Cardinals Individual UZR, overall WAR, and positional value optimization
We have two months of data for 2009 UZR on Fangraphs. By analyzing the Cardinals' defensive team audit page, we can see who's been giving the team some love with the glove, who's been sub-par, and what we can expect going forward. I hope that by looking at the fielding data and overall WAR, I can reinforce a consensus opinion about the team's optimal utilization of resources. I'm only looking at the data from players from this year. More after the jump.
|
Car. Innings |
Car. UZR/150 |
BRAA |
ZiPS ROS BRAA/(600 PA) |
||||||
|---|---|---|---|---|---|---|---|---|---|
|
LF |
2 |
14.0 |
0.8 |
63.0 |
161.0 |
-32.7 |
-0.2 |
7.8 |
|
|
CF |
26 |
215.2 |
-0.1 |
3.5 |
1119.0 |
-6.5 |
|
|
|
|
RF |
7 |
49.0 |
-0.1 |
11.9 |
248.2 |
14.8 |
|
|
|
|
1B |
59 |
522.0 |
-1.7 |
-3.8 |
7804.0 |
6.4 |
23.8 |
61 |
|
|
LF |
|
|
|
|
686.2 |
7.4 |
-1.0 |
18.2 |
|
|
CF |
|
|
|
|
269.2 |
5.1 |
|
|
|
|
RF |
39 |
315.1 |
2.0 |
-2.6 |
1899 |
8.3 |
|
|
|
|
SS |
29 |
238.1 |
-5.9 |
-26.3 |
5939.2 |
-3.9 |
-6.4 |
-7.8 |
|
|
1B |
2 |
9.0 |
0.4 |
53.4 |
246.1 |
-0.5 |
-1.0 |
4.8 |
|
|
LF |
51 |
396.1 |
-1.0 |
-4.6 |
1792.0 |
-7.9 |
|
|
|
|
CF |
2 |
10.0 |
1.1 |
52.9 |
664.1 |
-10.3 |
-1.8 |
-2.2 |
|
|
RF |
8 |
21.1 |
-0.1 |
-3.5 |
423.1 |
-8.7 |
|
|
|
|
LF |
22 |
45.2 |
-3.1 |
-68.9 |
592.0 |
3.1 |
|
|
|
|
2B |
48 |
349.0 |
-8.6 |
-26.3 |
349.0 |
-26.3 |
|
|
|
|
RF |
9 |
60.1 |
0.0 |
-0.9 |
80.1 |
5.0 |
-1.9 |
-14.6 |
|
|
LF |
6 |
27.0 |
-0.6 |
-87.1 |
80.0 |
0.1 |
|
|
|
|
CF |
2 |
2.2 |
-0.2 |
3.0 |
- |
- |
-2.1 |
-29.2 |
|
|
RF |
6 |
41.1 |
-1.0 |
-35.1 |
- |
- |
|
|
|
|
LF |
1 |
2.0 |
|
|
- |
- |
|
|
|
|
3B |
44 |
203.2 |
2.2 |
13.5 |
232.2 |
13.9 |
-2.3 |
-14.6 |
|
|
SS |
4 |
27.0 |
-0.2 |
-5.7 |
- |
- |
|
|
|
|
2B |
1 |
7.0 |
-0.4 |
0.0 |
- |
- |
|
|
|
|
SS |
12 |
92.2 |
1.4 |
19.8 |
- |
- |
1.8 |
-20.8 |
|
|
3B |
6 |
32.0 |
-0.9 |
-128.3 |
- |
- |
|
|
|
|
2B |
4 |
6.1 |
-0.9 |
-145.7 |
- |
- |
|
|
|
|
SS |
23 |
176.0 |
5.9 |
42.5 |
595.0 |
21.9 |
1.4 |
-16.8 |
|
|
2B |
15 |
84.2 |
1.7 |
24.5 |
358.2 |
1.9 |
|
|
|
|
3B |
|
|
|
|
169.1 |
-4.1 |
|
|
|
|
2B |
29 |
87.0 |
2.2 |
30.3 |
133.2 |
13.0 |
0.0 |
-5.8 |
|
|
3B |
38 |
268.0 |
-3.3 |
-12.4 |
268.0 |
-12.4 |
|
|
|
|
3B |
5 |
30.1 |
0.4 |
12.0 |
- |
- |
|
-2 |
|
|
CF |
40 |
305.2 |
10.3 |
38.2 |
- |
- |
2.6 |
0 |
|
|
RF |
6 |
46.2 |
1.3 |
83.3 |
- |
- |
|
|
|
|
LF |
7 |
40.0 |
0.1 |
-10.6 |
- |
- |
|
|
Here are a few assumptions behind this chart:
1) I am including each player's 2009 UZR fielding data and their career data for each position. Any player who only has 2009 data has a dash in their career columns. This should help overcome some sample-size issues, but by doing so I also introduce into the sample any improvement or deterioration of defensive skills.
