Author Topic: Stats. Giggity!  (Read 38844 times)

0 Members and 1 Guest are viewing this topic.

Offline Jordanz Meatballz

  • Posts: 4996
Re: Stats. Giggity!
« Reply #75: September 15, 2012, 11:04:36 PM »
Thank you sooo much! Still interesting despite problems. Does fangraphs have adjusted ERA for the older teams? (ERA+ I think?)

Well, they have something called ERA- , which adjusts to the league and for park effects.  This was the result, again, for pennant winners:



If you're wondering (I sure was), the Nationals' ERA- is 81, third lowest in MLB.  Lower being better.

Oops, made a mistake with my data.  Fixed.  I knew the Rangers had to be wrong.  Changed it to after 1920, because the list was still dominated by older years.

Offline lastobjective

  • Posts: 4751
  • Natitude
Re: Stats. Giggity!
« Reply #76: September 15, 2012, 11:32:05 PM »
Well, they have something called ERA- , which adjusts to the league and for park effects.  This was the result, again, for pennant winners:

(Image removed from quote.)

If you're wondering (I sure was), the Nationals' ERA- is 81, third lowest in MLB.  Lower being better.

Oops, made a mistake with my data.  Fixed.  I knew the Rangers had to be wrong.  Changed it to after 1920, because the list was still dominated by older years.

Thanks again :clap: props to you. Have fun playing with the data. If I have any more strange requests I'll be sure to post 'em here.

And when I was talking about ERA+ - http://en.wikipedia.org/wiki/Adjusted_ERA%2B Never heard of ERA- but seems like the same thing.

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39378
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #77: September 16, 2012, 12:07:48 PM »
Main difference is scaling.  Folks thought it would be easier to understand with the lower number being better.   An ERA+ of 200 is the same as an ERA- of 50.  Both indicate league average adjusted for ballparks is twice what the team / player did.

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39378
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #78: February 04, 2013, 01:33:52 PM »
Nice article on the use and misuse of WAR as a stat, written in response to Jim Caple's ESPN piece objecting to WAR.  I'll excerpt pieces, and highlight some in red.  First, Cameron agrees the stat is misused:
Quote
we’ve probably all seen instances where WAR was used to end a discussion rather than promote one. Caple’s correct that WAR was never designed to be the only statistic that matters, nor should we view it as some kind of infallible truth. It is not a perfect metric; it contains some imperfect inputs and it does not sum up the entirety of baseball in a single number.
Then he asks a basic question about stats:
Quote
However, I think it’s worth talking about why WAR has become so ubiquitous over the past few years, and what its sudden rise in popularity should tell us about every kind of statistic. And I’ll start by leading off with Caple’s final two sentences:

"We need to look at many stats to assess players, and one of them should be WAR. But it shouldn’t be the only stat we look at or cite."

To that last statement, I’ll respond with a question: what is the point of any useful statistic? To me, a good stat is simply the answer to a question that is commonly asked.

Whether it is batting average, strikeout rate, swing percentage, or average velocity, each one was designed to answer a pretty simple question. How often does that player get a hit? How often does he swing? How hard did that pitcher throw his fastball? These are questions that are worth asking, and so we track things in baseball that allow us to answer those questions with data, assuring us of getting a pretty accurate answer in most cases. In fact, I think a litmus test for the usefulness of a statistic can be simply translating the definition into a question and figuring out how often anyone ever asks that question?

Let’s take Wins for a pitcher, for instance. For a long time, they’ve been hailed as one of the most important statistics in baseball, but the actual question that statistic is answering goes something like this:

“How many times did that pitcher complete at least five innings, leave the game with his team having outscored the opponents through the point at which he was removed, and then watch his relievers finish the game for him without surrendering the lead that his teammates helped create in the first place?”

No one would ever ask that question. It’s not something that’s worth knowing, nor does it help anyone understand what actually happened in any real way.


What does WAR answer?  Try this example:

Quote
If someone who didn’t know anything about baseball came up to you and asked you who the best player in baseball was right now, your answer would probably be someone from the group of Mike Trout, Miguel Cabrera, Buster Posey, Ryan Braun, Robinson Cano, Joey Votto, or Justin Verlander. But, what would you tell them if they then asked why?

Well, Miguel Cabrera has the highest batting average of any qualified player over the last three years, but that doesn’t answer the question of why he’s a better hitter than Joey Votto, who is number one in on base percentage. And how would you even begin to explain why Cano’s slugging percentage relative to other second baseman is more impressive than Verlander’s ERA? It’s not just apples and oranges anymore; now we’re mixing in things like sriracha, oreos, and Copper River Salmon, and it’s becoming a question of individual preference.

