The LLictionary Project: Genesis
by seattlebruin on Jul 14, 2009 10:13 AM PDT
Email
PrintIt's occurred to me that a lot of the terms we use around LL can be confusing even to established regulars, much less new posters. We throw around terms like "expect to regress," "tRA," and "J/M/G."
Many of these things are general baseball or statistics ideas, and some are LL-specific, but all contribute to making LL both an incredibly informative, if at time confusing place to read and post.
I propose that we start a FanPost collection of all the phrases we use, a Lookout Landing Dictionary. I know we tried it before with the Lookout Landing wiki, LLikipedia, but that didn't work as expected. What I propose is that we use this one FanPost as a giant dictionary and continue to add to it as we find new things that should be here, or stumble across old ones that need to be included.
The dictionary would include both baseball and LL specific jargon, and would be intended mainly to help new posters figure out what the hell we're talking about, and as a handy reference for older posters who might have been confused by something and never asked.
I think we should use a format in which we'd provide a brief definition, as well as a link to a relevant subthread or FanPost which might explain the idea in more detail.
EXAMPLE
Regression - Regression is a statistical concept that relates to non-conforming results. It essentially states that given an expected normal distribution, an outlier will, over a large sample size, move towards or "regress" to the mean value.
A good example of this is provided by Kermit. in this subthread.
Let me know what you guys think. Option B is to just re-create LLikipedia, but with less insanity (and probably a restriction on who can edit pages)
168 comments
|
9 recs |
Do you like this story?
Comments
Just as an FYI, I was imagining this would be a community effort, collected into a FanPost and then linked back to as necessary
as such, I need ya’lls help =)
Twenty minutes was just too long without a comment?
by Kirsten Schlewitz on Jul 14, 2009 10:34 AM PDT up reply actions
I don't think I mentioned it in the FanPost, so I put it here
also, I’m no longer sorry.
by seattlebruin on Jul 14, 2009 10:35 AM PDT up reply actions
It's still in the tags!
Also, I got my rogue membership card. Alas that is off topic.
by Kirsten Schlewitz on Jul 14, 2009 10:51 AM PDT up reply actions
This is a good idea
I would like to see a handy-dandy description of “replacement level” contained in this resource.
Nice Guys Finish Third - My semantics are a waste of time.
Stop being so needy
I want to poop at your house - Thingray
by tootthekazoo on Jul 14, 2009 11:03 AM PDT up reply actions
Can we once and for all declare what tRA means?
by Kirsten Schlewitz on Jul 14, 2009 10:51 AM PDT reply actions
I'm forever referring to it as True Run Average
by seattlebruin on Jul 14, 2009 10:57 AM PDT up reply actions
Tawdry Run Average?
I want to poop at your house - Thingray
by tootthekazoo on Jul 14, 2009 11:03 AM PDT up reply actions
yes please
I’m sure there are glossaries out there for some of this stuff.
by Snuffleupagus on Jul 14, 2009 11:05 AM PDT up reply actions
tabulation RUN AMOK.
Amirite? Fucking nerds.
Sure.
tRA means assigning run and out values to all events under a pitcher’s control and coming up with an expected number of runs allowed and outs generated in a defense and park neutral environment.
??
events under a pitcher’s control = ? (SS, BB, FB, GB ??)
expected number of runs and outs generated = how is this then turned into a single number? what’s good? what’s bad?
This is a stat I would love to understand.
by Snuffleupagus on Jul 14, 2009 11:10 AM PDT up reply actions
In seriousness.
Do people who are not deep into maths and other stats get and understand that article? Isn’t that what this piece is for—more of a people’s understanding of what is going on?
by Kirsten Schlewitz on Jul 14, 2009 11:22 AM PDT up reply actions
I think that's the intent, yes
a layperson’s thumbnail guide to terms.
Nice Guys Finish Third - My semantics are a waste of time.
I realize the basic point behind tRA.
What I don’t think y’all realize is that for non-math or non-stats proficient people, reading that post can be difficult.