2) I am using the ZiPS Rest of Season projections to predict a players wOBA from here on out. I am then converting that wOBA to to BRAA (batting runs above average) over 600 PAs by using the formula (wOBA-.330)xPAs/1.15.
3) I didn't try to monetize value or introduce any positional adjustment because I'm solely concerned with the current roster and absolute performance...not any sort of replacement, trade or free-agent win computing exercize.
4) I also didn't include any minor league fielding data, primarily because I didn't know where to look and I wanted to stick with UZR, which I am most familiar with. Be cautious of the small samples generated for some of the players.
5) Mather and Glaus probably deserve to be on there, but I got lazy and I'm only looking at folks who have played this year.
Positional Analysis
C: Catching is a difficult position to capture through any metric, so I didn't even bother putting it on there. Yadi's defensive prowess is almost universally praised.
1B: Albert
2B: The first position where there is controversy. Schumaker: has gotten the bulk of playing time at 2B, but has cost the team 8.6 runs below an average second baseman according to UZR. According to the data on fangraphs, his biggest liability has been a lack of range (-9.5 range runs) and his average sure-handedness doesn't make up for it. Among qualified second basemen, Skip is the worst second baseman in baseball, 1.6 runs behind Dan Uggla. His bat is pretty much league average, and this year he's hitting RHP only slightly better than LHP (which is a commendable improvement for the Schu Man) but still. His reputation for defense saves in LF hasn't born fruit, either, since he's -3.1 runs this year and merely a career 3.1 runs average in the Dunn Siberia. Barring any sort of breakthrough in his defense, It might be time to stick a fork in this turkey of an experiment.
Thurston: Over 88 innings at 2B this year, Joe has contributed above-average defense and his career fielding data is in the +13 runs/150 games range. His offensive skills are ok according to wOBA and is only a few runs below average from here on out...should be about a league average-ish second baseman from here on out based on his ZiPS ROS wOBA and positive defense at 2B. He's always deserved a starting role, but has been blocked by elite 2B talent at LAN PHI and BOS
Ryan: Another good 2B, although not as good as Thurston. His career UZR/150 at 2B is only 1.9 RAA. He's been above average with the bat this year and is stinging the ball at a 23.7% LD rate, but that and his .376 BABIP should regress. ZiPS ROS is skeptical of his success this year.
T. Greene: Too early to say anything about his second base skills, but he's a SS. ZiPS is also skeptical of his success thus far. More later.
Barden: Not enough data, but he's likely a better 3B.
3B: Again, there are many suitors for this position on the Cardinals. I'll start with Barden who has been playing a very good defensive 3B so far this year. Over his major league career he's like 13 runs above average at the hot corner. ZiPS hates his bat for the rest of the season, though. Despite his whopping 3HRs this year, he's kind of like Cesar Izturis of the 3B.
Ryan: He hasn't played any 3B this year, but he's been -4.1 runs/150 games below average for his career.
Thurston: UZR really doesn't like Joe at 3B. -3.3 UZR runs this year. Pales in comparison to his 2B defense.
T. Greene: Too early to draw any conclusions about his D at 3B.
Freese: He's injured now (woe is the Future Redbirds reader), but Freese posted good numbers defensively. Furthermore, I remember reading somewhere that his minor league defense was in the vecinity of like +13 runs defensively. ZiPS also loves his offense, projecting him to be only 1 run below average for the rest of the year. If he hit like ZiPS called for him to over an entire season, he might be worth 2-3 wins. Nice asset.
SS: Controversy continues here.
Ryan: UZR loves Brendan Ryan at SS. Based on the data, he's 21 defensive runs above an average SS over 150 games. He's also been an above average offensive player this year, primarily based on his %23 LD rate and high BABIP. He does have a Expect that to regress, somewhat, but even if he hits 8 runs below average from here on out (like ZiPS calls for) he's still a great talent.
T. Greene: Tyler "First Round Pick" Greene has also been very good with the glove and has displayed some nice power and increased plate discipline this year in very limited playing time. ZiPS is skeptical of his performance from here on out, but I think his offensive performance might be legit due to his career .180 ISO. Lets hope he keeps getting fastballs.
