Sabermetrics | Introduction and Discussion

[Slot Machine]

Something that has been bugging me in regards to pitch framing is since you basically are trying to take advantage of s***** umpiring how can you decipher between calls that were influenced by the catcher and calls that weren't? I've seen plenty of close pitches that were well framed and called balls and also pitches that were well of the plate and poorly framed (if at all) and called strikes, I suppose this is the danger of strictly going of an eye test and I'm sure in a big enough sample all this stuff works itself out but still its been bothering me and I haven't thought as highly of pitch framing of late.

[GD]

Threw this into the OP. Very important. So unless it's drastically different between hitter/pitcher, we can't just look at Hard%/BABIP and determine Hutch's BABIP has been higher because he has previously given up harder contact.

 

http://www.hardballtimes.com/offensive-batted-ball-statistics-and-their-optimal-uses/

So how can we use the BIS contact data?

 

• Not for BABIP. This is seriously the wrong data to use if so-and-so has a low BABIP. Don’t say, “But he’s making hard contact (Hard%).” These stats do so very little to predict BABIP—in part because “hard contact” can be deep fly balls, and fly balls have the worst BABIP of all non-infield-pop-ups. And typically, weak or medium contact results in ground balls, and those have a higher BABIP. But ground balls can be hit hard too. Just stay away from BABIP with these stats.
  

 

• For ISO and SLG variations. Is your team’s prized slugger no longer lashing doubles and homers? Check the BIS data. Major fluctuations there might indicate he’s declining. Otherwise, give it some time.
  

 

• And to a degree, wRC+ variations. But a lot goes into a total-offense metric like wRC+. I’d be more inclined to look at a contact rate than a contact strength measurement. Contact is a clearly delineated event. Contact strength has a lot of noise. But in bigger samples, it can be useful. For instance: Nobody has even hit below 100 wRC+ when his Hard% is 35.5 percent or higher. In fact, very few hitters over 33 percent have been bad hitters—as a group, they average a 121 wRC+. Look at this:
  

[TABLE=width: 100%]

[TR]

[TH=bgcolor: #8E001C, colspan: 2, align: center]WRC+ BY HARD-HIT RATE QUARTILE[/TH]

[/TR]

[/TABLE]

[TABLE=class: sortable, width: 100%]

[TR]

[TD=bgcolor: #D4D4D4, align: center]Quartile[/TD]

[TD=bgcolor: #D4D4D4, align: center]wRC+[/TD]

[/TR]

[TR=bgcolor: #ffffff]

[TD=align: center]Max (43.2%)[/TD]

[TD=align: center]118[/TD]

[/TR]

[TR=bgcolor: #ffffff]

[TD=align: center]Q3 (31.4%)[/TD]

[TD=align: center]102[/TD]

[/TR]

[TR=bgcolor: #ffffff]

[TD=align: center]Q2 (27.8%)[/TD]

[TD=align: center]94[/TD]

[/TR]

[TR=bgcolor: #ffffff]

[TD=align: center]Q1 (24%)[/TD]

[TD=align: center]82[/TD]

[/TR]

[/TABLE]

 

• So fellas hitting under 24 percent Hard-rate are probably not doing well. But remember: there’s a lot of volatility here. The standard deviation in that bottom quartile is 13.6—meaning about 68 percent of the data lies between 68 wRC+ and 96 wRC+. It’s a wide swath.
  

What it tells us:

Taken together, these stats can give us a good feel for a hitter’s style—especially when it comes to groundball or flyball tendencies. Andrew Koo found a few years ago that the Oakland Athletics were leaning heavily on flyball hitters—and doing so to great effect at the time. A hitter’s GB/FB ratio might very well inform us how a hitter will perform in given stadiums or against given pitchers. The problem with these data, though, is that we are far to quick to look at line drive percentage and make bigger conclusions.

 

1. We can’t use LD% to rationalize a BABIP. You know, good for Dee Gordon that he is setting a career high in LD% during the 2015 season. That’s no reason to think he can keep his BABIP above .400 or above his career norms. Change “Dee Gordon” to “Starlin Castro” and “2015” to “2014” and we will see why LD% is a fickle master.
   
2. We can’t use LD% to rationalize a wRC+. Yes DJ LeMahieu has an enormous LD%, but he had an even higher rate in 2013—back when he also had a 68 wRC+.
   

[wamco]

As a former pitcher I find fip to be hilarious. No credit given to the pitcher at all when pitching under stress.