Now, you can make the argument that ok, we’re a stats oriented blog and so why would people that don’t get it want to come here? While I don’t want to go around dumbing things down, I think it would be possible to write some things down in shorter, more concise ways, that would allow people to at least get a basic feel for things.
by Kirsten Schlewitz on Jul 14, 2009 11:36 AM PDT up reply actions
No, I get the point of it.
I just do not see how to make it simpler/shorter.
I'll give it a shot
I think I was able to help NOLA
by seattlebruin on Jul 14, 2009 11:56 AM PDT up reply actions
It's not so much "don't follow" as "don't want to read so much"
Look at sb’s example definition of regression above. Can tRA be summed up in two sentences like that? That’s what I think the point of the post is.
Nice Guys Finish Third - My semantics are a waste of time.
Yes, and as was stated down below, it could have a definition then a link
Matthew sort of did it above, but that definition of tRA seems not quite right to me. And there’s no short definition for UZR.
by Kirsten Schlewitz on Jul 14, 2009 11:42 AM PDT up reply actions
UZR will be a disaster and I am not volunteering to write that one
I don’t have nearly the grasp of it to even consider explaining what it means besides that it makes Franklin Gutierrez even more awesome
by seattlebruin on Jul 14, 2009 11:43 AM PDT up reply actions
UZR is a measure of defensive ability based on range, outfield arm/double play ability, and errors.
It is measured in runs saved/lost with respect to an average fielder.
I don't see how to re-word the definition of tRA I stated above.
That’s the two sentence version.
Try something like
tRA is a method of assigning run and out values to all events under a pitcher’s control, correcting for defense and park factors, and producing an expected number of runs allowed and outs generated by that pitcher.
It’s not substantially different than yours, but it gets the point across in a little less jargon-y fashion.
Nice Guys Finish Third - My semantics are a waste of time.
The problem for me is that tRA is a number that looks like ERA, meaning 1.23, correct?
So how does that come from "producing an expected number of runs allowed and outs?
by Kirsten Schlewitz on Jul 14, 2009 12:44 PM PDT up reply actions
Wow if only you could get in touch with it's creator
by Poochie on Jul 14, 2009 12:49 PM PDT up reply actions 1 recs
That's the point I am trying to make
The creator and those who work with these things all the time don’t necessarily know how to explain these in the best manner—because they have never not understood what is going on.
by Kirsten Schlewitz on Jul 14, 2009 12:53 PM PDT up reply actions
If this becomes an official "thing" with a set of terms to work with I would be glad
to volunteer my services of turning tech-speak into layman’s terms – that’s what I do in my job, no reason I shouldn’t do it here.
Nice Guys Finish Third - My semantics are a waste of time.
I don't contribute much to the serious baseball discussion 'round these parts
so I gotta do what I can.
Nice Guys Finish Third - My semantics are a waste of time.
I'm trying to get away from that myself
It’s so easy to do and really difficult to get away from
Comparative advantage.
We’ve already got the best dick joke sally around; no need making pdb try and reinvent the merkin.
I will be sending out a Facebook message asking for help soon
by seattlebruin on Jul 15, 2009 12:56 AM PDT up reply actions
I'd be more than happy to provide some translation efforts.
You can look me up on Facebook. The name is: Robert Lintott.
by Robert Lintott on Jul 15, 2009 3:59 AM PDT up reply actions
I'm just giving you a hard time
My guess is that like wOBA each batted ball, each result, has an assigned run expectancy to it and the average is derived from that.
I'm going to be an amazing teacher of advanced stats someday.
by Kirsten Schlewitz on Jul 14, 2009 1:21 PM PDT up reply actions
I'm allergic to hard work.
I break out in hives that can only be cured by scotch and beer.
by Sec 108 on Jul 14, 2009 1:36 PM PDT up reply actions 1 recs
I don't know, I was just editing what was written above
I have very little understanding of the mechanics of the statistic. I ain’t so bright with the advanced stuff.
Nice Guys Finish Third - My semantics are a waste of time.
I'm about the same.
