Author Topic: Stats. Giggity!  (Read 38887 times)

0 Members and 1 Guest are viewing this topic.

Offline Squab

  • Global Moderator
  • ****
  • Posts: 4528
  • me lookin at the bullpen
Re: Stats. Giggity!
« Reply #300: July 26, 2014, 10:24:51 AM »
Somewhat related to stats, but mostly not, feel free to help support my grassroots rebellion attempt to replace Carson Cistulli as managing editor of NotGraphs.

You have my support.

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39413
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #301: July 28, 2014, 01:57:16 PM »
how many PAs before walk rates stabilize for hitters?  Span is having a month where his BB% is at 13%, nearly double what he had going into the month.  92 PAs.  FP mentioned some stance changes, getting his front foot down quickly, and waiting on the ball longer.  He's had 3 multiple walk games this month.  I'm wondering if this could be the new normal.

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #302: August 04, 2014, 08:43:50 PM »
Anybody know where to find split leaderboards? In particular I'm interested in career leaders in batting with the count 3-0 (to compare to Werth)

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39413
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #303: August 12, 2014, 09:40:20 AM »
Houston / PC / Linty - this interest you? 

PC - I don't know if you are interested in doing this, but I'd like to see a bar graph of the Nats scoring 0 to 1 runs, 2 runs, 3 runs, 4 runs, 5 runs, 6 runs, and 7 or more, and compare that to the league total rates (divided by 12 if you exclude the Nats).  I'm not so interested in the MLB total because the AL should be a higher run scoring with the DH.   I would guess that it is very rare to win with only 1 or fewer runs (1 being close to 0% and 0 of course being 0%), that there is a bit of a bell curve in the 2-6 range, and that the tail of 7 or more is close to 100%. 

I'm not so surprised that we have relatively few games of 3 or less.  Our run total should bias high. We probably do blow people out more often than most teams.  I'm going to guess that we are not doing as well for wins with 3 runs as the league average.

Maybe throw this in the stats giggity thread, but it would be interesting.

This is another thing I think Blue mentioned when I was making good eye-contact but paying more attention to my beer.  This goes to the utitliy of the Pythagorean win estimators and whether teams with thicker tails in their run distribution (more big run games, more low run games) would have their win percentage exaggerated in the Pythag when compared to teams that score 4-6 runs and have thinner tails in the distribution. 

I dimly recall that for that for the same runs scored, teams that are SLG heavy in their OPS have a narrower distribution of runs scored than teams that are OBP heavy in their OPS.  I think it was in a Fangraphs article several years ago.

Offline blue911

  • Posts: 18486
Re: Stats. Giggity!
« Reply #304: August 12, 2014, 09:53:19 AM »
I think we were drinking Jack Daniels

Offline ChasingShadows

  • Posts: 84
Re: Stats. Giggity!
« Reply #305: August 12, 2014, 10:04:06 AM »
I think we were drinking Jack Daniels

Johnny Walker's Red for me.

Offline PebbleBall

  • Posts: 3440
  • Now that right there is baseball.
Re: Stats. Giggity!
« Reply #306: August 12, 2014, 10:40:16 AM »
Anybody know where to find split leaderboards? In particular I'm interested in career leaders in batting with the count 3-0 (to compare to Werth)

PI has it. Here's 100 PA.  4 guys have 100+ 3-0 PA with 100% walk (or HBP) rate. 

3-0 Count, 100 PA

Min. 10 AB, only Wieters is better by BA, Werth 8th in OPS

3-0 Count, 10 AB

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #307: August 12, 2014, 10:47:13 AM »
PI has it. Here's 100 PA.  4 guys have 100+ 3-0 PA with 100% walk (or HBP) rate. 

3-0 Count, 100 PA

Min. 10 AB, only Wieters is better by BA, Werth 8th in OPS

3-0 Count, 10 AB


Thank you so much!!!

By the way, on the all-time list, sorting by at-bats, Werth is 8th in BA and Wieters is first.

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #308: August 12, 2014, 10:53:52 AM »
Also, if you include inactive players, who shows up on the leaderboard? Matt Wiliams. No wonder Werth has a green light.

Offline JCA-CrystalCity

  • Global Moderator
  • ****
  • Posts: 39413
  • Platoon - not just a movie, a baseball obsession
Re: Stats. Giggity!
« Reply #309: August 12, 2014, 02:52:25 PM »
I think we were drinking Jack Daniels
I drink enough J.D. that I got a degree for it.

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #310: August 12, 2014, 08:13:17 PM »
PI has it. Here's 100 PA.  4 guys have 100+ 3-0 PA with 100% walk (or HBP) rate. 

