Saturday, October 09, 2004
An Ad Hoc AL Fielding Metaranking Study
Believe me, I understand the complexities of cobbling together existing fielding statistics to arrive at a definitive assessment of how many runs players and teams save or allow by their defensive play. Or more exactly, I understand that I don't know it, and nobody else really does, either. Much greater minds than you or I have been stumped by this riddle.
However, I'm on the leading edge of the MTV Generation, so I want an answer, like, now. So rather than spend the next several years studying this enigma, I'm going to serve up an ad hoc system that took me all of about three hours to think about and develop.
In this system, I take three basic defensive measurements -- Fielding Percentage, adjusted Range Factor, and Zone Rating -- and mix and match a team's ranking within each for each position. (Adjusted Range Factor reflects the factoring out of strikeouts from total outs, since current the RF measurement penalizes teams that play with high strikeout pitching staffs.)
The blended ranking -- or "metaranking" - that results essentiallys sum the rankings in each. But rather than summing them up directly, I weight the rankings for each position based on the relative importance each metric counts for the position.
For instance, for SS and 2B, the weighting is: FPCT*1 + aRF*3 + ZR*5 = metaranking. The ability to get to maximum balls hit in your area is very important, and the actual error rate on those balls are relatively negligible. On the flip side, for 1B, the weighting is FPCT*2 + aRF*1 + ZR*1. For this position the ability to handle all types of throws cleanly is most valued, with less emphasis on range, thus more emphasis on fielding percentage.
Please note that the weightings I apply are not based on any statistical analysis I did for each position. It was more like, wet my finger, stick it in the air, and use the prevailing winds (and my 35-year knowledge of baseball) to arrive at what seems to be a reasonable weighting for each position.
Here is a link to how each team in the AL metaranks at each position:
http://www.geocities.com/chuck_hildebrandt/Positional_PVT_2004_post.htm
And, this being a Tigers blog, the Tigers' 2004 final fielding stats:
http://sports.espn.go.com/mlb/teamfieldstats?team=det
Here is how the Tigers rank against the best and worst teams at each position. The set of three numbers next to each position indicate the weighting of ranks in FPCT, aRF and ZR respectively; the set of three numbers next to each team indicate that teams rank in each of those three metrics:
1B (3/1/1)
First: Detroit 1/9/5 = 17
Last: Seattle 12/14/10 = 60
Carlos Pena has been a solid, sharp 1B for the Tigs, but Dmitri Young surprisingly turned in 211 innings of perfect play at 1B, so that helped Detroit achieve a #1 ranking here.
2B (1/3/5)
First: Toronto 5/1/6 = 38
Detroit: 6th 13/4/7 = 60
Last: Seattle: 12/14/10 = 104
Omar Infante drove most of the Tiger rating with over 800 innings, and he was a shade above average. Jason Smith and Fernando Vina pushed up the team's aRF in their 500 innings. And you gotta admit: Seattle is awfully consistent (and consistently awful) so far.
SS (1/3/5)
First: Oakland 4/2/2 = 20
Detroit: 5th 6/4/7 = 53
Last: Texas: 9/13/13 = 113
Another respectable showing for the Tigers as Carlos Guillen clocked 1,150 innings here. Couple with his .318/.379/.542 A-Rod-imitative offensive performance, he was clearly the best shortstop in the league this year. In fact, if he could have played 160 games, Guillen would have had a fairly legitimate shot at an MVP award.
3B (1/2/2)
First: Oakland 2/1/1 = 6
Detroit: 12th 13/5/14 = 51
Last: Anaheim 11/8/13 = 53
This is what not having a major league-quality 3B will do for you: put you two points from the bottom in fielding metarankings. When you need Brandon Inge to outperform everyone else to keep you out of the basement, you know you have trouble. This is a key reason I added 3B to our list of Tigers' needs.
How about Eric Chavez -- #1 in aRF and ZR, #2 in 3B fielding. The guy is worth his weight in gold, baby. (And while we're at it, Bobby Crosby doesn't suck at SS, either.)