I understand why the stats are good, but not so much the processes or anything. I’m eager to learn, so I hope I can check this thread again this evening and find myself a wealth of knowledge to soak up
I want to poop at your house - Thingray
by tootthekazoo on Jul 14, 2009 12:51 PM PDT up reply actions
Sometimes the comment threads are the best resources for explaining the how's and why's of a stat
I had a lot of good back and forth conversations bookmarked and sorted, sometimes the conversations are the best resource. They often pertain to a specific incident, due to the circumstance (say Gutierrez acquisition in the offseason) a lively discussion of the benefits of run prevention vs. batting value will spring up.
Lots of misconceptions will be aired, either as a question or as an incorrect statement of fact. The following back and forth will be very illuminating. People often ask questions I never thought of asking, it’s brilliant. Often I find the conversations are much better at defining a thought process or particular stat than a simple definition.
I think these are the discussions we might be looking for.
I myself learn better from specific examples, as in relating things back to a player.
by Kirsten Schlewitz on Jul 14, 2009 1:19 PM PDT up reply actions
Exactly! I hate it when I use 300 words, and someone says it better with 15
Stupid stupid (not you, me)
ERA=
Earned Runs / Outs * 27
tRA =
Expected Runs / Expected Outs * 27
So would this be fair?
The number of runs you would expect a pitcher to surrender assuming an average defense based on pitch results.
by Aaron Campeau on Jul 14, 2009 1:04 PM PDT up reply actions
average defense in an average park
based on outcomes under a pitcher’s control.
See, yes, that way it makes sense.
I like the definition created below (you and ac together) as well.
by Kirsten Schlewitz on Jul 14, 2009 1:21 PM PDT up reply actions
Well, for what it is worth,
That exact formula is listed in Graham’s piece.
What I was trying to get at is that the way the formula is set up and the use of words to describe it seemed contradictory.
Not that I hadn’t read the article or missed that formula.
by Kirsten Schlewitz on Jul 14, 2009 1:28 PM PDT up reply actions
.
The expected number of runs a pitcher will surrender over nine innings, assuming average defense and park, based on outcomes under a pitcher’s control. Said outcomes are batted ball types (ground balls, fly balls, line drives, home runs,) strikeouts and base on balls.
by Aaron Campeau on Jul 14, 2009 1:25 PM PDT up reply actions
I even wrote one without numbers or equations just for that and posted it here
It’s not really my fault that y’all misplaced it
by Graham MacAree on Jul 14, 2009 4:42 PM PDT up reply actions
tRA question
Looking at the tRA tables, I noticed that the out value of a K was 1.000. Are you taking into account batters who reach on strikeouts, or is the value so low as to be negligible?
by seattlebruin on Jul 14, 2009 5:07 PM PDT up reply actions
Remus with a headcold?
I will smash your face into a jelly.
by Phil Hatzenbuehler on Jul 14, 2009 11:26 AM PDT up reply actions
I'd be happy to contribute but I'm not hip to the lingo
So, I’m happy to let others contribute so I can become hip to the lingo
I want to poop at your house - Thingray
Does this mean
that somewhere in this fanpost I will learn what UZR is? O happy days.
Thanks
these are great. For me, the idea of this fanpost would be to collect links like this and preceed them with a short explanation of what the term is. Then people could follow the link if they want more.
Have a good one for BABIP? Another stat which would be great to understand. (sorry if this is annoying, but what I find via google seems to be really hit or miss as to quality.)
by Snuffleupagus on Jul 14, 2009 11:39 AM PDT up reply actions
Yup, that's basically what I want to do
I’m willing to put the time in, but I need help from others too
by seattlebruin on Jul 14, 2009 11:42 AM PDT up reply actions
BABIP is batting average for balls in play.
Basically it takes out any strikeouts, homeruns and walks and gives you their average without those. It is a good means of measuring whether or not a player has been lucky.
I hope I am right.
Just on a cursory glance
Here’s a link from THT
In brief, BABIP measures the rate at which batted balls (excluding home runs) turn into hits. When used with something like line drive percentage, it can give you an idea of how lucky or unlucky a batter is.
this is based on the fact that players can't aim?
