Where should they be playing?

What follows is (mostly) just for fun. Don't take it too seriously.

One of the themes on this blog over the past year or so has been to stress how important fielding performance is to not only how we evaluate players, but also to how well a team can be expected to perform. The '07 Reds are often said to have several players playing out of position (certainly Dunn, but possibly Griffey and Encarnacion, and maybe Phillips, depending on who you're talking to). I thought it might be fun to use data from the Fans' Scouting Report to try to quantify where each player on the Reds' team would be expected to best perform, and whether there was a better defensive alignment than the one the Reds most often used in 2007.

Some quick background on the methods: the Fans' Scouting Report is based on survey data of (ideally) a large number of "hardcore" fans who see lots of games each year. Each participant is asked to rate players that they saw more than 10 times that season on seven fielding skills, encompassing everything from first-step quickness and maximum sprint speed to throwing arm strength and accuracy. Participants are asked to rate players relative to all other players, not just players at their position. Once the averages are in from those rankings, one can estimate a players' rating at a specific position using a specific set of custom skill weightings for that position.

Below, I've estimated 2007 Reds' fielding ratings at each of the eight positions. Values are on a scale of 0-100, though few players ever top 80. Bold-faced numbers indicate that a player's weighted skill average is above average at a position, relative to MLB norms in '07 (Notes: players had to "qualify" for a position to be bold-faced, so left-handed throwers were not eligible to play C, 2B, 3B, and SS. I also assumed that anyone who could play catcher was already playing catcher, so only David Ross and Javier Valentin were eligible for getting bolded at that position).

Player	C	1B	2B	3B	SS	LF	CF	RF
A Dunn	24.4	16.5	18.7	25.3	20.7	21.9	21.9	23.1
A Gonzalez	67.0	61.3	64.1	63.3	64.0	59.8	59.1	60.3
B Phillips	78.4	75.9	80.5	76.3	79.1	79.8	78.7	79.6
D Ross	62.8	49.2	51.3	60.3	53.6	46.5	46.5	48.6
E Encarnacion	46.9	52.0	55.2	50.7	52.1	56.8	56.6	55.9
J Valentin	26.7	24.2	25.5	26.0	24.0	24.6	24.4	25.0
J Keppinger	45.8	43.2	45.7	42.8	43.8	44.2	43.4	44.3
J Hamilton	72.8	63.0	67.0	72.6	67.7	67.4	66.8	68.9
J Castro	60.0	55.4	58.2	55.8	56.3	55.0	54.1	55.2
K Griffey Jr.	63.9	55.8	57.6	56.5	55.7	48.6	48.2	49.2
N Hopper	49.0	51.9	56.5	48.6	53.3	61.2	59.8	60.0
R Freel	46.1	51.0	55.8	47.7	51.6	62.7	61.2	61.3
S Hatteberg	39.5	42.9	44.9	36.1	40.1	42.0	41.1	41.0
League Average	45.8	41.3	57.4	52.5	58.7	46.4	63.0	50.9

As you'd expect, there are fewer above-average candidates for the positions near the top of the defensive spectrum--catcher, center field, and shortstop--than there are for those near the bottom, like first base and left field.

Let's take a quick tour of the Reds' 2007 Starting Lineup:

• Dave Ross - Ross is rated as an excellent defensive catcher, and that's clearly the only place his meager hitting could even be slightly tolerated. And even that's questionable.
  
• Scott Hatteberg - Hatteberg used to be a catcher before blowing out his arm. Now, his arm skills are rated below his other four skills...and overall, he's only rated above average at first base. He's a fine hitter, but his lack of defensive ability really limits his value.
• Brandon Phillips - Phillips qualifies as an above average fielder at every single position, with the possible exception of catcher (I'm assuming that Fans' ratings of catchers relative to other positions is incomplete). His single best rating, however, was at second base. ... though he'd likely be more valuable at a more challenging position, with little apparent drop-off in skill.
  