At that point, when the conversation is just beginning, something like WAR is exactly what you need, because it attempts to answer the very basic question that is being asked: “Is this guy better than that guy?”
...
There are thousands of baseball statistics. Most of them are narrow, precise instruments that provide a single answer to a simple question. Pretty much all of them are incapable of answering the single most asked question in baseball, which is why an all-encompassing metric like WAR has risen to prominence in the first place.

Rather than passing WAR off as the be all, end all, stat to make judgments on a player, Cameron suggests:

Quote
There are simply times when someone asks a general, imprecise question about the value of a certain player, and WAR is the perfect tool to provide a general imprecise answer to that question.
...
WAR is a great place to start that conversation. It is a fantastic filter, grouping players into manageable sizes of comparable performances, allowing for further evaluation of those who are candidates for the answer, depending on what the question is. It is not precise enough that anyone should be declaring a definitive answer based on a decimal point difference. Just like every other statistic in baseball, a one year result does not equal a player’s true talent level, so don’t claim that WAR thinks that Chase Headley is a better player than Miguel Cabrera. Depending on what question you’re asking, you may very well want to give less weight to a player’s defensive rating.

WAR is not the be-all, end-all of baseball statistics. However, it serves a great function as a good answer to a commonly asked question, and at the end of the day, that is really the entire point of a statistic. There are times when one wants to do a deep dive into every aspect of a player’s overall game, and WAR is not the right tool for that job. But, as a quick summary of a player’s overall value, it is the best tool for the job.

Great comment:

Quote
Tangotiger says:
February 4, 2013 at 10:12 am
There is no flaw in WAR. If you use a hammer on a screw, it’s not the hammer’s fault.

Offline comish4lif

  • Posts: 2934
  • Too Stressed to care.
Re: Stats. Giggity!
« Reply #79: February 04, 2013, 02:27:38 PM »
Tangotiger is awesome, that's a great comment.

Offline Lintyfresh85

  • Posts: 35130
  • World Champions!!!
Re: Stats. Giggity!
« Reply #80: March 08, 2013, 12:40:53 PM »
Gio struck out more pitchers last year (41) than any other pitcher has since 1972.

For the season, pitchers hit .019/.037/.019 against Gio.  :shock:

http://www.fangraphs.com/blogs/index.php/gio-gonzalez-pitcher-abuse-and-a-modern-day-record/

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #81: March 11, 2013, 03:36:11 PM »
Alright, which one of you coulrophiles is "King of the Byelorussian Square Dancers" on FanGraphs?

Offline blue911

  • Posts: 18482
Re: Stats. Giggity!
« Reply #82: March 11, 2013, 04:04:57 PM »
Alright, which one of you coulrophiles is "King of the Byelorussian Square Dancers" on FanGraphs?

Such an excellent word. I congratulate you.

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39378
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #83: March 13, 2013, 09:01:00 AM »
Is there a site to figure out win probabilities for the WBC games?  Fangraphs isn't tracking them, I think.  I'm really curious about Torre's early inning small ball approach.  I'm especially curious about the Phillips sacrifice bunt in the 5th with a 2 run lead.  I'm pretty sure in terms of run expectancy man on first, no outs, is greater than man on 2d, one out.  Is the greater likelihood of a single tack on run (the bunt) better for win probability than the lower likelihood of more runs to blow the game open (Phillips swinging away).

There's some talk that the international game is more of a small ball game.  Well, it looks like there are at least as many big scores as low scores in this tournament, it seems.  Also, my impression is that the pitching depth of most squads is not that great.  The USA seems to have a big bullpen advantage.  It strikes me as not smart to be playing for 1 run early when scores get big late off of bad bullpens.

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #84: March 27, 2013, 07:52:37 PM »

Offline Tyler Durden

  • Posts: 7970
  • Leprechaun
Re: Stats. Giggity!
« Reply #85: March 28, 2013, 07:08:34 AM »
My FanGraphs community research debut.
http://www.fangraphs.com/community/index.php/bill-moneyball-veeck/

Nice writing.  I enjoyed it.  Keep up the good work.

Offline Tyler Durden

  • Posts: 7970
  • Leprechaun
Re: Stats. Giggity!
« Reply #86: March 28, 2013, 07:12:39 AM »
http://www.washingtonpost.com/sports/nationals/nationals-rely-on-scouts-first-approach-but-take-information-from-elsewhere-too/2013/03/27/5681a1da-966d-11e2-8b4e-0b56f26f28de_story.html

Washington Post has an article up on the Nats use of sabermetrics.  The sabermetrics section of the scouting office consists of two twenty-somethings.  I guess that's better than nothing, which is what it used to be.  I kept picturing Jonah Hill in Moneyball as I was reading the article.


Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39378
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #88: March 28, 2013, 11:04:44 AM »
Goat Anderson = Dennard Span Spring Training 2013?
GA - .206 / .343 / .225, DS - .211 / .348 / .263

Nice job, Houston.  Glad to see you published.

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39378
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #89: March 29, 2013, 01:41:57 PM »
B-R and Fangraphs have modified WAR based on a team with a .294 winning percentage as replacement level.  Essentially, 47 - 48 wins.  This is based on about 1000 wins over 30 teams palying a 162 game season.  This is still roughly what a team of Rick Ankiels, Chris Youngs, Chris Snyders, Dice-Ks, etc... could win.  Guys who signed over the offseason to minor league contracts with invites or were waived. Minimal cost to acquire.  It is explained here.  Also, it is the level at which all but one player who has played 10 years has exceeeded (these guys were able to stick in the majors a long time soe were likely better than replacement). 

A lot of the difference in Br-WAr and f-WAR was based on a different assumption of what replacement level was.  FG tended to give more credit for longevity because it assumed a lower replacement level.  Thus, guys like Brooks and Yaz who played a long time were rewarded more in their career WARs at FG.  Was it Woody Allen who said half of life is just showing up?  That was why FG had Jack Morris at a higher WAR than BR. It was not due to the use of FIP v. an adjusted runs/ 9.

Cameron writes about why use a replacement player as a baseline here.  Basically, this rewards a player who may be below average but plays full time over the guy who was even worse that he beat out. 

Offline nobleisthyname

  • Posts: 2709
Re: Stats. Giggity!
« Reply #90: March 29, 2013, 04:27:41 PM »
I noticed Zim's 7 win seasons were dropped down to the high 6's.

Offline Lintyfresh85

  • Posts: 35130
  • World Champions!!!
Re: Stats. Giggity!
« Reply #91: March 29, 2013, 06:35:23 PM »
If BR just named their stat something other than WAR, we wouldn't have this problem.

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39378
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #92: March 29, 2013, 06:47:35 PM »
What's interesting is to see the numbers using different methodologies coming together.  As Cameron points out, the 1000 win level also is close to what Tom Tango had in his model when he first wrote about replacement level.

Offline blue911

  • Posts: 18482
Re: Stats. Giggity!
« Reply #93: April 05, 2013, 08:17:35 AM »

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #94: April 05, 2013, 08:43:03 AM »
"Bloomberg announced similar deal with the Los Angeles Anaheim"

Yeah, screw that team's stupid name.

Sounds cool though. Wish I could have a peek at the result. Oh, and every scout having an iPad? LAC  8)

Offline blue911

  • Posts: 18482
Re: Stats. Giggity!
« Reply #95: April 05, 2013, 09:05:48 AM »
"Bloomberg announced similar deal with the Los Angeles Anaheim"

Yeah, screw that team's stupid name.

Sounds cool though. Wish I could have a peek at the result. Oh, and every scout having an iPad? LAC  8)

They really should change their name to "The Los Angeles of Anaheim" and be done with it.

Offline comish4lif

  • Posts: 2934
  • Too Stressed to care.
Re: Stats. Giggity!
« Reply #96: April 06, 2013, 07:26:14 AM »
"Bloomberg announced similar deal with the Los Angeles Anaheim"

Yeah, screw that team's stupid name.

Sounds cool though. Wish I could have a peek at the result. Oh, and every scout having an iPad? LAC  8)
It's a used iPad 1.

Online PowerBoater69

  • Posts: 14265
    • Twitter
Re: Stats. Giggity!
« Reply #97: April 06, 2013, 07:54:01 AM »
Oh, and every scout having an iPad?

Who thought that was a good idea?  Scouts are on the road all year and Android is much better for pictures.

Offline zimm_da_kid

  • Posts: 7931
  • The one true ace
Re: Stats. Giggity!
« Reply #98: April 07, 2013, 08:17:14 PM »
Quote
Jon Morosi ‏@jonmorosi  3h 
Nationals and Astros are the only MLB teams without a stolen base in 2013. Odd couple, don’t you think?

Offline Tyler Durden

  • Posts: 7970
  • Leprechaun
Re: Stats. Giggity!
« Reply #99: April 07, 2013, 09:14:27 PM »
Bo Porter connection?