Soria and Leverage Index
by Jeff Zimmerman on Apr 29, 2010 1:28 PM EDT
in Kansas City Royals News
Email
Print
I planned on doing this piece before that debacle Tuesday night as I usually write my pieces while listening to the game the night before they run. I am sure that Hillman's inability to bring in Soria will be discussed to death, as it should be. That game was sickening.
If the rest of you are like me, the bullpen exists of Soria and a rotating pile of dung. Not one of the relievers is considered a real setup man. With this situation, Soria needs to be used whenever the situation is most critical late in a game.
Leverage Index is a stat that is used to show how important the situation is at that time. The Leverage Index is higher late in a close game with players on base and lower early in a game or during a blowout. Recently a reporter asked Hillman why he didn't use Soria in a high leverage situation and here is his response:
"There's a thought there but, No. 1, it's a very unusual time for Joakim Soria to pitch in a ballgame. No. 2, you've still got those same bats coming up in the ninth in a higher-leverage situation - because it is the ninth, even if there are no runners on base."
The reaction on this exact quote was already discussed, but I want to see if Soria is being used in the highest leverage situations. Fangraphs tracks a pitcher's average LI and here is the season values for the relievers
Explanations:
pLI: A player's average LI for all game events.
phLI: A batter's average LI in only pinch hit events.
gmLI: A pitcher's average LI when he enters the game.
| Name | pLI | gmLI | exLI |
| Bruce Chen | 3.06 | 2.65 | 5.73 |
| Joakim Soria | 2.17 | 2.23 | 2.38 |
| Josh Rupe | 2.01 | 1.78 | 1.98 |
| John Parrish | 1.58 | 1.87 | 1.54 |
| Dustin Hughes | 1.15 | 1.32 | 1.82 |
| Robinson Tejeda | 1.14 | 1.72 | 1.82 |
| Brad Thompson | 1.00 | 1.45 | 1.07 |
| Kyle Farnsworth | 0.91 | 0.94 | 1.03 |
| Juan Cruz | 0.77 | 1.45 | 0.85 |
| Luis Mendoza | 0.73 | 1.24 | 0.25 |
| Roman Colon | 0.70 | 0.69 | 1.70 |
Well, besides Bruce Chen's game a couple of days go skewing his data, Soria, on average, has be used the most in high leverage situations. I graphed all the games and the pLI for all the relievers, but Chen and Rupe because of the recent call ups (This is the first time I have used this chart and is a work in progress).
Not in a single game this season Soria has been used in the highest leverage situation. In three of the games, 2 pitchers were used in higher leverage situations instead of Soria.
Using Soria only in the 9th is the safe bet for Hillman because he can show that Soria is used in High Leverage situations. He needs to start looking into situations where Soria can be used when the game is more on the line.
27 comments
|
7 recs |
Do you like this story?
Comments
I think you meant to put in
exLI: A pitcher’s average LI when exiting the game.
instead of
phLI: A batter’s average LI in only pinch hit events.
P.S. isn’t gmLi a dwarf?
P.P.S. if i’m reading that chart right, not only is Joakim appearing in lower leverage situations, he’s appearing in fewer games period
P.P.P.S. rec’d
batter nine you sucky
Hillman seems like a manager that can be influenced by the manager
I mean one of his quotes earlier this year was about how he put DJ in leadoff because media types kept crowing about it. Well, the drumbeat is starting to beat about using Soria in non-conventional ways. He will cave, just keep the drumbeat going!
Relive Royals History at royalsretro.blogspot.com
the pLI could be skewed for the relievers
since they always let runners on base in the 7th or 8th, increasing their leverage. So by letting the runners on, they are bumping up their leverage stat.
I would be interested in seeing the gmLI graphed, to show if they are bringing in other relievers in higher leverage situations other than Soria.
I might actually create both from now on to see if there is any discrepancies
I like the pLI in that is also show the reliever stayed in as the LI increases. Here is another one I might use instead. Discussing it over at the Book Blog.
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Apr 28, 2010 11:10 AM EDT up reply actions
Maybe go with innLI, then?
The worse the pitcher, the higher his average pLI (given the situation he started with) because he’ll allow more baserunners.
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Apr 28, 2010 12:19 PM EDT up reply actions
So if the exLI is worse than the gmLI,
the pitcher didn’t do his job?
Glad I came, just wish I hadn't stayed so long.
People ask me what I do in winter when there’s no baseball. I’ll tell you what I do. I stare out the window and wait for spring.
Or could have done the job. They can in with the bases loaded, down by 1 and got everyone out.
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Apr 28, 2010 11:11 AM EDT up reply actions
Just awesome.
Jeff, it’s not perfect, but can you average out the highest gmLI from each game, the second highest, etc.? This would be a rough approximation of what Soria’s gmLI could be if used optimally (and everyone else). Then maybe plug the actual situation and the hypothetical situation in a chaining analysis to see how much this costs the Royals…
Beyond the Boxscore Not a member? Sign up.
Or give me the raw data ;)
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Apr 28, 2010 10:31 AM EDT up reply actions
Didn't quite work:
1.83
1.29
0.96
0.98
0.66
Obviously, Soria’s not going to pitch in the highest LI situation of every game. You need a way to pick the highest 15% (or whatever) situations for Soria, then the next 15% for the main setup guy, etc.