• Alex Gonzalez - Gonzalez is interesting in that, while not as good at Phillips at any position, his skill set seems a bit more tuned to shortstop than Phillips'--mostly due to his high arm rankings relative to his other skills.
• Edwin Encarnacion - Here's one that's interesting. Edwin is rated as a tad below-average at third base, mostly due to his poor ratings on release and throwing accuracy. His good ratings for speed and acceleration, however, indicate that he could potentially become a good left fielder, perhaps upwards of +7 runs above average. That would more than make up for the -3 run value penalty associated with moving from 3B to LF.
• Adam Dunn - Yikes. Dunner is rated below average at every position. His highest absolute score was at third base, but there he'd be 27 points (~19 runs) below average. The smallest difference between his rating at a position and MLB averages at that position was... left field, at -24.5 points! I was surprised he didn't do better at first base (-24.8 points), which has a lower threshold, but his terrible ratings for Hands, Instincts, and Acceleration absolutely killed his rating at that position. Maybe moving him to first base isn't the answer...?
• Josh Hamilton - The Fans loved Josh, and if he threw with his right hand, he's be just like Phillips--above average anywhere you put him. As it is, even though he's just slightly above average in center field according to these numbers, he's most valuable there...though his outstanding throwing arm makes him a more naturally superb right-fielder.
• Ken Griffey - Junior's poor speed scores make him a poorly suited for the outfield, but the fact that he throws with his left hand rules out most infield positions. Because of the weak fielders at first base and left field, however, he rates as above-average in those spots.
  

Of course, just because a player is best suited for a particular position doesn't mean that's where the Reds should play him. They have to determine where each player can be placed (or hidden) to maximize the total fielding of the team. Assuming that the above eight players were going to play, I tried every combination I could think of to maximize the team fielding score. Here's my best alignment, along with the expected runs saved vs. average per season according to the Fans' Scouting Report data:

Pos	Name	Runs
C	D Ross	11.9
1B	S Hatteberg	1.1
2B	B Phillips	16.1
3B	E Encarnacion	-1.3
SS	A Gonzalez	3.7
LF	A Dunn	-17.1
CF	J Hamilton	2.7
RF	K Griffey Jr.	-1.2

Huh. Looks familiar, eh? I did try, but I couldn't find a better alignment than what the Reds used last season. The biggest roadblock is Hatteberg, who prevents me from moving Griffey, Dunn, or Encarnacion to first base.

Ok, just for fun, here's a lineup of '07 Reds that maximizes defensive skill and ignores offensive value altogether. Think of this as the ultimate run-prevention lineup for the '07 Reds. Runs values are per-season numbers, so this is what Fans say we'd get from Castro defensively if we played him a full year at third base:

Pos	Name	Runs
C	David Ross	11.9
1B	K Griffey Jr.	10.1
2B	B Phillips	16.1
3B	J Castro	2.3
SS	A Gonzalez	3.7
LF	N Hopper	10.3
CF	R Freel	-1.3
RF	J Hamilton	12.6

Fans data indicates that this lineup would save 50 more runs per season than the standard '07 lineup! That's primarily done by benching Dunn and Hatteberg, moving Griffey to first base, and starting Hopper and Freel in the outfield (there was also a very small boost to starting Castro and third instead of Encarnacion, but we'll ignore that because that's clearly not a worthwhile exchange).

Ok, so would it be worth it to use that lineup instead of the one the Reds actually used? Well, if we assume a full season is 700 PA's, and then pro-rate '07 offensive production using runs above average per plate appearance (RAA/PA, which, I was recently told, is the best-behaving rate stat for offense) for these four players, here's how a season of Dunn and Hatteberg compares to a season of Freel and Hopper on offense:

Name	PA	RAA	RAA/PA	pr-RAA
Dunn	632	31.0	0.049	34.3
Hatteberg	417	13.7	0.033	23.0
Hopper	335	-0.4	-0.001	-0.8
Freel	304	-11.3	-0.037	-26.0

Ok, so based on this (very crude) estimate, the offensive difference between a full season of Hatteberg and Dunn vs. Hopper and Freel is roughly 84 runs! That difference outweighs the fielding bonus we estimated using the Fans' data by something on the order of 3.5 wins. Even if we just substitute Hopper in RF for Hatteberg at 1B (moving Griffey to 1B), the offensive dropoff (-24 runs) isn't enough to pay for the defensive bonus (+18 runs). So it looks like the Reds were right to play the guys they played, where they played 'em. :)

...

