Converting OPS to wOBA
by colintj on Jun 24, 2009 1:36 PM CDT
Email
PrintDoes it get any more exciting than this? I bet not. Why write about this? OPS is used by everyone but Joe Morgan as shorthand representation for batter skill. wOBA, on the other hand, has no such following, as much as fangraphs has helped.
So what is wOBA? It's short for weighted On Base Average. It takes all possible batter outcomes (single, double, strikeout, etc.) and weights them according to their run value. The total number of runs is divided by plate appearances to give, in effect, runs per plate appearance. The final figure has been manipulated to an on base percentage scale where .340 is about average.
Using the standard approximation of 10 runs per 1 win, we can find the value of a batter's plate appearances. OPS was never designed with that in mind, but it turns out that wOBA and OPS correlate very very well.
The graph is the correlation of OPS to wOBA of 125 major league hitters over the last 3 calendar years, the data for which I nabbed from fangraphs. If you can't see, the r-squared value is 0.967 and the exact equation provided is
wOBA = 0.362*OPS+0.059
For ease of remembering, .36 and .06 are perfectly acceptable values. Plugging .340 for league average wOBA, we get that an average bat is about .775 OPS. 20 runs worse over the course of a season (600 PA), or about replacement level, is about .670. As long as we're talking about a fairly standard line (the regression had 3 consecutive seasons per batter, which significantly normalized batting lines), this conversion should not only be useful, but show perhaps that more advanced stats aren't getting us as far away from the hoi as we think.
On the other hand, that's an extremely high r-squared value, almost absurdly so. Perfect correlation is 1. Why so high? Power and on base percentage are dependent on each other. Pitchers tend to stay away from big sluggers, but don't mind busting weak hitters inside. So, there's a margin of error associated with avoiding the big bats. I would bet this holds particularly for those who mash fastballs. As support for this notion, I ran a second regression comparing ISO and BB% using the same data set and found an r-squared of .314, which is fairly robust. There are obviously skills involved outside of hitting for power that allow some to take more or fewer walks than opportunity allows, but there's a reason why you don't see any .300/.400/.400 hitters.
SouthSideSox is a community driven site. As such, users are able to express their thoughts and opinions in a FanPost, such as this one, which represents the views of this particular fan, but not necessarily the entire community or SouthSideSox editors.
88 comments
|
2 recs |
Do you like this story?
Comments
tim raines
and mike hargrove were pretty close.
Nice post Colin
its fun learning how to use numbers that make us feel superior to regular fans :D
Flowers – who is apparently the motherfucking boogeyman of the southern league
by omnipotent grab on Jun 24, 2009 2:07 PM CDT reply actions
yes, very pretty major - but can they fight?
So Colin finally broke down and took a stat class eh? r-squareds and regressions, look at him go!
Coach says he needs more toughness. I gotta step up.
ot: blog night 7/21
is anyone else going this year?
"Yes, the pitch in between the 2 home runs was HOF material.
Linebrink needs to be impaled by a trident." - Ozzie Montana 6-18-09
by e-gus on Jun 24, 2009 2:10 PM CDT reply actions 3 recs
It IS larry! And John Popper!
I can’t figure out the first guy – Paulie, Jim, Liney, or EG?
Ah, stocky white guys with goatees all look alike to me.
We’re a pack of a-holes.
by rhythm on Apr 14, 2009 1:45 PM EDT
wu is more like curly joe
"Yes, the pitch in between the 2 home runs was HOF material.
Linebrink needs to be impaled by a trident." - Ozzie Montana 6-18-09
sHe really needs to give my mullet back.
Just [expletive] Fuck It.
by homesickalien on Jun 24, 2009 2:42 PM CDT up reply actions
haha! *musket. not mullet. hahahahahaha.
Just [expletive] Fuck It.
by homesickalien on Jun 24, 2009 2:42 PM CDT up reply actions
ROFL.
Hey, it ain’t Billy Ray Cyrus, HSA. LOL.
