wOBA Against: What If We Ignore Luck?
We've got a million different statistics for pitching. Some, like ERA, stay on the simple side, and measure one aspect of pitching (allowing runs) well. Others, like FIP, try to go deeper and figure out how well a pitcher actually pitched, and what we can expect in the future. Statistics like FIP purposefully ignore certain aspects of pitching to try to focus on the factors that are predominantly skill-based, rather than luck-based. But what if we tried to go the other way? What about a stat that is purely about measuring past performance, with no regard to luck, skill, or ability? What about a stat that doesn't care if a pitcher's hit rate on ground balls is unsustainable in the future? Luckily, we've got just the thing.
Well, sort of. What we can do is repurpose an offensive stat, wOBA, and use it for pitchers. A quick refresher—wOBA takes offensive events like walks, singles, doubles, triples, and home runs, and weights them with numeric values relative to their actual worth. Sure, a bloop single is sometimes just bad luck, but let's throw all of that out of the window for now and focus on the actual results. For all Oakland pitchers with more than 100 batters faced, here are their wOBAs against.
Andrew Bailey's wOBA against is .245! He's essentially turned the batters he's faced into a lineup full of Everth Cabreras. Or Jack Wilsons. Even Yuniesky Betancourt has never put up numbers that low in any season. In fact, we've got four pitchers who have turned their opposition into a replacement level scrub: Bailey, Anderson, Cahill, and Breslow. And three of those four are likely to be building blocks we'll see for a long time.
The A's attempt to come off of the floor swinging as they face the Red Sox at 7:05 PM. Dallas Braden is scheduled to make his first start since coming back from the Disabled List, and he opposes knuckleballer Tim Wakefield.
108 comments
|
0 recs |
Do you like this story?
Comments
I think it should be
wOBAA, just for fun.
Simple, yet effective comparisons. Good stuff.
"Do I talk to myself? No, I just remind myself of what I'm trying to do. You know, I never answer myself so how can I be talking to myself?" - Rickey
Athletics Nation - WE'RE ALL GONNA MRIIIIIIIIIIIII!!!!!
So we should trade Breslow as soon as possible before he reverts to being worse than Blevins?
Actually no, because his tRA is better than Blevins, and for that matter Bailey and everyone else in the pen.
Don't be such Ynoa et al. -- oblique
Breslow has pretty much always been good outside of about 14 innings with Minnesota last year
www.zekeishungry.com
by thejd44 on Jul 20, 2010 6:58 PM PDT up reply actions
He hasn't really thrown enough strikes
But he’s been nails lately, so woot woot molecular biologist!
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
Pretty pretty graphs.
The funny thing about baseball is that people will believe what they want to believe. -Joe Posnanski 8/29/09
So I guess Mazzarro is living on borrowed time?
The A's are a fairly quiet team, and then there's Ben Sheets. Sheets, as a kid, must have been thrown out of every library in Baton Rouge. ~ Scott Ostler
Or he's just young and hasn't matured yet?
by LoneStranger on Jul 20, 2010 10:21 AM PDT up reply actions
he was young--he needed the money.
sock puppets have never successfully defended castles, except when working with squirrels, which would never happen because squirrels know better than to trust sock puppets, except when their nipples have magical powers. -nm
by Leopold Bloom on Jul 20, 2010 11:06 AM PDT up reply actions
well he wanted it anyway.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
Mazzaro is fine...
As I understand it, wOBA is set up to have about a 0.345 mean, so that the cognoscenti can easily interpret it. (It’s on the same scale as OBP, for which they apparently have a better-developed intuitive feel than they do for say, BA.) If that’s right, Vin isn’t as amazing as his MAC buddies, but still solid. At least, it makes me feel good to think so! :)
by BaltimoreAN on Jul 20, 2010 11:31 AM PDT up reply actions
Oops
This article says the American League average is 0.327. He’s still above average, which we’ll take from Vin especially in his age 23 year.
by BaltimoreAN on Jul 20, 2010 11:36 AM PDT up reply actions
i feel better then
its always amazing to me that these stats sometimes say one thing and my eyes see something else.
The A's are a fairly quiet team, and then there's Ben Sheets. Sheets, as a kid, must have been thrown out of every library in Baton Rouge. ~ Scott Ostler
Average is usually .330 or so (although this year this numbers are down for whatever reason).