Now, admittedly, there are about a billion holes where I could tear this little study to pieces. But it was a fun exercise, and I think we'll be able to do this sort of thing with much greater accuracy over the coming years as our ability to quantify and assess both fielding skill and fielding performance continues to improve. If nothing else, though, these data indicate that there's not an obvious way in which the Reds could have gotten more value out of their players last year, which has to be considered the null expectation in any study of this sort. I'd be more suspicious of this study if it showed a clear inefficiency!! :)

Fixing the Reds' bullpen

John Fay has been predicting that the Reds will target relievers in the free agent market this offseason in an effort to build up the ballclub's pitching, mostly because the market for starters is so dismal. The Reds most certainly need better run prevention, and their bullpen has been terrible since '03 or so. So it makes sense to target relievers, right?

Well, perhaps not. Tom Tango posted this about an article on the San Diego Padres' Kevin Towers by Tom Verducci at SI.com:

Looks like Kevin Towers and the Padres know what they are doing. Free-agent relievers are incredibly overpaid. You can pick out the teams that have no idea that the replacement level for pitchers as relievers is so high. Orioles last year, and Phillies this year. I’m always shocked when starters are released, and not given a try as a reliever. If they were fighting for a job as the 5th starter (and basically, the 7th best pitcher on the team), doesn’t it make sense that they could fight for the #3 or #4 guy in the bullpen?

I'll have more on valuation of relievers and starters in my next article on player value (sorry for the delay--been distracted with parents visiting, etc). But this argument here is not new. Think about all the ex-starters (and often failed starters) who have found their niche as relievers. Just naming those who have been associated with the Reds in recent years, we have David Weathers, Kent Mercker, Mike Remlinger, Chris Hammond, Josh Hancock, Ryan Dempster, Ryan Franklin... The list of those successfully going the other way is much more limited (Derek Lowe, .... uhhh... wait, I'm sure there's more...Danny Graves? No, that didn't work... hmm...).

Relieving is easier than starting. There are a variety of potential reasons for this, and I'm not sure that we really know the underlying causes. But there have been a good number of studies verifying that the little anecdote-based argument about this actually does correspond to the real differences in value.

Therefore, mining the failed starter ranks--especially those from the superior American League--might be a more efficient way to build up the Reds' bullpen than shelling out 3-4 million dollars at a time on established relievers. Might not get you a Cordero-esque closer, but it seems to be a good way to get some quality middle relief on the cheap.

Nuxhall in Hospital

Joe Nuxhall is in the hospital and is undergoing treatment for pneumonia. Early indications are that he is doing well, but I wanted to send out a quick get-well wish...'cause you know, I'm sure he's surfing around reading the various Reds blogs tonight. Hope he makes a quick recovery.

While we're at it, give Nuxhall a Hall of Fame vote. He sure as heck deserves it.

Update: Apparently, while the pneumonia seems to be under control, Nuxhall is due to get a pacemaker due to low pulse rate. I'm honestly not sure how substantial of a surgery a pacemaker installation is. But at the very least, it sounds like he's getting excellent care.

PMR on the Reds

David Pinto continues to release his PMR data at a steady clip. I consider PMR superior to RZR because of the additional adjustments it makes for batter handedness and park factors, so I will be updating my composite fielding estimates and total value estimates for Cincinnati Reds based on his data once he's run through the seven non-catcher/pitcher positions.

For now, though, it's nice to see some Reds getting love from Pinto's system:

• Brandon Phillips - +46 outs, or ~+35 runs. Wow.
  
• Josh Hamilton - +4 outs (+3.6 runs) in CF. This will push his overall fielding estimate in CF above average in 2007. The Fans also rate him as just above average, so I believe it. Maybe he'll be ok to have out there, at least for the next few years.
  
• Norris Hopper - +1 outs in CF.
• Alex Gonzalez - +0 outs at SS. That's a disappointment, though he'll still be above-average based on the other stats.
  
• Jeff Keppinger - -6 outs at SS (-4.5 runs).
  