I always thought a player could try to pull or push a ball depending on how the defense lined up. Is that just not accurate? (maybe this isn’t the place for this discussion)
by Snuffleupagus on Jul 14, 2009 11:53 AM PDT up reply actions
This is the place, and essentially yes, this is correct
a batter can generally try to hit a ball to right or left field, but he cannot aim a ground ball between the shortstop and third baseman – this is entirely luck.
by seattlebruin on Jul 14, 2009 11:57 AM PDT up reply actions
A player can go with the pitch and have some influence on whether they pull the ball or go up the middle or opposite field
but there’s really only so much a player can do when they put a good swing on the ball and hit a line drive.
From my understanding, BABIP and LD% work well together. If you see a player with a high LD% but a low BABIP (see Adrian Beltre: 2008), you can assume they’re hitting into some tough luck.
Does anyone know of a one-stop-shop for explanations on MLB rules relating to a team's mangament of players?
I’m thinking of things like how arbitration works, free agent loss compensation, demotion “options”, “under team control”, and the rule 5 draft. I believe I’ve seen something like the actual MLB rulebook for these things, but a resource that explains these concepts to the layman would be really handy.
by Manzanillos Cup on Jul 14, 2009 2:26 PM PDT reply actions
Nice.
Exactly what I was looking for. What service!
by Manzanillos Cup on Jul 14, 2009 3:56 PM PDT up reply actions
So I think I am going to set a goal of adding 10-15 definitions per day to this FanPost
between myself and whoever wants to help. If you want in, please just reply to this comment with whatever you want to write up, or ask and I’ll put suggestions out there.
Format wise, I think it works best if it’s kind of brief, like the regression definition in the OP. Two-six sentences should be plenty, and if you want, a link to a relevant discussion would be awesome as well.
So Matthew has given me the OK to use the LL Wiki to compile all this information
I’ll get started when I get the chance to clean it up a bit, and as we get definitions, we can start adding them to the wiki. You need to be given permission to edit, because we want to avoid the Great LLikipedia Disaster of 2008 repeat if at all possible.
by seattlebruin on Jul 14, 2009 5:05 PM PDT up reply actions
I could do Batted Ball Profile, since that comes up a bunch
Batted Ball Profile- a batted ball profile for a player (in its most basic form) is the percentage of balls the player hits on the ground, in the air to the outfield, popped up, or as a line drive. These can be useful in determining a player’s luck (and therefore their ability to sustain a trend). FanGraphs (here for Ichiro) and StatCorner (the same) have very useful and accessible batted ball profiles.
by Robert Lintott on Jul 15, 2009 10:06 AM PDT up reply actions
but you might want to mention that while batted ball profile is much more predictive than results-based analysis,
it still has the problem of scorer subjectivity, especially on line drives
by seattlebruin on Jul 15, 2009 10:10 AM PDT up reply actions
How's this
Batted Ball Profile- a batted ball profile for a player (in its most basic form) is the percentage of balls the player hits on the ground, in the air to the outfield, popped up, or as a line drive. These can be useful in determining a player’s luck (and therefore their ability to sustain a trend). Scorer subjectivity (what is a line drive? What is an outfield fly?), however, leads to a margin of error that must be taken into account. FanGraphs (here for Ichiro) and StatCorner (the same) have very useful and accessible batted ball profiles.
by Robert Lintott on Jul 15, 2009 10:27 AM PDT up reply actions
Great!
this is perfect, but the last thing we probably want is a discussion somewhere on how batted ball profile is used. Don’t stress too much looking for one though – I’m guessing these are going to come naturally from other discussions and then get added to the LLictionary as we go.
by seattlebruin on Jul 15, 2009 10:44 AM PDT up reply actions
I didn't have to stress much, and you even provided it.
From today’s Branyan post.
by Robert Lintott on Jul 15, 2009 10:50 AM PDT up reply actions
It's almost like this was intentional =)
by seattlebruin on Jul 14, 2009 4:01 PM PDT up reply actions
not surprising given the nature of this place.
i’ve been lurking for years. I laugh as much as I learn here. I’m one of the “regs” in terms of reading and one of those who get confused by every post from Graham and Matthew =). This idea would help tremendously.
br
br
by sirbrianwilson on Jul 14, 2009 4:05 PM PDT up reply actions
From the responses I hope you keep tight reign on the editing
Are you looking for specific definition, or is something like that link to wOBA in regards to Nick Swisher ok or even wanted? Those kind of conversation threads are a bit vague at times, and not always specific (the Swisher bit involved about 3 different metrics), but I find them insightful as they kind of show metrics in action. Maybe a separate thread for that type of stuff? I don’t want to pollute your thread with unwanted material
Something like that is perfect
but would be trying to link to the top comment of a subthread to give a reader specific direction as to how a metric is used.