K. Greene: I hate to suggest this, but it might be time to cut ties with K. Greene. His has not performed defensively or offensively this year and I'm not excited about his defensive rebound potential. Indeed, his career UZR is -3.9/150 games. ZiPS thinks he'll cost the Cardinals ~4 more runs below average from here on out, only about 4 more than Ryan, but that's assuming that he gets back into playing shape. I like Khalil, but still... I'll take the vacuuming Ryan and his LD-enhanced BABIP than K. Greene at this point.
This get a little bit more clear in the OF and I'm getting a little verbose:
LF: Ryan Ludwick: Ludwick has performed admirably in LF throughout his entire career even though his defense is a run or two better throughout the whole year in RF (and isn't bad in CF for his career). ZiPS likes his bat going forward. He makes the most sense in Left given our other options.
Ankiel: Ankiel gets better defensively as the numbers increase from LF-RF. He's actually not a bad CF, but he doesn't have a huge sample size in LF to base performance off of. I also like his man-cannon arm out in the right side of the outfield. ZiPS likes his bat despite the slow start this year.
Duncan. Meh. His defense is below average, but not terrible, and his offense only projects to be ~5 runs above average for the whole year. I'm not sure what a replacement level LF is this year, but Duncan is probably only 1 or 1.5 wins above that.
Schuman: Probably a little bit more valuable in LF than Duncan, but he's kind of a 4th OF, imo.
Shane & Stav: Kind of fun to look at, but I am sick and tired of having Aaron Miles in the OF. At this point they're both replacement. I'm not sure about Shane's ceiling, however.
Colby: Colby has a great glove, but rules CF.
CF: Like I said before...Colby. He's been worth one win with his glove, alone, and should provide league average-ish offense from here on out. All hail Colby. Ank actually isn't that bad CF.
RF: While Ludwick is probably the better RF and Ankiel only has 250 innings to support this: I like Ankiel in RF. Career UZR numbers are best there, and his bat is more RF than LF.
Not really surprising, but here's what I've supposed to be the optimal line-up factoring defense and offense:
2B: Thurston--the best defensive 2B and has close to league-average bat. Ryan next.
SS: Ryan--Crazy UZR/150 numbers combined with a career high LD% make him a valuable asset this year. He did hit 4HRs in 2007. T Greene is close, but his offensive success is more of a question mark due to high K%...does posses a nice career ISO mark.
3B: Murkier, due to organizational injury and depth, but I'll give you a WAR order: Glaus>Freese>Barden>T Greene One caveat is whether or not Greene becomes a plus defender at 3B. If T Greene be a +7 run defender at 3B and wOBA near league average, then I'd take him over Barden.
LF: Ludwick, due to Ankiel's prowess in RF. Ludwick is still a plus defender in LF, although his bat is relatively weaker due to the Manny-factor.
CF: Rasmus. ZOMG
RF: Ankiel: better than Juan Encarnacion. It also might be worth it to see Mather thrown in there on a semi-regular basis.
Schu: 4th outfielder, no business at 2B.
Duncan: DH
36 comments
|
6 recs |
Do you like this story?
Comments
good work!
well thought out fanpost. there isn’t, as you said, much surprising here, but it is nice to have the statistics thoughtfully compiled and presented. thanks!
cardinals are the things with feathers that perch within the soul.
Do you think that anyone would be upset
If we had Duncan DH this year? It would be nice if someone would tell Tony to go ahead had user the Ludwick-LF Rasmu-CF Ankiel-RF all year.
As far as skippy goes; it would be interesting to see how his range is at this time next year.
I definitely
think the Schu experiment warrants some more work given strong DP skills. There is little doubt that Ludwick 7 Ras 8 and Ank 9 is our warpiest OF
"Never judge a taco by its price" - Dr. Gonzo
I disagree
When you consider defense, Skip in a corner is probably as valuable as Ankiel. However, is there any doubt that Ankiel would bring back more in a trade?
St. Louis relievers... defying win expectancy since 2008
http://www.drivelinemechanics.com/
by vivaelpujols on Jun 12, 2009 1:27 AM EDT up reply actions
I'm really torn on Schu
But I feel our best chance of winning THIS year is to platoon him with a RH in LF. However, I don’t feel that his defence has been as bad as UZR suggests so far. I guess we’ll see at the end of the year.
My concern with Skip at 2B is that he simply isn’t going to get any better – he’s pretty much mastered the DP ball IMO but his problem is his first step, speed & agility getting to the ball (i.e. range) and I just don’t think that’s something he’s going to learn.
Because chicks dig the intentional base on balls.
by Felonius_Monk on Jun 12, 2009 6:44 AM EDT up reply actions
Also
we have a LH-hitting 2B with similar offensive skills to Schu who is, regardless what you make of Schu’s 2B performance so far, a superior fielder, and that’s Thurston.