I still don’t like the decision to artificially increase the scale of wOBA to a OBP scale, as the original scale basically sat at batting average scale—much more convenient for most people. But we can’t change it now, I suppose.
She lives by the wall, and waits by the door.
She walks in the sun, to me.
Interesting take but its not ignoring luck as the title says.
rather its embracing luck. That bullshit inside the park homerun that was misplayed 27 times but never was given an error counts the same as the McGwire blast into the third deck of the Skydome.
Also, Im going to take umbrage with the term actually as you used it. Studies have shown that pitchers have very little control of what happens once a ball gets put into play. Therefore saying that their wOBAA is their actual performance seems to me to be a misnomer. I suggest that you call it the outcome of their performance.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 9:47 AM PDT reply actions
There are no "true outcomes"
Balls and strikes are at the mercy of the whims of umpires. Home runs are usually hit on mistake pitches. Both of these rely on “luck.”
Dan is attempting to smooth away luck as much as he can.
I don't think he's trying to remove the luck at all. He's just taking the results with no regard to how they came about.
For example, W-L. Seeing how many wins a pitcher has doesn’t tell you if they were 1-0 games or 11-10 games. It’s still something people look at.
by LoneStranger on Jul 20, 2010 10:32 AM PDT up reply actions
True, the title is misleading. In the article dan says "with no regard to luck, skill, or ability"
which I translated as “we aren’t trying to isolate or remove luck, skill, ability or finger sniffing.” We’re merely looking at past results and assigning a number to them. Notice there is no analysis of why a pitcher has the number they have. We’re not predicting the future here.
by LoneStranger on Jul 20, 2010 10:30 AM PDT up reply actions
True, but wouldn't pitchers with a high GB% then have a lower wOBAA?
Groundballs are more likely to turn into outs, IIRC. So, it means something, just not something prognostic.
"Do I talk to myself? No, I just remind myself of what I'm trying to do. You know, I never answer myself so how can I be talking to myself?" - Rickey
Athletics Nation - WE'RE ALL GONNA MRIIIIIIIIIIIII!!!!!
by cuppingmaster on Jul 20, 2010 10:33 AM PDT up reply actions
Actually GB are more likely to be hits than FB
But if a GB is a hit, it is usually a single. When flyballs fall for hits, you get a double and sometimes a triple. That difference is what shows in wOBAA against the other pitcher stats, since it applies run values to every outcome.
On teams other than the A's
You can also get a HR
"There's never enough time to do all the nothing you want" -Bill Watterson
You can??????????
I must look into this.
Sincerely,
An A’s Fan
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
Ah, you may be right on the wording.
Perhaps “ignore” was the wrong word to use.
But I disagree on the second point. This is their actual performance, whether they have control over it or not.
She lives by the wall, and waits by the door.
She walks in the sun, to me.
If you distinguish "performance" from "outcome" then this is their actual outcome
I think that’s the point DFA is making.
"There's never enough time to do all the nothing you want" -Bill Watterson
I'm going to start calling DFA
Prof. Umbrage.
by LoneStranger on Jul 20, 2010 10:50 AM PDT up reply actions
Ooooo!
Can you call me Prof. Cucumber?!
sock puppets have never successfully defended castles, except when working with squirrels, which would never happen because squirrels know better than to trust sock puppets, except when their nipples have magical powers. -nm
by Leopold Bloom on Jul 20, 2010 11:08 AM PDT up reply actions
Why take umbrage when you can take infield?
You don’t see Ellis preparing for a game by taking umbrage.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
its actually a really important sabermetric point
what core set of events are actually related to performance enough to credit the pitcher and how should those events be measured as an effect on the team compared to the outcome of their performance.
This is the point that I think dan misses.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 11:41 AM PDT up reply actions
and I think you're missing the point that in this instance, we don't really care.
by LoneStranger on Jul 20, 2010 1:13 PM PDT up reply actions
well if you really don't care you probably shouldn't even look at baseball statistics beyond the final score
since what im talking about is really the whole point.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 3:10 PM PDT up reply actions
No, we don't care about what happened to create the singles, doubles, etc. and who is assigned the 'blame' for them.