LF (2/5/8)
First: Tampa Bay 2/2/4 = 46
Detroit: 3rd 14/8/1 = 76
Last: Boston 12/14/14 = 206
Craig Monroe's poor performance in LF notwithstanding, Tigs fare pretty well here with three main people you wouldn't expect to do so hot. Rondell White's and Marcus Thames' outstanding ZRs make the difference here.
CF (1/3/5)
First: Seattle 6/1/3 = 24
Detroit: 7th 14/10/4 = 64
Last: Texas 3/14/13 = 110
Just how bad is Alex Sanchez as an outfielder? If he played every inning in CF this year, he would be dead last in every category, by a long shot. As it stands, Nook Logan, Brandon Inge and Craig Monroe (who inexplicably made no errors in 189 CF innings while making eight errors in 446 innings, with decent ZR) lifted the Tigers up to 7th place in the metaranking.
RF (2/5/8)
First: Detroit 13/6/1 = 64
Last: New York 9/9/13 = 167
Bobby Higginson got more comfortable in his new position and turned in a typically fine defensive effort out there, even tying for the league lead in assists. (I personally saw him get a great one in July in which he threw out Ross Gload trying to advance to third on a second, throwing the ball to third on the fly. Amazing.) Fact is, though, it was Craig Monroe that elevated the Tigers to 1st place, again with great aRF and ZR.
Always good to see the Yankees in last place. Thanks, Sheff!
P (1/2/2)
First: Texas 1/1/2 = 7
Last: Detroit 12/13/8 = 54
Mike Maroth did very well, and Gary Knotts was passable, but the rest of the starting staff was just awful awful awful.
You will notice that I don't have a table available for catcher, since the three measures don't really apply to the position, and the ESPN team fielding table for catchers doesn'thave unique catcher fielding metrics such as SB attempts, CS %, or PBs. But reviewing the top catchers listed on the individual catching stats table, we see Pudge's reputation kept steal attempts at the lowest per 9 innings of any qualifying catcher, although CS % was middle of the road (perhaps because only the very best base-stealers made attempts against him). He also had fewer-than average passed balls and more-than-average errors and assists. All in all, he looks to have been an average-to-high-average catcher for the Tigers, not bad considering the wear and tear he has put on his body in his 16 years of professional ball.
However, I'm on the leading edge of the MTV Generation, so I want an answer, like, now. So rather than spend the next several years studying this enigma, I'm going to serve up an ad hoc system that took me all of about three hours to think about and develop.
In this system, I take three basic defensive measurements -- Fielding Percentage, adjusted Range Factor, and Zone Rating -- and mix and match a team's ranking within each for each position. (Adjusted Range Factor reflects the factoring out of strikeouts from total outs, since current the RF measurement penalizes teams that play with high strikeout pitching staffs.)
The blended ranking -- or "metaranking" - that results essentiallys sum the rankings in each. But rather than summing them up directly, I weight the rankings for each position based on the relative importance each metric counts for the position.
For instance, for SS and 2B, the weighting is: FPCT*1 + aRF*3 + ZR*5 = metaranking. The ability to get to maximum balls hit in your area is very important, and the actual error rate on those balls are relatively negligible. On the flip side, for 1B, the weighting is FPCT*2 + aRF*1 + ZR*1. For this position the ability to handle all types of throws cleanly is most valued, with less emphasis on range, thus more emphasis on fielding percentage.
Please note that the weightings I apply are not based on any statistical analysis I did for each position. It was more like, wet my finger, stick it in the air, and use the prevailing winds (and my 35-year knowledge of baseball) to arrive at what seems to be a reasonable weighting for each position.
Here is a link to how each team in the AL metaranks at each position:
http://www.geocities.com/chuck_hildebrandt/Positional_PVT_2004_post.htm
And, this being a Tigers blog, the Tigers' 2004 final fielding stats:
http://sports.espn.go.com/mlb/teamfieldstats?team=det
Here is how the Tigers rank against the best and worst teams at each position. The set of three numbers next to each position indicate the weighting of ranks in FPCT, aRF and ZR respectively; the set of three numbers next to each team indicate that teams rank in each of those three metrics:
1B (3/1/1)
First: Detroit 1/9/5 = 17
Last: Seattle 12/14/10 = 60
Carlos Pena has been a solid, sharp 1B for the Tigs, but Dmitri Young surprisingly turned in 211 innings of perfect play at 1B, so that helped Detroit achieve a #1 ranking here.