Update: Edwin Encarnacion was rated just -7 plays below average (-5.6 runs) at 3B according to PMR, which is much better than his -21.9 RZR rating this season, and much more consistent with his Fans' and ZR ratings. Why? I'm not sure, as PMR and RZR use the same hit location data; maybe there were an unusually large number of lot of hard-hit balls through his position this year? Replacing Eddie's PMR value with his RZR value will give Eddie's total value a boost of ~5 runs or so, which will shoot him into Scott Hatteberg territory. Very nice to see some more numbers supporting the conventional wisdom that he improved significantly this season with his glove.

Also interesting are his estimates of fielding behind pitchers, which can be considered a more precise way of assessing BABIP (or DER, depending on your preference) because it considers hit location, batted ball type, etc:

• Aaron Harang, +5 outs
• Matt Belisle, -0.5 outs
• Kyle Lohse (when with Reds), -3.7 outs
• Bronson Arroyo, -5.6 outs

If we convert those out values to an approximate runs stat (I'll use 0.8 runs/missed out), we can then subtract those values from actual runs allowed and get an approximate fielding-independent R/9 (listed below as PMR-R/9). I'll list that along with FIP, another way of trying to normalize pitcher performance:

Name	ERA	FIP	R/9	PMR-R/9
Harang	3.73	3.70	3.88	4.04
Belisle	5.32	4.54	5.62	5.60
Lohse	4.58	4.50	5.19	4.99
Arroyo	4.23	4.51	4.66	4.47

FIP is an estimate of ERA based strictly on readily available pitcher peripherals -- k/9, bb/9, and hr/9. Differences between FIP and ERA, in general, are due to poor fielding, "clumping" of offensive events in a manner that deviates from the norm, or non-hr park effects. Differences between R/9 and PMR-R/9, on the other hand, can probably be considered to be strictly due to differences in fielding.

Generally spreaking, the differences between FIP and ERA are much larger than differences between R/9 and PMR-R/9, at least among these four pitchers. In fact, differences between the two pairs of stats only move in the same direction twice! This indicates that the timing of offensive events (which is largely the result of luck) may have more to do with differences between actual and DIPS-based ERA estimates than the Reds' below-average fielding. That jives with Tom Tango et al.'s breakdown of variation in BABIP:

luck: 44%
pitcher: 28%
fielding: 17%
park: 11%

Neat stuff.

Total player value around MLB

Skyking162 is doing a wonderful position review series at his site, and he's ranking players following a very similar methodology to that I've discussed in the player value series. In fact, our e-mail conversations last month were a big part of the reason that I decided to do my series--he really helped me take my thinking to the next level. If you see differences in our rankings, it's because a) he uses a slightly different base runs model than I do to generate linear weights, and b) his fielding estimates are based on ZR and RZR only, whereas I threw in FSR as well. As far as Cincinnati Reds are concerned, the biggest discrepancy between his and my datasets is Brandon Phillips, for whatever reason.

Here are links to the positions he's covered thus far: First Base, Second Base, Left Field, and Center Field.

His blog has really become a great read for general baseball analysis and opinion since his return a few months back.

Player Value, Part 4: Position Player Wrap-Up

To view the complete player value series, click on the player value label on any of these posts.

Let's go over the main principles we've covered thus far:

1. A run saved is worth the same as a run scored. Therefore, estimates of player value require one to consider both offense and defense.

2. We often find it useful to compare players to replacement players, who we assume will hit at 73% of league average and play league-average defense at a given position.

3. A player's total offensive contributions can be estimated using linear weights, relative to replacement player offensive production levels.

4. A player's fielding performance can be best estimated using the combined estimates of several different statistics. Because replacement players can be assumed to be league-average fielders, we simply report all fielding statistics as +-average at a position.

5. We also need to add a position adjustment to a player's fielding estimate to signify the differences in value (i.e. average difficulty) of playing one position vs. another. This should be pro-rated for the player's playing time (defensive innings).

Now, let's put it all together and look at a case study.

The 2007 Reds Position Players

Fielding

While I've already posted offensive value estimates for the '07 Reds, I haven't yet posted fielding data, so let's start there.

Below I've used combined data from the Fans' Scouting Report, ZR, RZR, and the catcher statistics (C-Runs below) to estimate fielding runs saved vs. average. Methods are as described in my previous posts. ZR and RZR data are the summed values across all positions a player played. FSR data are calculated using Tango's custom weights for each position a player plays, pro-rated for his defensive innings at each position (Player innings out of 1440), and summed. I've also added a position adjustment (using Tom Tango's numbers) to the data to get an estimate of total fielding value.

