Introducing a New Stat: The Kendrick ERA (kERA)
by David S. Cohen on Jun 18, 2010 1:25 PM EDT
Email
PrintKyle Kendrick did it again last night. As I noted in the introduction to the game thread, he's been maddeningly inconsistent this year. Or, actually, he's been quite consistent, just in two different ways. I wrote:
Kendrick has had a true Jekyll and Hyde season so far. In five of his starts, he's been phenomenal, posting a 0.75 ERA in 36 innings. However, in his other seven starts, he's been absolutely atrocious, posting a 9.76 ERA in 31.3 innings.
Well, last night, Dr. Kyle Jekyll showed up, as he was excellent once again. He gave up 1 run in 7 innings, holding the Yankees to 4 hits and 2 walks. After last night, his bad seven starts remain at a 9.76 ERA, but he now has six good starts for a 0.84 ERA. He's been so good in his good starts, that giving up just 1 run in 7 innings last night actually made his good-start ERA go up!
Kendrick's split personality this year got me wondering about other starters of his quality, so I decided to take a look. For the year, Kendrick has a 4.48 ERA. There's a nice grouping of 20 starters who I consider roughly comparable to him so far this year, with ERAs between 4.21 and 4.71. With ERAs in the mid-4s, these starters are neither world-beaters nor slop-throwers. Going by ERA+, they're just below average, as ERAs in this range are in the 90s for ERA+.
What I decided to do was calculate each starter's kERA, or Kendrick ERA. kERA is the difference between the ERA for the top half of the starter's games versus the bottom half. Kyle Kendrick himself sports a whopping 8.93 kERA. Is anyone besting him at his own game this year?
Turns out, the answer is yes, but just slightly. Here's a chart of these 20 starters ranked by kERA, from highest to lowest:
| SP | ERA | kERA |
| Joe Saunders | 4.70 | 8.97 |
| Kyle Kendrick | 4.48 | 8.93 |
| Joel Pineiro | 4.45 | 8.54 |
| A.J. Burnett | 4.33 | 7.77 |
| Bronson Arroyo | 4.53 | 7.03 |
| Mark Buehrle | 4.71 | 6.79 |
| Nate Robertson | 4.57 | 6.54 |
| Felipe Paulino | 4.50 | 6.11 |
| Dave Bush | 4.67 | 6.05 |
| Dan Haren | 4.71 | 5.79 |
| Johnny Cueto | 4.50 | 5.77 |
| Mitch Talbot | 4.21 | 5.75 |
| Scott Baker | 4.41 | 5.30 |
| John Lackey | 4.53 | 5.26 |
| Chad Billingsley | 4.34 | 5.10 |
| Rodrigo Lopez | 4.70 | 4.94 |
| Gio Gonzalez | 4.21 | 4.79 |
| James Shields | 4.55 | 4.43 |
| Chris Volstad | 4.25 | 4.33 |
| Kenshin Kawakami | 4.42 | 4.18 |
Joe Saunders is the leader here, posting an 8.97 kERA. In Saunders' 7 top starts, he's posted a 1.22 ERA over 51.67 innings. In his 7 horrible starts, he's posted a whopping 10.19 ERA in 32.67 innings. With that kind of spread, he's our leader in kERA.
On the other end of the spectrum is Kenshin Kawakami. His 4.18 kERA leads this group, showing that he's been pretty consistent on his way to his unlucky record of 0-9. In Kawakami's six best starts, he has a 2.27 ERA. In his 7 worst, he's got a 6.45 ERA. He's rarely dominant, but also rarely atrocious.
Without going further into analysis, I'll pose a question for discussion. For starters in this ERA range, which would you rather have: A starter with a high kERA, who is either dominant or horrendous, but never in between; or a more consistent starter, one with a lower kERA, but who never truly sparkles?
47 comments
|
3 recs |
Do you like this story?
Comments
I’d be interested to see how the W/L records match up (team win/loss, not pitcher) to see if one style is more likely to lead to regular season wins.
That’s what I was thinking. I’m not sure whether boom-or-bust is better than consistently mediocre in terms of the team getting wins.