We’re a pack of a-holes.
by rhythm on Apr 14, 2009 1:45 PM EDT
also its more rifle than musket
muzzle loader if you prefer.
have you had it checked out yet as far as when it was made or its value?
"Yes, the pitch in between the 2 home runs was HOF material.
Linebrink needs to be impaled by a trident." - Ozzie Montana 6-18-09
Haven't taken it anywhere (been really busy with moving) yet, but my friend's dad saw the pics and analyzed possibly late-1800s, Middle Eastern origin...he said def. not American....
and of the google image searching I did, the Middle Eastern versions were the only ones with the thick barrel rings, like mine…
Still going to take it a couple places, but I’m not in a huge rush with everything else happenin’. You better believe I’ll update when I know more…
Just [expletive] Fuck It.
by homesickalien on Jun 24, 2009 4:52 PM CDT up reply actions
I'm hiking the Argentinian Appalachian Trail that week
Twitter: @SouthSideCheat
SouthSideSox on Facebook
with the governor of SC?
"Yes, the pitch in between the 2 home runs was HOF material.
Linebrink needs to be impaled by a trident." - Ozzie Montana 6-18-09
shouldn't be a surprise to anyone
considering your name.
White Sox Minor League Updates: http://twitter.com/SouthSidelarry
by larry on Jun 24, 2009 2:26 PM CDT up reply actions 1 recs
thats awesome gus
takes some real stones for the one certifiable cocksucker on here to jump into this convo.-mm
larry = obama
i don't really consider a raspberry from sliding into home an injury
by U-God on May 31, 2009 4:22 PM CDT to parent up reply reply actions actions 0 recs
by Where Triples Go to Die on Jun 27, 2009 7:06 PM CDT up reply actions
Not sure
I call a r-squared value of .314 ‘fairly robust’, but other than that this is good stuff.
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
.3 is usually considered pretty solid
from what i’ve read. below .3 is poor. i wasn’t expecting much above that when i ran the data. but i do believe you can perform rough profiles and generate an expected BB% from ISO. throw in some age data if you want to get a little more complex.
THIS STORY ONLY ENDS ONE WAY
I dont think the lines
are that clear cut . . . let me put it this way. If you worked for me and handed me that, I would tell you to get back to fucking work and come back when you have something useful because I sure would not be confident in making predictions based on a regression with an r-squared of .314. To me it means there are clearly important, missing variables. That is not to say you can’t make note of the positive correlation and move forward.
What was the siginficance level, that is fairly, if not more, important as well?
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
i'm doing this for free
and claim no expertise. and i’m not claiming that as indisputable evidence; it’s definitely not. it’s there because i ran it and it adds to the discussion. you definitely would tell that to somewhat who worked for you. it’s incomplete research. the full extent of it is beyond my capacity.
for instance, i don’t know how to run a p-test, though i know what the results indicate.
THIS STORY ONLY ENDS ONE WAY
No, I know you are doing this for
free . . . I would just be careful posting about some of this stuff because there is a lot more to it than r-squared’s. I agree with you that intutively, I would expect those to be positively correlated . . . doesn’t always mean that is the case, which is why we even bother with this stuff. Without having a p-value, to understand the significance level, I do not think you can reliably state any case. If you get a decent r-squared, but the variable is not significant, does it matter? Not really. So, if you have that info, it would be useful, that was my point. Obviously, there is a shit ton more to regression analysis than that, but if you can at least do 2 things: 1) show the variable is significant and 2) have a decent r-squared (which is a measure of the model’s explanatory power) than you have at least covered 2 major steps.
Even assuming it is statistically significant, you would be stating that 31% of the variability in BB% is based on changes to ISO . . . which is fine, but kind of meh, imo.
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
like i said, i don't know how just yet
i’m working on it. you’re welcome to provide expertise. i do appreciate the advice and that follows from what i’ve gathered from Tango. what is it that you do at work anyway?
THIS STORY ONLY ENDS ONE WAY
you can calculate the P directly from Rsquared if you know the # of observations.
you say 125? if that’s the case, it is very highly significant.
not quite
you are going to at least need a tstat as well
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
you realize
this is a simple correlation, not a multi-variate regression?
