of precedents for Nunez. For example, Felipe Lopez did have several seasons as a shortstop after his 2003, and also in 2002 was able to catch the ball, showing that having a horrific 1/3 of a season can just be a small sample size issue. But overall the Yankees no doubt hope for Nunez to have a better career than these guys had.

I found this table interesting for 2 reasons:

• Historical perspective - it shows how bad Nunez has been.
• Catharsis - for someone who has anguished every time a ball is hit in Nunez' direction, it made me laugh to discover that his performance has been the worst in decades. 

That being said, I would not make too much of this table alone in terms of gauging whether Nunez will ever develop defensive competence. There are some other ways of using historical fielding stats to get some context on this question, which a followup blog post later this week will discuss.

Imagine There's No Depth

Mariano Rivera's torn ACL was to me the worst loss I have ever experienced as a Yankees fan; especially in the hours when it was unclear whether last Monday's win would be the last time we saw Mariano. His accomplishments, his grace and athleticism, his calm and confident presence, and his contribution to the local economy of wherever Louisville Slugger makes bats - well, there is no replacement in my heart for Mariano Rivera.

However, Mariano likely being out until 2013 does not destroy the 2012 Yankees season. It does harm the Yankees chances, no doubt. Even if it is as few as the 2 wins Tom Tango estimates Mariano is worth, 2 wins could be the difference between playing in October or not.

But losing #42 should not destroy the Yankees' chances, because the team still has a strong late-inning bullpen. This is due to both David Robertson's emergence as a top reliever and to last year's controversial signing of Rafael Soriano.

There was much wailing and gnashing of teeth over Soriano's signing, since it cost the Yankees a first-round draft pick to obtain a reliever for the 7th-8th inning.

There also was the financial cost of the contract; the Yankees had to pay closer money - 3 years at $35 million - to lure Soriano into a setup role. That is just tipping money for the revenue-soaked Steinbrenners, so only immense greed would have made finances an excuse from not improving the depth of the bullpen.

Imagine however if the Yankees were operating in the mode that was leaked to the press this winter, which is that by 2014 the Yankees will be under the salary cap because of incentives in the new Collective Bargaining Agreement. In that mode, the 8th-inning guy now would not be Rafael Soriano; it would be some combination of Boone Logan and Corey Wade, or some mediocre and cheap reliever pulled off the scrap heap. Ugh.