If we obtain a 3B, the sensible thing to do is use Thursty in exactly the role that Schu’s had so far. In fact, because Thurston’s a bit better vs LHP, and brings more value with the glove, you could argue that he could play every day and be average or slightly above at 2B. Great FAT pickup by Moz.
Thurston at 2B, T Greene at SS, Ryan as backup MIF, Albert at 1B and a legit 3B (DeRo? Beltre?) and sudeenly we’ve got an excellent infield defence. Ludwick in right every day, Rasmus in centre every day, one of Schu/Ank in left platooning with a right-hander and the other guy sitting on the bench every day, now we have an excellent OF defence. Duncan to AAA. Pineiro, Welly and Thompson say “thankyou”.
Because chicks dig the intentional base on balls.
by Felonius_Monk on Jun 12, 2009 6:48 AM EDT up reply actions
yes, please.
the truth can't hurt you, it's just like the dark/ it scares you witless, but in time you see things clear and stark -- macmanus
bud selig? he might object that we're not in the AL.
the truth can't hurt you, it's just like the dark/ it scares you witless, but in time you see things clear and stark -- macmanus
Not an issue
We’ll just come up with some graphs showing it’d increase “parity” and lead to larger T.V. advertising revenues.
There's no "I" in team. There's also no "I" in "B-g Mac Land".
"1B: Albert"
Brilliant.
"If I prepare myself, my stuff is good and I'm going to get outs. That is a fact." - Chris Carpenter
by spants on Jun 11, 2009 11:19 PM EDT reply actions 1 recs
what more do you need?
It kind of sounds like he’s [Duncan] just running around like a puppy out there – full speed ahead in random directions. – BTown Birds Fan
what I thought
4B - beer baseball bands blog
"OOHHHHH!!!! He knocked out the I in Big Mac Land!!"
by Cards Fan in Chitown on Jun 12, 2009 2:32 AM EDT up reply actions
this post is a slam dunk
it pretty much confirmed what I was thinking of for the positions as of late. Dunc and Schu need to be riding the bench a lot. Thurston is a great second baseman, Ryan has looked spectacular at short, Ank can play center but Rasmus is a force to be reckoned with there, Ludwick needs to play nearly every day (unless he’s not as durable as we think he is), 3B is still up in the air. I think Tyler Greene could use a little more playing time in the minors, unless Brendan starts to come back down to earth (I personally think he’ll keep up the good work). Barden is the de facto third baseman until something happens with the 3 big ifs (Glaus, Freese, trade).
the good: what if TLR has been treating the first part of the season as a research period trying to get info (as you just did) on where to play people, as well as who should be getting the bulk of the starts (so far he has been giving some really equal playing time to everyone as far as I can tell)
the bad: Tony feels the need to continue tinkering the entire season, much like he did last year, strangling the team with control issues
the ugly: well, a lot of things could make the lineup worse, won’t even go there
conclusion: I really hope we get to see more of Thurston at second and less of Dunc and Schu (unless Ank stops heating up, he can be scary bad at times, but it’s still impressive that he can play anywhere in the outfield at least capably). third base remains an interesting question, because none of the options there are really working that well. I have no idea what to do about Spockoli, he’s a bit of an albatross at this point.
4B - beer baseball bands blog
"OOHHHHH!!!! He knocked out the I in Big Mac Land!!"
by Cards Fan in Chitown on Jun 12, 2009 2:42 AM EDT reply actions
Maybe I don't understand UZR/150
but how can you have a positive UZR yet be negative when extrapolated over 150 games?
"People call me El Hombre," Pujols said. "But only Stan is the Man."
It's based off of the RngR component of UZR
So you can have a positive UZR if your ARM rating is good enough to offset bad range, but only your range will be extrapolated. Still, I really don’t like using UZR//150 this early in the season.
St. Louis relievers... defying win expectancy since 2008
http://www.drivelinemechanics.com/
by vivaelpujols on Jun 12, 2009 8:26 AM EDT up reply actions
so UZR and UZR/150 are really different measures
since one uses ARM while the other doesn’t? Also I have heard bad things about using defensive metrics with small sample sizes so I usually give them slightly more weight than a grain of salt.
"People call me El Hombre," Pujols said. "But only Stan is the Man."
Yes
St. Louis relievers... defying win expectancy since 2008
http://www.drivelinemechanics.com/
by vivaelpujols on Jun 12, 2009 9:14 AM EDT up reply actions
I wish they would have used different acronyms then
Because I always assumed UZR/150 was just UZR expanded to 150 games much like ERA is your earned runs averaged out over a 9 inning game.
"People call me El Hombre," Pujols said. "But only Stan is the Man."
