Another interesting study on possible racial bias in officiating...this time studying umpires and pitchers. I haven't read the study, evaluated the quality of its science or read the peer reviews of it (if any). So I won't make a judgment about it based on my beliefs, uninformed suppositions or assumptions about how the study was done or its accuracy. Just thought it was interesting enough to share.

Time magazine article:
http://www.time.com/time/health/article ... 38,00.html

The actual study:
http://www.eco.utexas.edu/faculty/Hamer ... uthors.pdf

I can't wait for the study done on possible racial bias in studies done on racial bias.

1% of pitches thrown?

That is completely statistically insignificant.

_________________
"Play until it hurts, then play until it hurts to not play."
http://soundcloud.com/darkside124 HOF 2013, MM Champion 2014

For a small data sample, 1% would be meaningless. But not for a sample this large. This dataset has 2,120,166 pitches. One standard deviation is 1,456. Two standard deviations is typically considered significant and six (six sigma) is very significant. So while 1% (21,000 pitches) seems insignificant on its face, in this case it represents 14 standard deviations. That is very statistically significant in a dataset of 2.1 million.

Again, I have to disagree. I don't think that you can call this statistically significant. The size and shape of the strike zone varies on a pitch by pitch basis. Also, 1% of pitches, isn't it possible that the umpire actually made an honest mistake? A home plate umpire must make over 300 accurate calls a day between calling balls and strikes, fouls, interpretation of the rule book, ect. An accuracy rate of 99% seems quite exceptional. Racially motivated, this is getting silly.

I am losing patience that we will look for racism EVERYWHERE and find it weather it exists or not.

This story should be a non story.

_________________
"Play until it hurts, then play until it hurts to not play."
http://soundcloud.com/darkside124 HOF 2013, MM Champion 2014

"the lowest percentage were called between a white ump and a black pitcher. " - Time

I disagree with Time's analysis. From Table 2 (page 26) of the actual study the lowest % seems to be black umps and Asian pitchers (30.19%).

_________________
Paws Up

Coast - How do you get a standard deviation for attribute data? I didn't see this number.

_________________
Paws Up

bw...correct me if I'm wrong here. I would calculate it as: square root of the sample size = 1 SD of the difference between two groups being compared.

I don't know whether the variance in all combinations is significant, but the author notes, as you pointed out, that the variation is statistically significant for the Black umpire/Asian pitcher combination.

I guess the obvious question is what did Dontrell and C.C do to piss off umpires??

on second thought...wouldn't there be NO standard deviation? The sample size is every single pitch so there is no staistical error over the 3 year time frame.

_________________
Paws Up

_________________
SideshowBob311: "Sadly enough, I think we're the "intelligent" portion of the sports radio fanbase".
"I make fun of whoever sucks, including me"- Harry 11/30/07

I am saying it is hard to reconsile mathmatical principals with a subjective data set. it would be different if a strike zone were a simple definitive thing, but a strike zone changes on a ump by ump, or batter by batter basis.

We're assuming, in analyzing this data, that the zone is absolute. In real life it is not. Therefore any qualification of derivations on a pitch by pitch basis using the race of the batter or the pitcher as a unique variable is flawed.

_________________
"Play until it hurts, then play until it hurts to not play."
http://soundcloud.com/darkside124 HOF 2013, MM Champion 2014

ROONEY, Pops, you told me never to believe statictics. You taught me that Stats lie. You don't have a stupid son, you have a Ditto of you at age 30.

_________________
"Play until it hurts, then play until it hurts to not play."
http://soundcloud.com/darkside124 HOF 2013, MM Champion 2014

_________________
We all have private ails. The troublemakers are they who need public cures for their private ails.- Eric Hoffer

I think a study like this was done to with the conclusion already determined. Is there possibly a racism umpire who determines close calls based on race ? Possibly, but I think nearly all umps are credible, and the last thing they are thinking about is the race of the players.

Does the recent NBA kafuffle with the ref shake your view of umpire integrity? If it happened in the NBA, could it happen in the MLB?

_________________
"Play until it hurts, then play until it hurts to not play."
http://soundcloud.com/darkside124 HOF 2013, MM Champion 2014

A ref (or player/manager) betting on games can happen in any sport, and the only thing you can really do about it is punish the ref who got caught, and keep up the security checks that they probably were doing already or add new procedures that make sense.