3-0 Count, 100 PA

Min. 10 AB, only Wieters is better by BA, Werth 8th in OPS

3-0 Count, 10 AB

New question. How do I see if somebody swung and missed?

Offline blue911

  • Posts: 18486
Re: Stats. Giggity!
« Reply #311: August 28, 2014, 09:12:48 AM »
Jason Parks is leaving Baseball Prospectus for a scouting job with the Cubs.

Online HalfSmokes

  • Posts: 21606
Re: Stats. Giggity!
« Reply #312: August 28, 2014, 12:35:43 PM »
http://www.fangraphs.com/blogs/so-lets-talk-about-alex-gordon/

good article (of course, I don't agree with the nothing to see here tone- Cameron is pointing out how flawed the defensive component of WAR is, but also defending WAR), but kind of interesting to compare this to the arguments in favor of Trout in 2012 based on WAR (including the defensive component)

Offline tomterp

  • Global Moderator
  • ****
  • Posts: 33783
  • Hell yes!
Re: Stats. Giggity!
« Reply #313: August 28, 2014, 04:56:29 PM »
Jason Parks is leaving Baseball Prospectus for a scouting job with the Cubs.

BP's going to have to develop its own talent pipeline, keep getting raided.

Offline dracnal

  • Posts: 1696
Re: Stats. Giggity!
« Reply #314: November 07, 2014, 06:24:27 PM »
Thought this was a fairly interesting article:  http://sportsworld.nbcsports.com/bill-james-statistical-revolution/

It's long so in a nutshell for the TLDR crowd: The guy who sparked Sabrmetrics kind of regrets it because people just use bs stats to cherry pick support for their position.

Offline blue911

  • Posts: 18486
Re: Stats. Giggity!
« Reply #315: November 07, 2014, 06:29:45 PM »
Thought this was a fairly interesting article:  http://sportsworld.nbcsports.com/bill-james-statistical-revolution/

It's long so in a nutshell for the TLDR crowd: The guy who sparked Sabrmetrics kind of regrets it because people just use bs stats to cherry pick support for their position.

Because nobody ever misused counting stats?

Offline welch

  • Posts: 16298
  • The Sweetest Right Handed Swing in 1950s Baseball
Re: Stats. Giggity!
« Reply #316: November 07, 2014, 10:21:39 PM »
The point: something might be true even if it's hard to measure. In baseball stats, the question should be "what does the stat mean?" Does it force a change in pitchers, hitters, fielding position? Does it determine what kind of pitch should be thrown next?

Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #317: November 10, 2014, 10:06:49 AM »
Because nobody ever misused counting stats?
Ryan Howard had a lot of RBIs this year.

Offline Squab

  • Global Moderator
  • ****
  • Posts: 4528
  • me lookin at the bullpen
Re: Stats. Giggity!
« Reply #318: January 13, 2015, 08:10:57 PM »
Fangraphs is hiring a writer or two. Houston you better get on this!
www.fangraphs.com/blogs/fangraphs-now-hiring/

Offline welch

  • Posts: 16298
  • The Sweetest Right Handed Swing in 1950s Baseball
Re: Stats. Giggity!
« Reply #319: January 13, 2015, 08:42:38 PM »
Repeating: how do we know what the derived stats really mean? What amount of difference in WAR, or several other stats-built-on-stats can be justified as a significant difference?

Example: in the early '70s, I administered the Psychiatric Status Schedule (PSS), developed by Spitser, Endicott, and Cohen. They were important people: Bob Spitzer was later the president of the American Psychiatric Association. As lead research assistant (I had quit the University of Chicago, and our big boss thought that meant I knew something. He was wrong, of course.) I gave test after test. Found out later that were were the calibration group. Much later, a friend, who is now a big-deal psychiatrist and psychiatric theorizing, told me that he used to carry our results over to the office at Columbia Presbyterian. "We didn't know what the stuff meant".

I see that WHIP tracks closely to ERA. Therefore, I trust WHIP, although I'm not sure it tells much more than ERA. For instance, I know that Walter Johnson's WHIP takes a staggering leap in 1921 and 1922, and I also remember that he hurt his arm about 1920, the year he pitched his only no-hitter. Then WHIP drops to a respectable 1.2 something until he retires.

How reliable is xFIP? When there is a difference, how much matters and how much is noise?

On WARfare, I've already concluded that it is a fair measurement of roughly how valuable a player was, but very rough.



Offline Lintyfresh85

  • Posts: 35130
  • World Champions!!!
Re: Stats. Giggity!
« Reply #320: January 13, 2015, 09:27:25 PM »
WAR is pretty solid for hitters. Not so much for pitchers.

