Tango on Baseball Archives

Clutch Hitters (January 27, 2004)

I figured a hitter's Linear Weights Runs without considering the context, and then I compared it against the actual game context (inning, score, base,out). I could have done it 2 ways. One is using win probability, and do it PA by PA. The other is to create "classes" of pressure, and use the average pressure level for each class. I did it the second way.

For example, Miguel Tejada


LIclass	 avgLI 	eventPA
0	 0.24 	913
1	 0.74 	842
2	 1.23 	503
3	 1.90 	358
4	 3.50 	120

had 913 PA with little pressure. Those PAs had an average LI of 0.24. He had 120 PAs with high pressure. The average LI of those PAs was 3.5.

Tejada's LWTS runs in LI class "4" were multiplied by 3.5, and those in LI class "0" were multiplied by 0.24. I did this for all the classes of pressure.

Given the pressure, Tejada produced 46 more runs than would have been expected by luck, from 1999-2002. That's Tejada's clutch performance.

Using this process, the spread of clutch performance for all MLB hitters was exactly the same as expected by luck. While I can't say that this means that clutch hitting does not exist, I can say that using this very rigorous process, that I can't identify them with any degree of confidence.

Here then are the 20 best and worst hitters in producing during clutch situations, compared to the overall talent.

Batter Clutch Runs
Giambi, Jason 59
Tejada, Miguel 46
Sexson, Richie 43
Burks, Ellis 43
Walker, Larry 42
Winn, Randy 33
Lee, Carlos 32
Vaughn, Greg 29
Rolen, Scott 29
Palmeiro, Rafae 29
Gonzalez, Alex 28
Long, Terrence 28
Boone, Aaron 28
Relaford, Desi 26
Cruz Jr., Jose 25
Clark, Tony 25
Goodwin, Tom 24
Spiezio, Scott 24
Lawton, Matt 24
Valentin, Jose 23

...

Ordonez, Maggli -21
Catalanotto, Fr -21
Hidalgo, Richar -21
Bichette, Dante -22
Tucker, Michael -23
Lankford, Ray -23
Stewart, Shanno -24
Benard, Marvin -26
Anderson, Garre -26
Fullmer, Brad -26
Green, Shawn -27
Giles, Brian -28
Castilla, Vinny -29
Glanville, Doug -32
Lugo, Julio -34
Perez, Neifi -34
Young, Eric -34
Beltre, Adrian -36
Cirillo, Jeff -39
Easley, Damion -46

--posted by TangoTiger at 02:46 PM EDT

Posted 3:27 p.m., January 27, 2004 (#1) - bob mong
Man, Cirillo has sucked.

Posted 8:05 p.m., January 27, 2004 (#2) - Robert S
Interesting stuff, tango!

Posted 9:30 p.m., January 27, 2004 (#3) - Alan Jordan
I'm sure that everyone else reading this could answer this question, but what is LI?

Posted 10:26 p.m., January 27, 2004 (#4) - tangotiger
Leveraged Index.

It's the amount of "swing win" impact each situation presents. Bottom of 9th, down by 1, men on base, 1 out, the probability distribution for the potential new win % swings about 10 times than a random situation. That's an LI of 10.

Bottom of the third, up by 18, the LI is almost 0.

Posted 10:49 p.m., January 27, 2004 (#5) - MGL
Tango, or anyone else, if you want to find out if you can identify clutch players, using your definition, why not take everyone's clutch/non-clutch ratio or difference or odds ratio (I'm not sure which would be best), and run a regression correlation from one time period to another? If it is zero or near zero, no clutch "ability" - right? Is your method any different or any better?

Posted 9:52 a.m., January 28, 2004 (#6) - Rally Monkey
Garret Anderson, not a clutch hitter?

You learn something new every day.

Posted 10:56 a.m., January 28, 2004 (#7) - FJM
Very interesting indeed. A couple observations.

1) From the data you provided I calculate an average LI for Tejada of 0.936. Since 1.00 is normal, that strikes me as surprisingly low, especially for such a large number of PA's (2,736). But perhaps I'm wrong. Could you post the average LI and number of PA's for each of your Top 20/Bottom 20? Thank you.

2) Wouldn't it be better to rank the players by Clutch Runs per a fixed number of PA's? The distribution appears to be skewed to the right. But that could be caused by the Top 20 getting more opportunities than the Bottom 20.

Posted 11:32 a.m., January 28, 2004 (#8) - tangotiger
I'll try to do that tomorrow.

I do it both ways, by total runs and total runs per 600 PA. I just presented it one way here.

Posted 12:37 p.m., January 28, 2004 (#9) - tangotiger
Among the 119 hitters with at least 2000 PA from 99-02, the LI for the hitters ranged from 1.09 (Phil Nevin) to .91 (Fernando Vina). Tejada was 110th among these hitters at .94.

Posted 7:35 a.m., January 29, 2004 (#10) - studes (homepage)
That's interesting to me; I'd like to see the list of leaders and laggards in batting LI.

Have you ever looked at batting LI before? You could potentially use it to (once again) see if batting order really has any impact (do cleanup hitters have the most LI?), or judge to see which hitters were most "on the spot."