[Captain Adama]

As a former pitcher I find fip to be hilarious. No credit given to the pitcher at all when pitching under stress.

 

FIP under high leverage splits?

 

Then again, things like these tend to even out over a large sample.

[GD]

As a former pitcher I find fip to be hilarious. No credit given to the pitcher at all when pitching under stress.

 

ERA doesn't either, lol. That's data, not a metric.

[ValiantJaysFan]

Is there a metric that measures productive at bats? ie if a runner is on second and the batter grounds out but the runner moves to third?

[Slot Machine]

Usually good pitchers start racking up K's when under "stress". Its great to watch, and FIP loves it!

[GD]

Usually good pitchers start racking up K's when under "stress". Its great to watch, and FIP loves it!

 

The good ones, at least!

[BigCecil]

http://m.mlb.com/video/topic/73955164/v466492483/torbos-martin-nabs-castillo-with-81-mph-cannon

 

Statscast on Martins throw Sept 8th. "arguably MLBs strongest C arm"

[RealAccountant]

Carlos Delgado hit 4 homeruns in a game. How many WARs was that 4 homerun game worth. Was it more than 1 win? How can a player be worth more than 1 win in a game.

 

Does WAR maximize at 1 a game?

[KevinGregg]

Carlos Delgado hit 4 homeruns in a game. How many WARs was that 4 homerun game worth. Was it more than 1 win? How can a player be worth more than 1 win in a game.

 

Does WAR maximize at 1 a game?

 

Remember that to find a player's war first you find their runs above replacement then divide that number by about 10. You can easily see how a player can be worth more than 10 runs per game.

[Olerud363]

Remember that to find a player's war first you find their runs above replacement then divide that number by about 10. You can easily see how a player can be worth more than 10 runs per game.

 

Player a) is in centerfield. Moves to left field. Are they just as valuable there??

 

If so why do teams (usually) move all their best defensive outfielder to center??

 

Why does fangraphs have a defensive penalty for left fielders, if the player can contribute just as much in left field??

 

Does the same hold for every position??

 

If you move Ryan Goins to first base is he just as valuable there as middle infield because he is way better then all the other first basemen??

Edited October 31, 2015 by Olerud363

[L54]

Holy f***

[GD]

This has already been explain to you several times

[Olerud363]

This has already been explain to you several times

 

So why not just patiently explain it again?? Put a link to an article. Cut and paste the previous explanation... People who are "confidently" right, patiently and politely explain things. People who aren't confident give one sentence snappy answers.

 

Why does fangraphs use a -7.5 run adjustment for left. Do you think they are wrong on this??

[Olerud363]

This has already been explain to you several times

 

As you go down the defensive spectrum there aren't as many hard plays. A good defensive player won't have as many opportunities to contribute, a bad one won't have as many to do damage.

 

I believe the former conversation was about Dexter Fowler. The argument was that he may be better as a left fielder because he won't do as much damage.

 

A positive defensive player won't contribute as much at an easier position.

 

A negative defensive player won't do as much damage at an easier position.

 

Ben Revere's best season was 2012 according to fangraphs -- he contributed 10 runs as a center fielder. I don't think he could do that as a left fielder.

 

IF a guy is +10 as a centerfielder are they still +10 as a left fielder?? Or is it more like they are neutral as a feft fielder??

 

The way I see it is a great centerfielder, +25, would still contribute in left but it would be more like +15. A horrible center fielder, -20 or something, wouldn't do as much damage in left.

[GD]

People who are "confidently" right, patiently and politely explain things. People who aren't confident give one sentence snappy answers.

 

You have been given links. You have been given full explanations. You are asking us to repeat ourselves until we are concussed from banging our head against the wall. I have literally no idea what to say to you. It's all there and you keep asking the same question that's been repeatedly answered while being increasingly condescending each time. Do some research, sit down and think for 30 minutes, idk, but we've tried and we're just a broken record at this point.

[GD]

IF a guy is +10 as a centerfielder are they still +10 as a left fielder?? Or is it more like they are neutral as a feft fielder??

 

I'm pretty sure NJH gave you the exact answer to this question right before you called him an *******.

[Olerud363]

I'm pretty sure NJH gave you the exact answer to this question right before you called him an *******.

 

"Theoretically, someone switching from CF to LF doesn't actually lose any value because they should be more skilled at their new position, relative to average. Revere is now compared to plugs like Hanley Ramirez and Kyle Schwarber in UZR calculations."