We’re just want to look at some stuff that happened in the past, assign a number to it, and compare it to other players. You’re over-analyzing the whole thing.
by LoneStranger on Jul 20, 2010 3:17 PM PDT up reply actions
no because using Dan's statistic to do what your saying is a shitty way of looking at it.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 3:25 PM PDT up reply actions
youre
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 3:25 PM PDT up reply actions
i dont believe in appostrophes
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 3:48 PM PDT up reply actions
I think what LS might be saying is lighten up frances
I didn’t read it as Dan claiming that it predicted anything.
"The ego, the super-ego, and the Ed" - dannycakes
im saying that it isn't even discriptive.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 3:47 PM PDT up reply actions
Why is trying to turn things around and look at them from a different view so 'shitty?'
There doesn’t have to be a predictive value in everything. We’re looking at the past here. Maybe it doesn’t tell you what you want to know, but this is dan’s article, not yours, so I think he knows what he wants to show better than you do.
If you don’t like wOBAA, then don’t use it. Come up with something yourself instead of just complaining “you’re doing it wrong” when you don’t agree with someone.
by LoneStranger on Jul 20, 2010 3:37 PM PDT up reply actions
because it neither has prescriptive nor descriptive value
why would you context neutral the past if the player has a huge role in context if youre trying to be descriptive?
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 3:46 PM PDT up reply actions
ok, that is a valid critique
what you were saying was a sematic debate between performance and outcome. this is smoething new.
"The ego, the super-ego, and the Ed" - dannycakes
no this is a part of the debate
is the results in context a part of the performance is a part of the question. I would say context nuetral pitching performances for starters is a really terrible way of judging “actual performance” pretty much however you describe it but especially how I do.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 4:06 PM PDT up reply actions
there is no debate
there is no right or wrong here. I wish you would stop tying to frame it that way. Its what makes you otherwise fantastic research and analysis tiresome.
"The ego, the super-ego, and the Ed" - dannycakes
yes there is a debate
its a important debate we don’t have a right answer yet that doesn’t mean there won’t be. Unfortunately, this measure does a worse job than the statistics that we currently have.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 8:09 PM PDT up reply actions
to be a little more clear
there is no debate in this thread. there is a debate overall.
ANy debate about it here is at best a side note, and at worst a strawman
"The ego, the super-ego, and the Ed" - dannycakes
I don't think it was meant to be a granular stat. It's more of a 20,000 ft view.
You can compare it to other pitchers. You can say that one pitcher looks better than the other. You can say that one guy was more likely in the past to have given up a double or a HR instead of a single.
It doesn’t make it worthless, no matter how much data you think it won’t tell you.
by LoneStranger on Jul 20, 2010 4:17 PM PDT up reply actions
yeah it is worthless
because it doesn’t tell you that one guy was more likely to give up a double than a home run though
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 4:30 PM PDT up reply actions
Cedrick Bowers was always more likely
to give up a homerun than anything.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
If im catching and the runner between stealling second gets hit by lightning right before he was slides in safely so i throw him out
should I get credit for the CS?
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 11:43 AM PDT up reply actions
Does it happen often enough to be statistically significant?
"PECOTA can pretty much kiss my ass."-Nico
yes you should get credit
but that doesn’t seem to be the point here. I think its a semantic argument, between the actual outcomes and “performance.”
"The ego, the super-ego, and the Ed" - dannycakes
it really isn't semantic at all if you look beyond the surface
i mean you could go even farther in a UZR like method. How many pitches of that variety with that velocity and break get hit. What is the average outcome of those hits. Why don’t we categorize pitcher performance like that?
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 12:15 PM PDT up reply actions
I don't see the problem with saying,
“70 times a runner has attempted to steal against Egbert Goatbody. 30 times the runner has been out.” You may need to look deeper to find out whether 28 runners tripped halfway to 2B or how well that team’s pitchers hold baserunners or how you should view a sample of 70 that isn’t a sample of 700 or of 7, but you have data about “outcome” and that’s worth something — and not more and not less.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
Wasn't he a 1970's-era
designated pinch-runner and sometimes porn star?
Do you have control of player names in The Guba?
Cause if you can, you’ve got some work to do.
by LoneStranger on Jul 20, 2010 3:39 PM PDT up reply actions
im not disagreeing with calling it an outcome, im disagreeing with calling it "actual performance"
you know if you read what im posting you might actually see that.