Here are the data (sorry about the missing cells--nature of the beast). Fielding is the +-Fielding value relative to a player's position. TtlFldVal is a player's total fielding value, which is the sum of Fielding and the position adjustment (PosAdj). All values are reported as +-runs.

Name	PriPos	Inn	FSR	ZR	RZR	C-Runs	Fielding	PosAdj	TtlFldVal
D Ross	C	837	6.9			6.3	6.5	5.8	12.3
B Phillips	2B	1371	15.3	-0.8	15.1		9.2	0.0	9.2
N Hopper	CF	607	0.6	4.3	11.2		6.0	0.9	6.8
A Gonzalez	SS	872	2.2	3.2	5.4		3.8	2.4	6.2
R Freel	CF	577	-0.7	-4.7	7.8		1.0	1.5	2.5
J Ellison	RF	82		1.8	3.5		2.1	-0.2	1.8
D Wise	CF	13		0.4	0.6		0.5	0.0	0.6
M Bellhorn	3B	14		0.2	-0.1		0.0	0.0	0.0
R Hanigan	C	20				-0.1	-0.1	0.1	0.0
J Keppinger	SS	509	-3.4	2.9	-3.2		-1.0	1.0	0.0
R Jorgensen	C	34				-0.2	-0.2	0.2	0.0
J Valentin	C	472	-4.4			-3.1	-3.4	3.3	-0.1
J Hamilton	CF	663	2.0	2.3	-8.4		-1.8	1.6	-0.2
C Moeller	C	87				-0.9	-0.9	0.6	-0.3
B Coats	RF	68		0.0	-0.5		-0.3	-0.1	-0.4
P Lopez	SS	93		-2.6	1.2		-0.7	0.3	-0.4
E Cruz	SS	2		-0.7	-0.7		-0.7	0.0	-0.7
J Castro	SS	178	0.0	1.0	-4.9		-1.5	0.2	-1.3
J Cantu	1B	112		-1.7	0.1		-0.8	-0.6	-1.4
J Conine	1B	435	-1.9	1.4	-2.8		-1.0	-2.4	-3.4
J Votto	1B	188		-0.7	-5.7		-3.2	-0.9	-4.1
S Hatteberg	1B	772	0.6	7.4	-11.1		-1.2	-4.3	-5.5
E_Encarnacion	3B	1168	-1.1	3.0	-21.9		-7.4	-0.8	-8.2
K Griffey	RF	1163	-1.0	-1.2	-17.1		-7.1	-3.2	-10.4
A Dunn	LF	1189	-14.2	-4.1	-17.5		-11.6	-3.3	-14.9

According to these data, the Reds' most valuable fielders in 2007 were David Ross, Brandon Phillips, Norris Hopper, and Alex Gonzalez. Phillips was the best at his position of any Red in terms of runs saved vs. average, but Ross (who was ranked the 8th most valuable catcher in MLB) played a more valuable position, and this pushed his overall ranking to the top. Strangely, Phillips was rated as merely average by ZR, which speaks to the differences between the BIS and STATS hit location datasets.

In contrast, the Reds' worst fielders were Adam Dunn, Ken Griffey Jr., Edwin Encarnacion, and Scott Hatteberg. Dunn got hammered by all of the fielding statistics, so there's clear consensus that he's Not Good in left field. In contrast, Griffey was rated as roughly average by both Fans and ZR, but was hammered by RZR. This is largely because the latter statistic places far more weight on plays out of zone, which is where Griffey's performance really fell short. Edwin was also rated differently by the two datasets, though in his case it's not clear why RZR dislikes him so much more than ZR.

Overall, I really like how total fielding value ranks these players. Compared to rankings that are strictly relative to positions, the total fielding value rankings recognize the inherent differences in value among different positions, which results in appropriate looking boosts to guys playing hard positions (C, CF, SS) and penalties to guys playing easier positions (1B, LF, RF).

Total Value

Now, the moment we all (or, at least, I) have been waiting for: let's put together our offensive and fielding numbers and rank the '07 Reds non-pitchers based on their total value!!!