The editing control will be very tight indeed
by seattlebruin on Jul 14, 2009 7:25 PM PDT up reply actions
I just put 2 and 2 together and realized what putting your definitions in the wiki means
You’ll be able to have all the links you want, footnotes, etc., wtf ever.
I have a relevant gif!
angels fan in seattle
by Eyebrows on Jul 14, 2009 4:17 PM PDT up reply actions 4 recs
Who is this shuh nareee?
And why do you want me to lick her?
Thug Life
by Slow Country on Jul 15, 2009 8:03 PM PDT up reply actions
Yiff-tionary.
I will smash your face into a jelly.
by Phil Hatzenbuehler on Jul 15, 2009 1:13 PM PDT up reply actions 2 recs
Something everyone can help me out with -
let’s compile a list of useful acronyms we use around here. These can include anything from the LL nickname of a player or user to advanced stats like wOBA to names for other places (DRB, HH, etc.)
I suggest even listing them if you don't know what they mean and would like a definition.
Nice Guys Finish Third - My semantics are a waste of time.
Hmmm...
Loafie = Jose Lopez
DTFT = Death to Flying Things = Sex in Midair = Frabklin = Gutz = Guti = Gut = Franklin Gutierrez
(I am) Beluga Tits = anagram of Miguel Batista
Sex Cannon = Russell Branyan
Sex Wagon = Jeff Clement
WAR = wins above replacement level (I couldn’t find a good one post Fangraphs explanation of this for position players but only for pitchers).
Pretty much everything with a handy-dandy roll-over definition on FanGraphs.
Maybe something like Pizza Cutter’s work on when statistical samples become reliable (hitters and pitchers) would be helpful too.
This actually brings up a couple fairly fundamental questions (WARNING: here be corporate-speak)
As far as document scope. First, how much of the wheel do we want to reinvent? I see things like
Pretty much everything with a handy-dandy roll-over definition on FanGraphs.
and I wonder if we need to replicate other work here, or if we should just say something in some sort of intro to the final product, like “Definitions and explanations for any statistical term you don’t see explained here can be found at Fangraphs” or something like that.
Second, about the nicknames and site in-jokes. Who will own them going forward, once the doc goes live, so that they’re updated and don’t become stale?
Nice Guys Finish Third - My semantics are a waste of time.
I actually think they should all be in there
but I don’t want to have everyone spend all this time to build a mammoth reference source and then have nobody maintain it going forward, and have it be out of date when players leave or jargon changes (dead memes for example), so that two years from now we have to rebuild the thing all over again.
Nice Guys Finish Third - My semantics are a waste of time.
Probably me
I could use some help.
Scope is of course a huge problem – how far do we want to extend it? If we put all the FanGraphs stuff on there, are we obligated to describe how everything is determined?
This is the biggest question for me right now. Clearly we need analysis-specific terms and information, but at what point do we stop?
by seattlebruin on Jul 15, 2009 2:33 PM PDT up reply actions
My guideline would be
if it’s been well and concisely defined elsewhere, list it and refer to the site which has the good definition in our doc, if not we’ll do it. Why do a bunch of work that’s already been done?
Nice Guys Finish Third - My semantics are a waste of time.
By "it" I of course mean
if a term or stat has been well and etc etc etc
Nice Guys Finish Third - My semantics are a waste of time.
totally Ridiculous Assumptions
It’s where you look at ERA and assume it is representative of the pitchers actual individual abilities, and is a repeatable skill. About as accurate as a National Guard artillery section, with the kill radius of a marshmellow to the face.
I believe it stands for true runs allowed.
...and now I'm here
Isolated Slugging and Isolated Discipline.