2B (1/3/5)
First: Toronto 5/1/6 = 38
Detroit: 6th 13/4/7 = 60
Last: Seattle: 12/14/10 = 104
Omar Infante drove most of the Tiger rating with over 800 innings, and he was a shade above average. Jason Smith and Fernando Vina pushed up the team's aRF in their 500 innings. And you gotta admit: Seattle is awfully consistent (and consistently awful) so far.
SS (1/3/5)
First: Oakland 4/2/2 = 20
Detroit: 5th 6/4/7 = 53
Last: Texas: 9/13/13 = 113
Another respectable showing for the Tigers as Carlos Guillen clocked 1,150 innings here. Couple with his .318/.379/.542 A-Rod-imitative offensive performance, he was clearly the best shortstop in the league this year. In fact, if he could have played 160 games, Guillen would have had a fairly legitimate shot at an MVP award.
3B (1/2/2)
First: Oakland 2/1/1 = 6
Detroit: 12th 13/5/14 = 51
Last: Anaheim 11/8/13 = 53
This is what not having a major league-quality 3B will do for you: put you two points from the bottom in fielding metarankings. When you need Brandon Inge to outperform everyone else to keep you out of the basement, you know you have trouble. This is a key reason I added 3B to our list of Tigers' needs.
How about Eric Chavez -- #1 in aRF and ZR, #2 in 3B fielding. The guy is worth his weight in gold, baby. (And while we're at it, Bobby Crosby doesn't suck at SS, either.)
LF (2/5/8)
First: Tampa Bay 2/2/4 = 46
Detroit: 3rd 14/8/1 = 76
Last: Boston 12/14/14 = 206
Craig Monroe's poor performance in LF notwithstanding, Tigs fare pretty well here with three main people you wouldn't expect to do so hot. Rondell White's and Marcus Thames' outstanding ZRs make the difference here.
CF (1/3/5)
First: Seattle 6/1/3 = 24
Detroit: 7th 14/10/4 = 64
Last: Texas 3/14/13 = 110
Just how bad is Alex Sanchez as an outfielder? If he played every inning in CF this year, he would be dead last in every category, by a long shot. As it stands, Nook Logan, Brandon Inge and Craig Monroe (who inexplicably made no errors in 189 CF innings while making eight errors in 446 innings, with decent ZR) lifted the Tigers up to 7th place in the metaranking.
RF (2/5/8)
First: Detroit 13/6/1 = 64
Last: New York 9/9/13 = 167
Bobby Higginson got more comfortable in his new position and turned in a typically fine defensive effort out there, even tying for the league lead in assists. (I personally saw him get a great one in July in which he threw out Ross Gload trying to advance to third on a second, throwing the ball to third on the fly. Amazing.) Fact is, though, it was Craig Monroe that elevated the Tigers to 1st place, again with great aRF and ZR.
Always good to see the Yankees in last place. Thanks, Sheff!
P (1/2/2)
First: Texas 1/1/2 = 7
Last: Detroit 12/13/8 = 54
Mike Maroth did very well, and Gary Knotts was passable, but the rest of the starting staff was just awful awful awful.
You will notice that I don't have a table available for catcher, since the three measures don't really apply to the position, and the ESPN team fielding table for catchers doesn'thave unique catcher fielding metrics such as SB attempts, CS %, or PBs. But reviewing the top catchers listed on the individual catching stats table, we see Pudge's reputation kept steal attempts at the lowest per 9 innings of any qualifying catcher, although CS % was middle of the road (perhaps because only the very best base-stealers made attempts against him). He also had fewer-than average passed balls and more-than-average errors and assists. All in all, he looks to have been an average-to-high-average catcher for the Tigers, not bad considering the wear and tear he has put on his body in his 16 years of professional ball.