Honor is no substitute for victory.
Top 5 highest kERA (7.00+):
.
39 team wins
31 team losses
.
.
Lowest 5 kERA (5.00-)
.
26 team wins
42 team losses
.
My gut was to take the consistent pitcher to get more wins, but it appears to be a significant advantage to take the more schizophrenic types. I was not expecting that.
Interesting. Thanks for posting that. I’ll have to look into it more closely at some other point in time.
by David S. Cohen on Jun 18, 2010 3:00 PM EDT up reply actions
I would think that a pitcher who is very good overall would benefit from a low kERA, but a pitcher who isn’t very good overall would benefit from a high kERA.
That makes intuitive sense. It would seem to be an incremental thing. The higher your ERA, the more you want your kERA to be high as well. The lower your ERA, the smaller you want your kERA to be.
by David S. Cohen on Jun 18, 2010 3:14 PM EDT up reply actions
as well as better job security, as we have seen.
by Wet Luzinski on Jun 18, 2010 5:18 PM EDT up reply actions
with the phils of late…might be better to take the bipolar guy. If he’s only giving up 1-2 per start you have a shot, but if he’s giving up >3 you’re done for.
It all depends on the makeup of your team. If you’ve got a decent offense, you can get wins from the guy who gives up 2, 6, 2, 8, 0, 2. You could still expect to be in all those except the 8 maybe.
If your offense is putrid you might lose all of those, except the gooseegg.
It's obvious
that you would want the inconsistent pitcher. Not many offenses will muster 5-7 runs a game to balance out the 4.2+ ERA. You will undoubtedly lose a decent number of those games. However, with the highly inconsistent pitcher, with an offense like the phils (the last few years, not the last few weeks), you will win nearly every one of the good starts, and also have a shot at some of the not-so-good starts
"My grandmom's favorite grandson, ask my grandmom" --Rone
by layout ultimate on Jun 18, 2010 9:29 PM EDT up reply actions
Of couse think about it in
Lets say in 900 innings you’re going to give up 50 runs
This would give you an ERA of 5.00 if you had the chance to give them all up in the first game and go you’re next 9 games without surrendering a run wouldn’t you take it?
by sowhatifitisasportste on Jun 18, 2010 6:21 PM EDT up reply actions
my bad ment to be 90.0 innings lol
by sowhatifitisasportste on Jun 18, 2010 6:36 PM EDT up reply actions
i think it would take more than 10 games to throw 900 innings.
your logic is correct, (minus that order f magnitude thingy) but not realistic because no pitcher would be allowed to give up 50 runs in a game.
It’s the most extreme but it makes a point if you’re gona take a loss don’t loose by 1 run go down in a blaze and come back with your A game
by sowhatifitisasportste on Jun 18, 2010 6:41 PM EDT up reply actions
7.79
He has a 1.40 ERA in his six best starts and a 9.19 ERA in his seven worst.
by David S. Cohen on Jun 18, 2010 2:23 PM EDT up reply actions
Checking if I'm calculating this right...
Halladay has a kERA of 4.12 – 0.47 in his 7 best starts, 4.59 in his 7 worst starts?
Honor is no substitute for victory.
Looks right to me
I wonder if a 4ish spread is the norm, regardless of how good/bad a pitcher is.
by David S. Cohen on Jun 18, 2010 2:59 PM EDT up reply actions
It might be the norm, but it's not a definite
A.J. Burnett – kERA of 7.77 – 0.92 in his 7 best starts, 8.69 in his 7 worst.
Fausto Carmona – kERA of 3.46 – 1.79 in his 7 best starts, 5.25 in his 6 worst
Honor is no substitute for victory.
FLAKE
Baseball Prospectus has a stat called FLAKE, that measures consistency of pitchers. There’s a story about it here – http://www.baseballprospectus.com/article.php?articleid=8579
Does Ubaldo even qualify?
"They say that nobody is perfect. Then they tell you practice makes perfect. I wish they'd make up their minds." - Wilt Chamberlain
Sure
He’s got a pretty miniscule kERA, though – 2.13. His good-start ERA is 0.17 (1 earned run over 7 starts and 54.33 innings!!!). HIs “bad”-start ERA is 2.30. That’s kind of insane.