So, while it could happen in any sport, it doesn't really ruin my view of umpire integrity unless these stories start becoming more wide spread. One bad apple doesn't spoil the whole bunch.

I wouldn't use this study to make any broad sweeping generalizations about the integrity of umpires, racism of umpires or the accuracy of balls and strikes being called. I do continue to find it amusing though when people attack science and scientists when the conclusions of a certain piece of rigorous study don't fit their beliefs or assumptions.

Is there ever a situation where "science" has a political agenda and should be more scrutinized?

I don't think I'm "attacking" science here. I think I'm pointing out a legitimate flaw in the scientific process by which the data was collected and analyzed.

_________________
"Play until it hurts, then play until it hurts to not play."
http://soundcloud.com/darkside124 HOF 2013, MM Champion 2014

Understood, Darkside. I wasn't referring to you, actually. You made a sensible point about a technical matter related to the study construction. While I would quibble with it a little, my disagreement is not serious enough to elaborate upon. I was referring more specifically to people like BD and Mike who questioned the motives, intentions or integrity of the scientist. I find those kinds of comments amusing and unfortunate. And BD and Mike...I'm not picking on you here. Many share your views.

To answer your question Darkside, yes, I believe there is plenty of junk science that is done without rigor or peer review that seem to be constructed merely to support the contentions of the author. Some media organizations give them airing, which really serves to devalue all science. (I don't think that's the case with this, by the way, but I haven't read everything or peer reviews of it yet either). If our media can't tell the difference or establish specific parameters to decide about the scientific merit of different studies, I am not surprised that many people don't know the difference either (especially those who don't understand statistics, the scientific method or the peer review process).

Mr. Reason used a variation on the theme that statistics lie and can be twisted to support one's contentions. Of course that's the case with bad science, but I believe that kind of statement or belief can be very destructive to science generally. (No offense intended, Reason. You are just the latest to say it. I just cringe when I hear it) People can throw that kind of statement out about any statistical or mathematical piece of work and in the process, they devalue and degrade all statistics and mathematics and all science..even the most worthy work. In many cases, I doubt the person throwing out that dismissive comment has even read the work, let alone evaluated it for statistical or scientific merit. I think that's sad.

Again, no disrespect meant toward Reason or anyone else here. I just think it's sad that we have become so cynical that many of us can't even trust any science except that which supports our beliefs or contentions. For as much as science has given us in so many ways, it's lamentable that so many people hold it in such low esteem....or don't understand the difference between good science and junk.

Coast, it is actually a quote from Greg "sacks don't matter" Blache. Murph uses it to start Stat of the Week with John Dewan.

Statistical studies have become far to politicized recently. Especially studies of the academic variety. You can't deny that sometimes researchers use data to prove a preconceived theory. Mankind's effect on climate change comes to mind. They look for data to buttress their position while ignoring other analysis that may not fit the template. I am always skeptical of any study until I read through the process and the results. Which I haven't done in this case and have no plans on doing so. My comment was not about this particular study. I was just trolling.

_________________
We all have private ails. The troublemakers are they who need public cures for their private ails.- Eric Hoffer

I haven't read any of the report. Does it use any information from the tracker system MLB established to create a more homogenous strike zone?

It would be interesting to see the deviations from that standard.

Yes, there some data for the Questec or whatever its called vs. non-computer dealie, as I like to call it.

_________________
Paws Up

Sorry Coast, I was just being a smart ass.
I just think studies like these are kinda silly when you think about potential benefits of the results. None really. I don't need a team a scientists to tell me that racism can exist anywhere, anytime & there's already a system in place to audit umpires' pitch calls. Who did the cost-benefit analysis on this study?

The author is funded through an endowed chair at the University of Texas (which usually means all costs associated with him and his work are paid by the endowment). He is a pretty accomplished scientist and author on labor economics issues. I suppose there are thousands of scientists doing work for which we don't understand the value. I usually don't worry much about the cost-benefits associated with how other people do their jobs. I have enough trouble managing that for myself. But if you're interested in that, perhaps you should email him and ask him. His email is right there on his website: hamermes@eco.utexas.edu