Isolated Slugging is SLG-BA and Isolated Discipline is OBP-BA.
by Aaron Campeau on Jul 15, 2009 2:27 PM PDT up reply actions
I'm really getting upset about losing nearly a years worth of bookmarks
They are really helpful, and the comments are sometimes isolated, located randomly during a game thread, or during an aside that has nothing to do with the original post. So basically all over the place.
I'm assuming you're running Firefox, if not go get it and use it instead of IE
Nice Guys Finish Third - My semantics are a waste of time.
Drat! Something is wrong with the profile folder
Eventually tried removing 3.5 and restoring 3.0 FFX but no joy. But thanks for the link, I didn’t know FFX could do this or the syncing function.
How many wins
Is a replacement level position player worth in a year?
by Snowman1025 on Jul 15, 2009 5:36 PM PDT via mobile reply actions
That's why it is wins above replacement.
Or how many more wins a player is worth above an average minor leaguer (AAA, i think) of the same position.
Thug Life
by Slow Country on Jul 15, 2009 8:16 PM PDT up reply actions
I'm pretty sure it's 46 wins on Fangraphs, right?
Fun Fact: the 2003 tigers went 43-119, and their position players were a half win below replacement and their pitchers were only 3.9 WAR above replacement (a 49 win true talent team, which coincides with their baseball reference pythag of 49-113).
The number varies depending on different conceptions of RL, but yes, I think that's what fangraphs uses.
(I just meant I wasn’t sure what Snowman was asking…. if he meant how many WAR, then the answer seems simple. If he meant how much does it cost to acquire one, well, that seemed simple too.)
Maybe some of this series by Dave Cameron would be useful to link to (I wish there were a page which summed up the series, but alas)
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-catchers
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-first-base
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-second-base
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-shortstop
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-third-base
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-left-field
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-center-field
http://www.fangraphs.com/blogs/index.php/2009-replacement-level-right-field
http://www.fangraphs.com/blogs/index.php/the-2009-replacement-level-position-players
http://www.fangraphs.com/blogs/index.php/wrapping-up-the-week
http://www.fangraphs.com/blogs/index.php/replacement-level-on-two-feet
Thanks; this is a really, really useful series for the LLictionary.
Esp. for those who want to tie the theoretical concepts to specific players.
I was going to attempt a layman's translation of wOBA, but hey. How about OBP and
Which is the definition (layman) as I look at these stats. It’s like trying to learn a foreign language, when you’re not very good at it you have to translate it from english in your head. It takes a long time, if ever, to start thinking in these terms.
Anyway, when I get to wOBA Tango pretty much has written the simplest explanation of what information wOBA is telling you, and how to look at it to see it, I’ve yet to find.
The bottom line is this: if you see a guy with a .360 OBP and a .360 wOBA, then you know he’s got a "normal" profile of extrabase hits and walks. If you see a guy with a .340 OBP and .360 wOBA, then you know there’s alot of power that his OBP is not capturing. If you see a guy with a .380 OBP and .360 wOBA, then you know he’s not a power hitter. That’s why I like wOBA.Link to source.
Now I’m thinking for someone new to SABR it might be most efficient if there were a primer or map. A list of basic terms to understand, in some type of order, that would get the average Joe (me) quickly up to speed. Off the top of my head in no particular order.
Rate stat, counting stat, appeal to ignorance, and then all the stats in StatCorner. Linear weights.
I’m going to make this way more complicated than it needs to be, so put the breaks on at any time. What I’m thinking of is not to bring a person up to speed of an analyst level of sabr fluency, but just fluent enough to be able to follow along with a conversation without their eyes glazing over. This probably already exists someplace, if so I’m going to be pissed I didn’t look for it a year ago.
Something that would be helpful for people just getting into more advanced metrics
would be explanations for why traditional stats are inadequate. If a person understands why BA, ERA/wins, RBI, etc. are lacking they’ll be more likely to leave them behind and have a better understanding of what wOBA, UZR and tRA are trying to measure.
Oooh, I like this
we can have pages for ERA where it will say something like
ERA is a traditional statistic used to evaluate pitchers. It is calculated as earned runs allowed per nine innings pitched.
However, ERA has major flaws in that it is both defense-dependent and fails to take into account park factors. Because of this, metrics that evaluate pitcher performance only should be used, as opposed to ERA, which measures combined pitcher and defensive performance.