Offline welch

  • Posts: 16298
  • The Sweetest Right Handed Swing in 1950s Baseball
Re: Stats. Giggity!
« Reply #321: January 13, 2015, 10:21:53 PM »
My gripe about WAR is that it does not punish for strikeouts. I think they are usually the worst batting outcome. Yes, it can be argued that a 12-pitch K is valuable because it tires a pitcher, but the game is still measured by runs scored. The 100-pitch max dates from Paul Richards, who maxed his young pitchers (were they nicknamed the "baby bulls"? Milt Pappas was the first and most famous). It's not obvious that 100 pitches, or 90, or 115 pitches should be the maximum. Seems arbitrary, something like pulling a starting pitcher for Drew Storen with two outs in the ninth if the starter has just walked someone and you "always" go to a closer with one on and two out on the ninth

OPS is interesting, but there should be a measure that counts a K as more than a simple out.

Many people, such as Boswell, claim that BAPIP should hover around .300. I think he used it to argue that Espinosa would show better as son as his BAPIP rose to "normal".

I happened to look up a few players, and my old hero, Harmon Killebrew, never managed a BAPIP of .300. In his first real season, 1959, his BAPIP was about .230, although he tied Mickey Mantle for AL home run champ with 42. Every one of us kids thought that our Killer would someday make the Hall of Fame. Killebrew was rushed back to the majors before he learned to field...the Nats were that desperate and his hitting was that good.

Maybe Killebrew is just a quirk. Frank Howard had a higher BAPIP before Ted Williams taught him "the science of hitting". His BAPIP in 1969 was about .285, even though his BA was about .295. If I could pick players from Washington's past, I would pick Harmon and Hondo, among others. Of the players I would pick, on what I've read, I would pick Sam Rice for CF. His career BAPIP was about .330. That was not luck: he was that good, and so said Bucky Harris, Ty Cobb, and Shirley Povich. I would also pick Cecil Travis, the best player never to get a single vote for the HoF. His career BAPIP was also .330, and it was no fluke. He joined the Army, served throughout WW2, from his feet during the Battle of the Bulge. In 1941, he was about 28, hit .359 (second to Ted Williams, who hit over .400) with a BAPIP of .366. He was getting better, pulling the ball more often. Discounting 1945 - 47, when he played on damaged feet, which cost him his balance at the plate and his speed, his career BAPIP would be even higher. (Yes, from what I've heard and read, Cecil Travis was a better SS than Derek Jeter).

What about xFIP? Agreed, it would be nice to measure a pitcher without considering the fielders, but that should over-value strikeout pitchers. Extreme example would be Strassburg compared to Zimmermann or Fister. Strassburg has spectacular stuff and a great ratio of K's to BB's, but Zimmermann and Fister are more effective pitchers.


Offline houston-nat

  • Posts: 19050
Re: Stats. Giggity!
« Reply #322: January 14, 2015, 11:23:36 PM »
Fangraphs is hiring a writer or two. Houston you better get on this!
www.fangraphs.com/blogs/fangraphs-now-hiring/
applied tonight!

Offline UMDNats

  • Posts: 18063
Re: Stats. Giggity!
« Reply #323: January 15, 2015, 10:37:22 AM »

Many people, such as Boswell, claim that BAPIP should hover around .300. I think he used it to argue that Espinosa would show better as son as his BAPIP rose to "normal".


http://www.fangraphs.com/library/pitching/babip/

"Typically around 30% of all balls in play fall for hits, but there are several variables that can affect BABIP rates for individual players, such as defense, luck, and talent level."

League-wide BABIP is usually around .300, but there are many players who consistently have a BABIP higher or lower than that.

Offline Mattionals

  • Posts: 5738
Re: Stats. Giggity!
« Reply #324: January 15, 2015, 11:28:13 AM »
http://www.fangraphs.com/library/pitching/babip/

"Typically around 30% of all balls in play fall for hits, but there are several variables that can affect BABIP rates for individual players, such as defense, luck, and talent level."

League-wide BABIP is usually around .300, but there are many players who consistently have a BABIP higher or lower than that.

Is there a way to show league average BABIP from year to year?  I would say that with the increase in strikeouts, there would be an increase in BABIP as well.  Basically, a guy from the low strikeout era put more balls in play, yet if batting averages have stayed the same, that would mean that BABIP as a whole should be lower.  In the high strikeout era, less balls are being put in play, but when they are, there is a better chance that the player makes it on base.

Basically to me, MLB has traded poorly hit balls for strikeouts.