UZR doesn't predict future performances
I am serious when I say 2 years of data to predict the future in regards to defense.
Stat Whore
by FlimtotheFlam on Jun 12, 2009 10:09 AM EDT up reply actions
I agree
St. Louis relievers... defying win expectancy since 2008
http://www.drivelinemechanics.com/
by vivaelpujols on Jun 12, 2009 10:18 AM EDT up reply actions
Defensive stats
Are always tricky to handle, but what should be the amount of innings that one needs in order to draw conclusions about defense? 2000?
"Never judge a taco by its price" - Dr. Gonzo
UZR has made us lazy.... No Offense
It makes us think we know more than we know. Too much data and too many variables. Compared to an offensive stat it might be as good as evaluating a player by batting average but honestly I don’t even think it is that good.
Stat Whore
Then it is completely useless
because BA is approaching meaninglessness…
Because chicks dig the intentional base on balls.
by Felonius_Monk on Jun 12, 2009 12:20 PM EDT up reply actions
are you saying there's a better stat to evaluate defense
or that we can’t really evaluate defense?
4B - beer baseball bands blog
"OOHHHHH!!!! He knocked out the I in Big Mac Land!!"
by Cards Fan in Chitown on Jun 12, 2009 1:05 PM EDT up reply actions
2nd one
defense right now has a bunch of stats that can be used over the life of a player but are almost worthless as snapshots in time. How do numbers tell you things like bad routes, bad dives, dumb-headed drops vs a good chance gone bad, lazy throws in, etc. Even UZR has some human component since someone has to determine which zone the ball was in, I don’t think we have concrete ways of saying “this ball was on this path past the SS” yet so someone has to assign a ball to a zone. Over the course of a season, or a few years the human factor can be worked out but over a few innings even a slight variation can cause huge issues. Just basic statistics, small sample sizes show outside influences much stronger than larger sample sizes.
"People call me El Hombre," Pujols said. "But only Stan is the Man."
I was thinking of doing something like
giving rating points to players depending on what their ranking is in each of the individual different defensive stats. but I seem to remember this already being done before, can’t remember.
4B - beer baseball bands blog
"OOHHHHH!!!! He knocked out the I in Big Mac Land!!"
by Cards Fan in Chitown on Jun 12, 2009 2:38 PM EDT up reply actions
The one that always stands out in my mind
Is Shifts in the field by player’s to a hitters tendency’s. It can make range look really good or really bad. Plus, When Rasmus is in CF and he has to cover Dunc on the left side what kind of effect does that have on his range?
There is just too much variables without a large amount of data.
A lot of it is Fangraph’s fault. Not in a bad way but some stats have become more available and easier. I still have major beefs with how they put a $$$ on value. That stuff just reeks of BS to me. 10 wins equals a run? Sounds like some metric BS to me. I spent this summer emailing back and forth between Sean “CHONE” Smith and Tom Tango about that exact topic.
I believe it is closer to something like (LgRunsAvg*2)*Cost of one win. So a Run in the AL would be worth less than a Run in the NL. Which makes sense the AL has the DL. But now I am just ranting.
Stat Whore
by FlimtotheFlam on Jun 12, 2009 4:25 PM EDT up reply actions
Ank vs Duncan
Why is Ankiel’s career UZR/150 of -6.5 in CF described as “isn’t that bad”, but Duncan’s -7.9 as “not terrible”. 1.4 isn’t that big a difference.
Ankiel looks better
Duncan looks sloppy
Stat Whore
by FlimtotheFlam on Jun 12, 2009 4:44 PM EDT up reply actions
A below average center fielder is WAY better than a below average left fielder
Ank is probably a + defender in left.
St. Louis relievers... defying win expectancy since 2008
http://www.drivelinemechanics.com/
by vivaelpujols on Jun 12, 2009 9:06 PM EDT up reply actions
CF is a more difficult position to defend than LF
Stat Whore
by FlimtotheFlam on Jun 13, 2009 2:10 PM EDT up reply actions
really?
4B - beer baseball bands blog
"OOHHHHH!!!! He knocked out the I in Big Mac Land!!"
by Cards Fan in Chitown on Jun 13, 2009 2:53 PM EDT up reply actions
Sarcasm?
"If I prepare myself, my stuff is good and I'm going to get outs. That is a fact." - Chris Carpenter
drip
4B - beer baseball bands blog
No results found for comming:
Did you mean Coming (in dictionary) or Commin (in reference)?
Dictionary suggestions: Coming
by Cards Fan in Chitown on Jun 16, 2009 1:03 PM EDT up reply actions

by 


