The question is if we are trying to properly evaluate players based on their contributions to the team, how should we judge them? Its similar to the GM who makes a trade of a mediocre closer with lots of saves for a ten top 20 prospects. If they get hit by a bus in the parking lot of the Coli how do you credit the GM’s performance? The outcome was bad but the performance was good.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 12:33 PM PDT up reply actions
so, semantics.
You agree that this is not a conglomertation of data that will lead anyone to predict what the likely hits and outs the pitcher will obtain in his/her next outing.
"The ego, the super-ego, and the Ed" - dannycakes
Not really
I mean to a certain extent the data is worthless. The whole reason that you use the wOBA weighting is that it makes the outcomes context neutral for batters, so if you want to embrace luck, of which context is certainly a part of for pitchers it really makes no sense to context neutralize. Furthermore pitchers have a far more direct outcome on the context than batters do as in most cases they were directly involved in creating the bases juiced no out situation that they find themselves in. If you are not trying to isolate controllable performance you should just look at R, because at the end of the day thats the actual outcome and any measure which had a purpose other than to determine actual performance by limiting factors to controllable performance for pitcher is going to half ass things in a really useless way.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 12:59 PM PDT up reply actions 1 recs
Is there anyway to look up the FIP and wOBA
for Catfish Hunter, Ken Holzman, Dave Stewart, Bob Welch, Mike Moore etc….
The A's are a fairly quiet team, and then there's Ben Sheets. Sheets, as a kid, must have been thrown out of every library in Baton Rouge. ~ Scott Ostler
FIP, yes.
Fangraphs has historical stats here (Catfish, for example). But for wOBA against, there’s no place to find that as far as I know. I had to calculate the wOBA against myself.
She lives by the wall, and waits by the door.
She walks in the sun, to me.
And what is the wOBA against yourself?
Having seen your fastball, I’d say it’s probably at least .430.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
itd would be interesting to see the park adjusted numbers.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 20, 2010 11:43 AM PDT up reply actions
I was thinking to completely figure out their performance
you would need to have some sort of combination of FIP and wOBA Against.
Ah, you mean
wFOIBPA.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
Huh. Hardball Times ran a very similar article this morning.
http://www.hardballtimes.com/main/blog_article/pitchers-woba-against-2010/
They have something I don’t—leaderboards for the whole league. Looks like Cahill is third in the league (Cliff Lee has a .256 and Jon Lester has a .273).
She lives by the wall, and waits by the door.
She walks in the sun, to me.
And a Google Spreadsheet of all pitchers.
http://spreadsheets.google.com/pub?key=0Aps4HA2pVu7GdF9QRkFkLWszTWNjczhQRDVLSURQS2c
She lives by the wall, and waits by the door.
She walks in the sun, to me.
I'm working on a pretty cool post right now, tentatively titled:
Slugging Percentage: What If We Ignore The First Digit?
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
And the answer you came up with was 'Chris Carter'?
Innnnnteresting
I suspect that you think tilting at windmills means something other than what it does.
The ninth fastest thirty year old in San Francisco
If you calculated these yourself...
What weights did you use? The classic ones from the Book or others? What is the league average using your weights and not, say, fangraphs?
"Loyal? I'm the most loyal player money can buy." - Don Sutton
The classic The Book ones. Which I believe is what Fangraphs uses.
She lives by the wall, and waits by the door.
She walks in the sun, to me.
Nope
Fangraphs uses one that includes SB and CS and they adjust the weights yearly or over a couple of years. If you’re using the ones from The Book it’s from a period of like five years from 8-9 years ago. Overall, it shouldn’t change much but I think league average for the classic weights will be around .335 while the Fangraphs average will be a shade under .330.
"Loyal? I'm the most loyal player money can buy." - Don Sutton
Ooooh.
I didn’t know that they adjusted the weights. That actually makes a lot of sense, and it’s nice to hear that they do that.
She lives by the wall, and waits by the door.
She walks in the sun, to me.
You always say that. Why don't you add something to the conversation?
I'm here to talk about the past.
And why don't I ever reply to myself?
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
Yeah, really.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
what a meaningless statistic
it establishes a staff pecking order we already see and know. You mean our All Star closer is our best pitcher? GET OUT!