The main two measures used on LL are tRA and FIP or xFIP. These seek to measure only events pitchers can control. For more information, please refer to the tRA or FIP pages.
by seattlebruin on Jul 16, 2009 9:52 AM PDT up reply actions
Blockquote fail, whoops
should be entire last three paragraphs
by seattlebruin on Jul 16, 2009 9:52 AM PDT up reply actions
I'd prefer for the LLictionary to have lots of links but also good information as a stand-alone
not to just a be a giant linkfest
by seattlebruin on Jul 16, 2009 9:57 AM PDT up reply actions
Awesome.
Just call dibs on what you want. So far the only for sure assigned pages for Ronny Ce:Deno, Jose Lopez and wOBA.
by seattlebruin on Jul 16, 2009 10:04 AM PDT up reply actions
Try this.
Replacement level is (the high end of) the talent level that a team may obtain for a minimum salary on the open market or at minimal cost in a trade. This is called “freely-available talent.” A full MLB team of replacement-level players, as FanGraphs calculates it, would be a true-talent 46-116 team.
Possible citations:
http://www.fangraphs.com/blogs/index.php/glossary/ (WAR Series by Dave)
http://www.hardballtimes.com/main/article/replacement-level-article/
http://www.insidethebook.com/ee/index.php/site/article/how_to_calculate_war/
http://www.lookoutlanding.com/2009/7/14/949000/the-llictionary-project-genesis#18273237
That’s all for right now. I’ll update it this weekend.
Not to keep sounding like a corporate whore
but at some point is somebody going to
1) cut off the suggestion period for a while while a first draft is put together and list what will be in that first draft
2) put together a template for what this document/page might look like?
I can take a stab at #2, at least structurally speaking, while we await the results of #1, but there are some great suggestions in this thread so far and I want to start getting them documented in some sort of first draft before the earliest ones get lost.
Nice Guys Finish Third - My semantics are a waste of time.
So, I've been thinking about this for a while, and thought this might be a place to bring it up...
I know BABIP. I know batted ball profiles. What I don’t know is the statistic that brings it together into a single number, to analyze someone’s luck. It seems like it should be very easy to determine an expected BABIP based on their profile and then compare it to their true BABIP. Something like this: (LD BIP x LD avg BABIP) + (FB BIP x FB avg BABIP) + (GB BIP x GB avg BABIP) = xBABIP. BABIP – xBABIP = “luckiness”
Does this exist somewhere and I just don’t know it? Is there some reason this information wouldn’t be valuable, like some weakness in the formula above?
It's call prOPS and it's not very useful because batter skills distort run and out values for batted ball profiles
by Graham MacAree on Jul 16, 2009 1:41 PM PDT up reply actions
For example, Derek Jeter has a .342 BABIP
Lucky, right? Not really. He’s at .361 for his career.
by Jeff Sullivan on Jul 16, 2009 1:51 PM PDT up reply actions
Yes but in reality this is really hard to do
by Graham MacAree on Jul 16, 2009 1:58 PM PDT up reply actions
You have to run the formula individually for every player with varying regression components
It’s hard.
by Jeff Sullivan on Jul 16, 2009 1:59 PM PDT up reply actions
You'd think this could be automated if the correct data is presented
by seattlebruin on Jul 16, 2009 2:41 PM PDT up reply actions
You need massive database indexing.
For instance, you have to keep track of the past X many LDs, GBs, FBs, etc for every player and update them on a rolling basis. Then you need to re-run your formula every day.
I’m working on it, but it’s a mammoth problem in scale.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Lookout Landing
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
In Association With
Big League Stew - MLB - Yahoo! Sports
- Ku Klux Klan, umpires and bowling: Why the minor leagues tossed out Jim Evans academy
- The Cardinals Hall of Fame can now be found at Bill Clinton’s library in Arkansas (Huh?)
- Will Bryce Harper start the season with the Nationals?
- Mitch Mustain puts football behind him, signs reported rookie deal with White Sox
- Robinson Cano and Denzel Washington face off in Jay Leno’s Yankees trivia contest
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.
