Warning: Creating default object from empty value in /home4/derok6/public_html/chop-n-change/plugins/system/cdnforjoomla/helper.php on line 27
measuring-pitching-success | August | 2004 Articles

2004 Archives

Measuring Pitching Success

on .

Win percentage, save percentage, ERA, and WHIP are good, but maybe there's something better waiting to be discovered. This article focuses somewhat on relief pitching, but the conclusion applies to all pitchers. * * * One of the toughest baseball skills to quantify is successful relief pitching. ERA works moderately well, but its accuracy breaks down with the consideration of inherited runners. Relief pitchers who allow inherited baserunners to score aren't charged with earned runs for the inherited runners; the pitcher who allowed them to reach base is. A temporary solution for that would be to divide earned runs that were also inherited runs scored in half -- half of an earned run for the pitcher who allowed the runner, and another half for the pitcher who allowed him to score. This would not be perfect by any means, but it would at least assign some responsibility to each pitcher without unfairly putting it all on the guy who left the game and didn't actually allow the runner to score. Then there are saves, holds, strikeouts-per-(9)inning(s), and other derived statistics that measure specific talents or tendencies. But none of them seem to measure how good a relief pitcher is at simply doing his job -- holding or saving a lead or keeping a game close. Improving Save Percentage to account for Setup Relief One thing we could do is improve save percentage. When a relief pitcher earns a hold, it is because he came into the game in a save situation and successfully preserved the lead, but was replaced by another pitcher before the game ended. If a pitcher enters a game in a save/hold situation and relinquishes the lead, however, he is charged with a blown save, which is unfair to setup relievers who are not in the game to earn saves in the first place. The quick and dirty solution is to include holds in a new-and-improved save percentage that I have temporarily dubbed "saves plus holds percentage" (SH%). It's simple: S% = Sv / (Sv + BSv) SH% = (Sv + Hld) / (Sv + Hld + BSv) All the new number does is incorporate holds into the percentage, allowing the game's premiere setup relievers a way to compete with closers. For instance, the top tens in 2003 in both categories (20 Sv+Bsv required for Sv%, 20 Hld+Sv+BSv required for SH%):
CloserSvBSvSv%
E Gagne
R Beck
B Wagner
J Smoltz
D Kolb
E Guardado
M Mantei
K Foulke
J Borowski
T Percival
55
20
44
45
21
41
29
43
33
33
0
0
3
4
2
4
3
5
4
4
1.000
1.000
.936
.918
.913
.911
.906
.896
.892
.892
 
RelieverHldSvBSvSH%
E Gagne
R Beck
F Rodriguez
S Hasegawa
T Martin
T Miller
O Dotel
D Weathers
B Donnelly
A Lopez
0
1
19
12
28
16
33
26
29
16
55
20
2
16
0
4
4
7
3
14
0
0
0
1
1
1
2
2
2
2
1.000
1.000
1.000
.966
.966
.952
.949
.943
.941
.938
Once again, though, saves and holds are not a measure of good relief pitching, but rather a measure of one's success (or lack thereof) at preserving his team's lead when he enters the game in a save situation. There are too many conditions to be met for these statistics to be useful as a measure for all relief pitchers, so perhaps there is a better way. Back to DIPP (Defense-Independent Pitcher's Percentage) Recall that DIPP "is the percentage of outs a pitcher records on plays that do not involve the defense". Over the long run, DIPP should be an excellent measure for dominant control of the strike zone. It is derived by dividing strikeouts into the sum of strikeouts, walks, hit batters, and home runs: DIPP = K / (K + BB + HB + HR) When I introduced DIPP three weeks ago, I did not yet have a complete list of pitchers and their statistics for the 2003 season. Now that I do, here are the DIPP leaders among qualified starters (162+ innings) and qualified relievers (60+ innings):
PitcherDIPP
C Schilling
M Prior
P Martinez
J Schmidt
M Mussina
R Halladay
J Vazquez
K Brown
A Pettitte
E Loaiza
.795
.778
.766
.765
.765
.756
.742
.725
.723
.719
 
PitcherDIPP
J Smoltz
E Gagne
L Hawkins
M Rivera
B Wagner
G Mota
E Guardado
M Timlin
B Shouse
D Riske
.890
.856
.798
.797
.784
.767
.759
.756
.755
.739
Looking at the names on these two top-ten lists for 2003 (and also seeing the names and stats for those at the bottom of the lists), there is no question in my mind that DIPP is a high-quality measure for of pitchers' effectiveness, especially for relief pitchers, where it does not simply reward a pitcher for his team's offense. I will add a more complete list of SH% and DIPP leaders for the 2003 season shortly, perhaps this weekend. (I did not intend to turn this into a "use my new stat" editorial, but after playing with the numbers over the 2003 season, I have become quite impressed with DIPP's simplicity and utility. Also, while lightheartedly researching the relief-pitching side of things, I came across one of Mike's Baseball Rants that hit on the subject pretty well. I definitely recommend reading that if the subject of this post interests you.)
0 comments

Warning: DOMDocument::load(): php_network_getaddresses: getaddrinfo failed: Name or service not known in /home4/derok6/public_html/chop-n-change/modules/mod_xmlarticles/helper.php on line 19

Warning: DOMDocument::load(http://www.bloguinmedia.com/curation/maintop/articles.xml): failed to open stream: php_network_getaddresses: getaddrinfo failed: Name or service not known in /home4/derok6/public_html/chop-n-change/modules/mod_xmlarticles/helper.php on line 19

You Might Like...