"The Few, The Proud, The A's Fan" - UncleLeo
It's ok. incarnate has a history of being kinda hostile toward you.
We know to just ignore him.
by LoneStranger on Jul 20, 2010 2:12 PM PDT up reply actions
Heh.
Not worried, trust me.
She lives by the wall, and waits by the door.
She walks in the sun, to me.
there there, dannycakes.
The funny thing about baseball is that people will believe what they want to believe. -Joe Posnanski 8/29/09
All right,
way to really contribute to the discussion and add something new. I will definitely be looking for more stuff from you in the future!
We need someone
to pay attention to what is going on… stats are what make up baseball in the end. And no one is looking at ours on a day-to-day basis. Nights like last night proved that.
Optimistic...
The best GMs are the ones who can take data from both the scout side of things and the numbers side of things.
For all intensive purposes, we’re arm-chair GMing every damn day.
by LoneStranger on Jul 20, 2010 4:19 PM PDT up reply actions
Well our GM
says our manager doesn’t have the weapons it takes to win. But, if he dug into the stats and matchups of games we lose, he would find out we could’ve given ourselves a better shot to win.
Optimistic...
I'm curious about a non-A's player: Carlos Marmol
He is the most unhittable pitcher I have ever seen. The numbers suggest it’s actually better to just go up there, not swing, and hope he throws 4 balls before he throws 3 strikes.
This year he has a K/9 of SEVENTEEN. Sure, he walks a lot of guys (but doesn’t give up hits or homers), but… what’s his wOBA?
www.zekeishungry.com
by thejd44 on Jul 20, 2010 7:01 PM PDT reply actions
This entire conversation shows me how much the
prediction mentality has dominated everyone’s perception of stats. Dan writes a whole paragraph in which he beats you over the head with the idea that this stat deliberately DOES NOT tell you anything about future performance — “no regard to luck, skill, or ability,” “a stat that doesn’t care,” “unsustainable in the future” — and yet in spite of that, the first few comments are people saying “oh, so we should trade so-and-so” and “ah, so so-and-so isn’t that good after all”, and then DFA follows up by telling us in about 50 different ways why this stat is no good because it doesn’t predict anything, and then in 25 more ways why it’s no good as a descriptive stat either because it doesn’t describe things about a player that really matter, like actual skill, skill being defined as something that you can count on him to show again some more in the … well, in the future, but who cares because descriptive stats are stupid anyway, and even though there are a billion predictive stats, if you’re interested in descriptive stats you may as well just throw them all in the trash except for runs because who care about descriptive stats anyway cause, you know, they’re stupid.
I mean, OK, I get that you guys like to predict the future, and that’s OK, but when Dan goes out of his way to say THIS STAT ISN’T ABOUT PREDICTING THE FUTURE, can you maybe at least hear what he’s saying?
There are more things in heaven and earth, Horatio, Than are dreamt of in your philosophy.
by iglew on Jul 21, 2010 1:01 AM PDT reply actions 3 recs
P.S.
(Not you, Lone Stranger. You’re cool.)
There are more things in heaven and earth, Horatio, Than are dreamt of in your philosophy.
Or you can completely ignore the crux of my argument which is about how its a shitty discriptive stat
because it doesn’t isolate the pitchers performance as an individual like FIP (which is a descriptive stat) but credits it as such. I never made any argument about how it wasn’t predictive if you actually read my argument, I just said it was worse than the stats we currently have at describing what dan wanted to. If you are trying to determine actual outcomes rather than true performance it makes absolutely no sense to use weighted values like wOBA does because any pitcher that starts a inning has a fair amount of control over the number of runners that are on base for example.
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 21, 2010 7:43 AM PDT up reply actions
I didn't ignore that.
It’s exactly what I was ranting about.
There are more things in heaven and earth, Horatio, Than are dreamt of in your philosophy.
then why the hell are you going on and on about predicting shit?
there are plenty of good descriptive stats like FIP, R, IP, this is not one of them. Can you please explain to me why you should use linear weights with a starting pitcher?
He's not missing bats and still giving up HRs like they're party favors at Chuck-E-Cheese - mikev
by designatedforassignment on Jul 21, 2010 12:06 PM PDT up reply actions

by 




























