Team Defense

by rglass44 on Oct 13, 2008 2:16 PM EDT

There has been a lot of talk this season about the team's defense, and how it compares to other teams and last year. Sky has introduced us all to Justin Inaz's Total Value list and it's defensive analysis. It seemingly has the best rating system for a player's defense because it uses the average of two measures and converts it to runs. Looking at the player ratings, the Rays seem to be about average as far as defense goes. Subjectively this does not make sense. Let's take a look at how defenses around the league look compared to their pitcher's true abilities.

*props to StatCorner and Justin Inaz of On Baseball for letting me use info garnered from their respective websites.

The following table shows the difference between the real runs allowed by a team versus the hypothetical runs the pitcher has allowed. Using the formula for tRA, I found the utRA for each team's pitching staff (utRA is unadjusted tRA because I did not want to introduce parkfactors because the real runs were scored in the real environments the LD, SO, BB, GB, etc. data came from). It also shows team FRAA, which is the sum of all players for each team's Fielding Runs according to Justin Inaz.

Team	RA	utR	tFR	tFR%	tFR Rank	Team FRAA	FRAA Rank
STL	725	819	93.75	11.45%	1	46.9	3
PHI	680	767	86.91	11.33%	2	51.4	1
MIL	689	765	76.06	9.94%	3	29.2	6
TOR	610	674	63.77	9.46%	4	38.1	5
TB	671	738	66.90	9.07%	5	25.1	9
NYN	715	768	52.52	6.84%	6	13.8	12
OAK	690	739	48.60	6.58%	7	49.1	2
SD	764	785	20.54	2.62%	8	26.4	8
LAA	697	709	12.36	1.74%	9	7.8	13
SEA	811	821	10.10	1.23%	10	-15.4	20
CIN	800	807	7.49	0.93%	11	-8.6	17
CHN	671	673	2.06	0.31%	12	20.3	11
BOS	694	695	1.13	0.16%	13	-31.3	25
SF	759	758	-0.76	-0.10%	14	3.1	15
LAN	648	646	-1.61	-0.25%	15	21.9	10
NYA	727	725	-2.46	-0.34%	16	-56.0	29
FLA	767	761	-5.75	-0.76%	17	-27.0	24
DET	857	850	-6.88	-0.81%	18	-10.9	18
WAS	825	818	-7.24	-0.89%	19	-0.4	16
MIN	745	737	-8.40	-1.14%	20	-42.4	28
HOU	743	733	-9.79	-1.34%	21	40.6	4
CLE	761	749	-11.53	-1.54%	22	7.4	14
BAL	869	855	-13.53	-1.58%	23	-38.6	26
PIT	884	861	-22.83	-2.65%	24	-25.4	23
KC	781	745	-35.80	-4.80%	25	-57.5	30
COL	822	776	-46.00	-5.93%	26	-11.8	19
ATL	778	734	-44.19	-6.02%	27	27.5	7
ARI	706	665	-41.39	-6.23%	28	-22.8	21
CHA	729	684	-45.28	-6.62%	29	-25.0	22
TEX	967	886	-80.78	-9.11%	30	-40.7	27

The interesting thing here is that a lot of these rankings are very similar to what the Fielding Ratings are. The outliers are the most interesting, though. The big movers are Houston and Atlanta that both drop considerably. This seems to make sense looking at the rosters of the two teams.

The Rays climb up the ratings into the top 5. Looking at the top and bottom 10s of each list the tFRA list seems more in-line with what I would consider the better fielding teams. One thing that sticks out is the teams with the strength up the middle tend to do fairly well.

It's too bad I'm not better with stats because I think some useful information might be gleaned from this. Maybe the statcorner guys want to embark on a true fielding measure next.

This makes me wonder even more about the reliability of any defensive metrics we currently have. HitFX!!!!!

Tweet Comment 5 comments | 0 recs |

Do you like this story?

Comments

very interesting stuff

and there’s a lot more you could do with this data, too.

one thing tRA doesn’t account for is the the location of each batted ball type. so two pitchers might both allow 50 ground balls, but one guy allows them in holes and the other allows them right at fielders. tRA says they’re even, making RA-utR show the at’em pitcher’s fielders to be better, while zone ratings would show the fielders to be similar. now, allowing ground balls in different locations probably isn’t much of a skill (but probably SOME skill) but it will certainly happen over a full season.

Beyond the Boxscore // Calling BJ Upton lazy is lazy.

by Sky Kalkman on Oct 14, 2008 10:50 AM EDT reply actions

One would think that with all the pitchers on a staff

That would even out. During the offseason I think I may try to see if any factors seem to “drive” the results in any way. We’ll see though.

by rglass44 on Oct 14, 2008 11:01 AM EDT up reply actions

it doesn't

Dave Pinto’s applied his defensive system to pitchers and found significant differences over the course of a season. That’s also one reason individual pitchers post ridiculously high/low BABIPs over one season — their batted balls happen to all go right at fielders or all go right at gaps. It’s not always good/bad fielding that explains BABIP.

Sure, combining five starters reigns in the craziness, but a difference of 30-50 runs is only 10 extra plays made/not made per pitcher and about .40 runs of ERA. nothing crazy.

Beyond the Boxscore // Calling BJ Upton lazy is lazy.

by Sky Kalkman on Oct 14, 2008 2:33 PM EDT up reply actions

the correlation between tRA-implied fielding and zone rating fielding is .67, pretty solid

the distribution of tRA-implied fielding is wider, with an SD of 42 versus 32 for FRAA. that’s likely because tRA-implied fielding includes both fielding and “luck” of batted-ball location while FRAA only inludes fielding (which accounts for batted-ball location).

Beyond the Boxscore // Calling BJ Upton lazy is lazy.

by Sky Kalkman on Oct 14, 2008 2:36 PM EDT reply actions