Yet Amaro smugly refutes it—as does Utley, apparently.
"Tortorella’s got it all wrong ... Gaborik shouldn’t be messing with our skilled player." -Peter Luukko
confusing
Join the NLL community at http://www.nationallacrosseleagueblog.blogspot.com/
Talk about Boston sports at http://www.bestbostonsports.com/
by bestbostonsports on Jun 18, 2010 6:15 PM EDT reply actions
I'll go for Jekyll and Hyde
Considering Kawakami is 0-6, I would much prefer Kendrick and his Jeckyll and Hyde act. At least we get a few quality wins, including a dominant effort against those dreaded Stankees. Imagine Kendrick coming in to pitch the World Series with Yankees… would that be a heart stopper or what? LOL!
by Fillyfaninforeignland on Jun 18, 2010 10:28 PM EDT reply actions
A logical extension of kERA would be to look at the distribution of all pitchers' starts.
How many starts fall in the 0.00 – 0.99 ERA range? the 1.00 – 1.99 ERA range? … the 9.00 – 9.99 ERA range? Etc.
ERA distributions would give a clearer picture of effectiveness than simply overall ERA.
This Article Was Sort of What I was getting at earlier
In a number of threads prior to last Ftiday night’s game with the Twins, I brought up the idea that it may be better for the Ohillies to keep Kyle Kendrick instead of Joe Blanton in the rotation because while Blanton was consistently giving up five runs, at least there was a chance with Kendrick that he would only give up a couple of runs, allowing the Phillies to win without having to score many runs (vital considering the offensive malfeasance this team has exposed us to in the last month). Someone else posted that teams were more likely to win games with pitchers of Kyle’s ERA range if they were erratic than if they are consistently giving up 4-5 runs a game, as often 4 or 5 runs is enough to have given up the loss. I think that probably until you get to pitchers in the sub 3- 3.5 ERA category, you are better with the ups and downs— pitchers with high kERAs are more likely to give you games where the score is held down enough for the team to win. That being said, I would think that for elite pitchers, you may be better off with taking their average, as it is a manageable number of runs.
by dannijd on Jun 21, 2010 11:42 PM EDT via mobile reply actions
ESPN Monday Night Baseball
Looks like this idea is catching on nationally. In tonight’s game the announcers were making a big point about how AJ Burnett’s numbers were widely divergent between his good games (first 8 if I remember right) and his bad games (the ones since then). Right on cue, Burnett got shelled in the first inning.
by phillyinportland on Jun 22, 2010 3:14 AM EDT reply actions
The fact that Burnett has regressed so much from the beginning of the season leads me to wonder if he is doing something differently, tipping his pitches, or if something else is going on… Unlike Kendrick, who has had good days and bad ones throughout the season (although none so disastrous as his first two), Burnett came flying out of the gate, doing well, and now can not seem to buy himself luck. This leads me to wonder if he is having bad luck (BABIP, fly ball/ homerun rate, etc).
True enough
What I thought was interesting was the way Burnett’s differences were described much the way we’ve discussed Kendrick. In fact I just noticed that Burnett is number four on the kERA list. I imagine that a split like Burnett’s, where the change came part way into the season, rather than like Kendrick’s, where the change comes game to game, means that he isn’t just erratic but probably hurting or tipping pitches. As long as he doesn’t get replaced with Cliff Lee I’m glad to see him struggle.
by phillyinportland on Jun 23, 2010 2:25 AM EDT up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
Next Game
In Association With
Big League Stew - MLB - Yahoo! Sports
- Marlins agree to unretire No. 5 for Logan Morrison, who will honor his father’s favorite player
- Weatherman Derek Holland does Dallas TV
- Ask Alex: Will Julio Teheran or Matt Moore win rookie of the year?
- Ku Klux Klan, umpires and bowling: Why the minor leagues tossed out Jim Evans academy
- The Cardinals Hall of Fame can now be found at Bill Clinton’s library in Arkansas (Huh?)
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