I love pie
Moe: “When the roll is called up yonder I’ll eat pie.”
Curly: “Pi r squared?”
Moe: “No, pie are round; cake are square.”
Curly: “Oh.”
Moe: “No, O are round, also.”
"I put up some beautiful number." – Sammy Sosa
"I wouldn't know a steroid from a reefer." - Lou Piniella
yes
though, I dont know what you mean by ‘simple correlation.’ A regression is a regression.
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
i was thinking the tests
were the same for correlation coefficient and a simple (1 variable) regression. but that is not the case.
nevertheless, i’d venture to say his coefficients would be highly significant.
i dunno, you may be right
i haven’t used this shit in 15 years. go ahead and teach him how to do it.
And to clarify regarding the t-ratio
you get that by dividing the coefficient on the variable by the std error of coefficient.
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
i had a std error once
a shot from the big needle cleared it right up
"Yes, the pitch in between the 2 home runs was HOF material.
Linebrink needs to be impaled by a trident." - Ozzie Montana 6-18-09
good to know there are options!
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
I work in finance at a healthcare company
on the R&D side of things . . . I have software that spits all this shit out automatically if you have the data set.
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
the data set is linked in the post
fangraphs lets you export pretty much anything into .xls files
THIS STORY ONLY ENDS ONE WAY
If you are running this in Excel
You normally gives you the p value automatically when you run the regression
all he's sayin, i believe is
iso and bb% are positively correlated, which is what we would expect. what to do with that info, i’m not sure.
Tonight's Lineup
Josh Fields was the starting White Sox first baseman and Paul Konerko the DH, as Jim Thome received the day off against Dodger left-hander Randy Wolf Wednesday.
Podsednik lf
Ramirez ss
Dye rf
Konerko DH
Pierzynski c
Anderson cf
Nix 2b
Beckham 3b
Fields 1b
Floyd p
That is one pretty, pretty lineup
...take your hobbit circle jerk to another thread
by larry on May 18, 2009 8:58 PM PDT
Wow...6 through 9 will be quite awful
I’ll say a combined 2 for 10 with 5 Ks.
Good thing I stocked up on Summer Shandy.
If anyone is going tomorrow…tailgate in Lot G starts at 11…party on!
"I put up some beautiful number." – Sammy Sosa
"I wouldn't know a steroid from a reefer." - Lou Piniella
Why would Thome need a day off after 9 games of sitting on the bench and PH?
Is it more of a case of Konerko getting a day’s rest from the hot, hot weather while still batting?
by Scotty Ballgame on Jun 24, 2009 4:42 PM CDT up reply actions
Let's hope so
Standing at first base is simply grueling.
"I put up some beautiful number." – Sammy Sosa
"I wouldn't know a steroid from a reefer." - Lou Piniella
In this weather standing in general in that uniform is pretty brutal.
Especially at Konerko’s size.
by Grinder in Training on Jun 24, 2009 5:40 PM CDT up reply actions
I think you can assume
it is Wolf’s ridiculous line thus far against LHB. Still have a hard time with it considering the talent difference and the fact he is not traditionally that difficult for LHB to hit.
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
Is Fields at 1B an upgrade over Konerko?
That’s the part I don’t get. I can understand getting Fields some at-bats but why would you weaken an already bad defense.
by Scotty Ballgame on Jun 24, 2009 4:54 PM CDT up reply actions
Fields has this problem all the time
by hoodlight on Jun 24, 2009 5:03 PM CDT up reply actions 1 recs
WTF?
"Jenks, who was never afraid to say "no" to a hamburger..."
by BobbySouthSide on Jun 24, 2009 5:04 PM CDT up reply actions
it's ozzie's attempt to get fields in the lineup and against a guy who josh should theoretically have some success against.
for whatever that’s worth.
White Sox Minor League Updates: http://twitter.com/SouthSidelarry
and Josh never lets us down....come on buddy!!!