Actuall, I’ll give it a whirl… there have been 63 relief appearances this year. Giving everyone nine (from different games, obviously) splits it up nicely between seven bullpen spots:
green: 2.87
orange: 2.25
purple: 1.75
blue: 1.28
pink: 0.71
grey: 0.28
white: 0.06
(It probably deserves more than a footnote, but to hold any manager accountable for not reaching the hindsight-is-20/20 perfect usage pattern is pretty darn unfair.)
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Apr 28, 2010 12:42 PM EDT up reply actions 2 recs
also the back-to-back games with high leverage and/or high pitch counts for the 'pen
cause some mandatory non-optimal usage
by SagehenMacGyver47 on Apr 28, 2010 1:26 PM EDT up reply actions
Definitely.
For example, the table above shows Soria pitching four out of the first five days. Probably wouldn’t/shouldn’t happen.
Beyond the Boxscore Not a member? Sign up.
question:
“Well, besides Bruce Chen’s game a couple of days go skewing his data, Soria, on average, has be used the most in high leverage situations.”
How much of this can be because we generally cannot GET to Soria, so then he has less chances to even pitch in a leveraged situation? Obviously, if we were winning more, he’d be used more, but when he sits on a shelf because we’re down 1 he doesn’t even get the option to pitch.
Obviously, you are not a golfer.
That is a little tougher to figure out
As a general rule, I went and looked in The Book and anytime the LI is at 1.5 late in the game, the ace reliever should be used.
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Apr 28, 2010 12:42 PM EDT up reply actions
and that's with an average bullpen (I believe) and "closer"
the delta between the Royals’ bullpen ability and Soria’s ability is much greater than average, which would call for using Soria more often.
by SagehenMacGyver47 on Apr 28, 2010 1:31 PM EDT up reply actions
Any chance
Rupe is going to be solid and just had an off night? At this stage of things, it’s impossibleto distinguish one of those off nights with terminal suckage.
Obviously Hillman is using Soria incorrectly.
However, I think there is still a good question about whether with the Royals awful bullpen, it is better to simply insert Soria for the highest leverage situations for the Eck-type inning of work or for the rest of the game a la Quiz. The one inning closer/stopper averages about 60-80 inning a year and diminishes their WAR potential. Quiz on the other hand in his typical years averaged 120-140 inning. I think that could almost double Soria’s WAR potential if his rate stats stay similar.
The trade off seems to be more games and fewer innings versus fewer games and more innings. But I think the games would not diminish as much as the innings would increase.
Finally the long save route would also be dependent on Soria’s rate stats staying about the same even facing people the second time through the order.
Leverage Schmeverage
Here is the problem with putting Soria in at other times: do you want to see any other member of the bullpen face Mauer, Thome, and Morneau in a single inning. Not me. These guys are just an example of what they could face. If these guys have a chance of coming to bat in the 9th, only Soria should be pitching in the 9th. If these guys come up in the 8th, then I agree, Soria can pitch then, as long as there are 6 batters before they come up again. Because I am sure what is left in the bullpen will walk every hitter in order to face Mauer.
This is sort of like trying to get Grienke a win. What is everybody’s solution to getting Grienke a W. Pitch 8 innings, then give it to Soria. If Soria comes in at inning 8, and somebody blows it in the 9th, the solution will be to have Soria pitch 2 innings (which will not happen). So instead of it possibly being blown in the 8th, it is now possibly blown in the 9th. What is the difference. It is just a little more heartbreaking when it comes in the 9th.
Eventually, the solution for everybody becomes, only let Grienke and Soria pitch, because the rest of the bullpen will blow it. Does it really matter what what the leverage situation is for the other pitchers when they screw it up regardless?
The real solution is to find any reliever that knows what home plate looks like and throws a freaking strike, more than once per plate appearance.
I don't think you understand what "leverage" means.
Glad I came, just wish I hadn't stayed so long.
People ask me what I do in winter when there’s no baseball. I’ll tell you what I do. I stare out the window and wait for spring.
by Warden11 on Apr 29, 2010 10:03 AM EDT up reply actions 1 recs
Not sure how to respond
Let me think about where to start
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Apr 29, 2010 11:04 AM EDT up reply actions
eakers - Finding good relievers would be a nice change
but given what the Royals have, using leverage to guide their decisions is a better process than the current strategy of just saying “Soria pitches the 9th. Everybody else can go first”.
Even if Mauer, etc might bat again in the game, you shouldn’t pitch a lesser reliever in a risky situation just in case some hypothetical riskier situation might develop later in the game.
by SagehenMacGyver47 on Apr 29, 2010 2:14 PM EDT up reply actions
The classic situation: Use your stud now, or wait for a later hypothetical?
If you are down 3 games to 2 in the World Series, do you pitch your stud or throw him in Game 7? Considering your ace may not pitch at all because there will be no game 7, you throw the ace in game 6.
_
I see no parallels between closer usage and that situation at all.
and World Series?
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Apr 29, 2010 9:40 PM EDT up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
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.