In the table below, RAR represents a player's offense (runs above replacement), while TtlFld represents a player's defensive value (+- fielding and a position adjustment). Total_Value is the combined value of offense and defense!

Name	POS	RAR	TtlFld	TotalValue
B Phillips	2B	30.2	9.2	39.4
A Dunn	LF	51.3	-14.9	36.4
K Griffey Jr.	RF	36.4	-10.4	26.0
J Hamilton	CF	25.2	-0.2	25.0
S Hatteberg	1B	26.9	-5.5	21.4
A Gonzalez	SS	13.3	6.2	19.5
J Keppinger	SS	17.1	0.0	17.1
N Hopper	CF	10.8	6.8	17.6
EEncarnacion	3B	24.8	-8.2	16.7
D Ross	C	-4.6	12.3	7.7
J Valentin	C	3.2	-0.1	3.1
J Cantu	1B	3.8	-1.4	2.4
J Votto	1B	7.3	-4.1	3.2
R Freel	CF	-0.2	2.5	2.2
D Wise	CF	0.3	0.6	0.9
J Conine	1B	4.6	-3.4	1.2
R Jorgensen	C	0.4	0.0	0.4
R Hanigan	C	0.3	0.0	0.3
J Ellison	RF	-1.8	1.8	0.0
E Cruz	SS	-0.2	-0.7	-1.0
M Bellhorn	3B	-1.4	0.0	-1.3
B Coats	RF	-1.4	-0.4	-1.8
P Lopez	SS	-3.6	-0.4	-4.1
C Moeller	C	-4.8	-0.3	-5.1
J Castro	SS	-8.2	-1.3	-9.4

Isn't that exciting? Seven long articles worth of methods just to produce this little table!!!! :)

After all of that work, I'm quite comfortable saying what many would probably have been willing to say from the get-go: Brandon Phillips was the most valuable position player on the Cincinnati Reds in 2007! He was their third most valuable hitter, and the second most valuable fielder. Overall, it was an outstanding season by the Reds' second baseman. Hopefully he can continue that success next year.

For all my concerns about Adam Dunn's defense, he still came in a close second as the Reds' second most valuable position player thanks to his tremendous offensive performance. Josh Hamilton also had a strong showing, especially when you consider that he only had 337 PA's. Just think of where he'd be if he could get 650-700 PA's next season while maintaining this level of performance. I get misty-eyed when I think about it.

Without going back to repeat the math throughout his career, Alex Gonzalez may well have had the best all-around season of his career despite getting only 430 PA's. And Hopper and Keppinger may have had what will, in fact, turn out to be the best seasons of their MLB careers (though I'm obviously hoping they have more in the tank).

On the flip side, despite early-season struggles offensively, Edwin Encarnacion did manage to come out with positive value over replacement in this analysis, which is probably an improvement over 2006. His hitting slipped a bit this season overall, but he did show some improvement in his defense. If he can continue to improve on defense, while getting his offensive production back to his '06 rates over a full season, he could be a force to be reckoned with. But if he can't get his offense back on track for a full season, or his fielding slips to '06 levels of misery, he might not keep his job much longer. The kid's only 24, but the Reds will only wait so long for him to put things together. Next season might be his make-or-break season, at least as far as the Reds are concerned.

Finally, David Ross shows up better than I expected thanks to his strong showing on defense. Nevertheless, 8 runs above replacement-level is not acceptable performance from someone who got ~350 PA's on this team. Catching was very clearly the biggest hole in the Reds' lineup, and the place where they could potentially gain the most ground via an acquisition. ... not that the free agent market for catchers is particularly rich.

Coming up next: pitchers!

Brandon Phillips Photo by AP/David Kohl
Edwin Encarnacion Photo by Cincinnati Enquirer

Moved to Feedburner

Hi folks,

I've updated my RSS and Atom feeds to route through Feedburner. Feedburner has a lot of nice features that I'd like to take advantage of, including better compatibility and the availability of some usage stats. :)

Nevertheless, I ran into a couple of problems while porting it over related to the size of the feed. I think I've resolved those issues, but if you're still having difficulties with my feed, you may find it worthwhile to re-subscribe using this feed (click on it):

Sorry for the inconvenience. Please let me know if you have any additional problems!