"Jenks, who was never afraid to say "no" to a hamburger..."
by BobbySouthSide on Jun 24, 2009 5:31 PM CDT up reply actions
since everyone would
have ripped him to shreds had Josh gone 0-4 and we lost by a run . . . definitely have to give him props, wow, cant say I saw that coming.
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
Neither did Ozzie,
to that extent. Mrs. Fields, either.
We’re a pack of a-holes.
by rhythm on Apr 14, 2009 1:45 PM EDT
Thome definitely needs a day off, been a long time for him
"I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it."
I've seen less shitty lineups in my toilet bowl.
Just [expletive] Fuck It.
by homesickalien on Jun 24, 2009 4:56 PM CDT up reply actions
Ramirez is my pick to click...it's CUBA hot tonight and he'll feel like home.
"Jenks, who was never afraid to say "no" to a hamburger..."
by BobbySouthSide on Jun 24, 2009 5:09 PM CDT up reply actions
You pick, I click
Who do you think is going to be the star of the game??
Secondly, how bout’ dem Yanks knocking off Spain! Yowsers!
"Jenks, who was never afraid to say "no" to a hamburger..."
by BobbySouthSide on Jun 24, 2009 5:18 PM CDT up reply actions
next year wouldn't it be nice
to have a DH that could actually play the field once in a while?
I would agree
That is what a bench is for and ours is filthy.
"Jenks, who was never afraid to say "no" to a hamburger..."
by BobbySouthSide on Jun 24, 2009 6:11 PM CDT up reply actions
i would phrase the question as
next year wouldn’t it be nice to have a DH that could put up a 150 OPS+?
if not, sure, he better be able to do something else. but one can best take advantage of the DH rule by having a guy who can rake, regardless of fielding ability.
White Sox Minor League Updates: http://twitter.com/SouthSidelarry
as an additional note,
the white sox are getting the second best production from the DH spot in the league. you can chalk that up to thome and the sox eschewing the ‘rotation’ that other teams favor.
White Sox Minor League Updates: http://twitter.com/SouthSidelarry
Im hoping we can get at least 110 OPS+ from our DH.
Depends on whoever else fills the role.
Flowers – who is apparently the motherfucking boogeyman of the southern league
by omnipotent grab on Jun 24, 2009 6:24 PM CDT up reply actions
Minus else*
Flowers – who is apparently the motherfucking boogeyman of the southern league
by omnipotent grab on Jun 24, 2009 6:25 PM CDT up reply actions
150 OPS+?
Aren’t those guys a dime a dozen?
Seriously, what about Dunn? From what I read the Nats are asking for a king’s ransom but wouldn’t he fill Thome’s roll next year?
by Scotty Ballgame on Jun 24, 2009 6:28 PM CDT up reply actions
*role
Apologies. I’ve been drinking already in preparation for tonight’s game
by Scotty Ballgame on Jun 24, 2009 6:31 PM CDT up reply actions
alrighty
wouldn’t it be nice if our DH wasn’t our one and only good hitter in the lineup?
OT
Flowers – who is apparently the motherfucking boogeyman of the southern league
by omnipotent grab on Jun 24, 2009 6:17 PM CDT reply actions
Thank Colin, this was helpful
wOBA = 0.362*OPS+0.059
For ease of remembering, .36 and .6 are perfectly acceptable values.
Quick typo (unless I missed something) .36 and .06 are perfectly acceptable. Not a big deal if you read the whole article, but if someone were to refer back to this at a later date it may screw them up big time.
-Jeeves Life in the Cell
Question
Aren’t the inputs to calculating wOBA and OPS more or less the same?
That would explain why the R was so high – a thing has a perfect correlation with itself.
Yawn.
i thought the weighting would make SOME difference
i expected a very high correlation though, for the reasons you say.
THIS STORY ONLY ENDS ONE WAY
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
In Association With
Big League Stew - MLB - Yahoo! Sports
- Photo: Padres fan regrets letting mother capture moment with Trevor Hoffman
- 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
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.