 

He made a pretty general statement. I am asking if the statement holds no matter how good or bad you are?? I guess you're claiming it does, as you won't go into any more detail.

 

OK. So you guys think that a +10 centerfielder will be a +10 left fielder. A +30 center fielder will be a +30 left fielder. A -30 center fielder will be a -30 left fielder??

 

Would Kevin Kiermaier still be +30 as a left fielder??

 

I assume you WOULD NOT put Kyle Schwarber in center. There would be more hard plays, if he's -20 as a left fielder, he'd be -35 or something in center because there are more hard plays that he couldn't make. Or is that incorrect??

 

Does it work the other way?? If Kevin Kiermaier is put in left is he less valuable because there are less hard plays??

 

I guess I have my answer. You guys think it doesn't matter if a guy plays left or center. Same value. However you aren't explaining it, you're just claiming it.

 

It has not been "explained" many times. It has been "claimed" many times. Big difference.

 

Yo Bro 2 WAR center fielder = 2 WAR left fielder!

[Olerud363]

Theoretically, the value doesn't change. That's the assumption of positional adjustments.

 

Tango talks about positional differences often on his blog. Here's one such discussion: http://tangotiger.com/index.php/site/comments/positional-adjustments-any-updates

 

Does the variance change?? Intuitively it seems it should.

 

Great Center Fielder +30 runs, Horrible Center fielder -30

 

You move the two players to left -- they are still an average of zero but..

 

Great Center Fielder (playing left) +20 runs, Horrible Center fielder (playing left) -20

 

Great Center Fielder WAR goes down, cuz he doesn;t have as much opportunity to make hard plays.

 

Horrible Center Fielder WAR goes up, cuz he doesn't have as much opportunity to screw up hard plays.

[Olerud363]

It seems like the theoretical variance should increase as you go to harder positions

 

variance of DH - 0 (no argument, not even Grant please)

 

variance of 1b ??

 

(potential) variance of lf ??

 

(potential) variance of ss ??

 

(potential) variance of cf ??

 

I use the term potential because the low end guys are going to be eliminated before they reach the low end.

 

If Colabello was in center maybe he could put a -80 runs behind Kiermaier. If you put them both in left the difference might only be 60. If you put them at DH it's obviously 0.

[Olerud363]

You have been given links. You have been given full explanations. You are asking us to repeat ourselves until we are concussed from banging our head against the wall. I have literally no idea what to say to you. It's all there and you keep asking the same question that's been repeatedly answered while being increasingly condescending each time. Do some research, sit down and think for 30 minutes, idk, but we've tried and we're just a broken record at this point.

 

You've given no links. Your claim is basically that it doesn't matter if a guy plays left or center.

 

So if you have Kevin Keirmaier and Nelson Cruz, both guys are 5 WAR guys, does it matter which one plays left and which one plays center?? Will it be 10 WAR either way??

[Olerud363]

You've given no links. Your claim is basically that it doesn't matter if a guy plays left or center.

 

So if you have Kevin Keirmaier and Nelson Cruz, both guys are 5 WAR guys, does it matter which one plays left and which one plays center?? Will it be 10 WAR either way??

 

Very interesting. Apparently some questions aren't answered on this thread... What is the purpose of it again??

 

from fangraphs - "In reality, while plate appearances will obviously dovetail pretty closely with defensive chances, it’s not an exact match."

 

Looking at basic range factors there are about 20-25% more plays in center then left...

 

For an average defender it shouldn't matter if they play left of center

 

An above defender should be more valuable in center (because they are getting more chances)

 

A below average defender (Dexter Fowler, as he looked after 2014) should do less damage in left (less chance to screw up)

 

Sorry for using to many words -- the counter point is this

 

YO BRO 2 WAR LEFT FIELDER = 2 WAR CENTER FIELDER

 

The great minds have spoken. This is the YO-BRO thead... show your genius in 8 words or less.... other wise do not speak

[Olerud363]

NJH - the above is drivel (4 words)

 

GD - YO BRO, NJH explained it all (6 words)

[KingKat]

NJH - the above is drivel (4 words)

 

GD - YO BRO, NJH explained it all (6 words)

 

I just want to state this plainly. The quote above is you playing the victim and being a HUGE *******. You have been provided links, you have been provided explanations. When you either failed (or refused) to understand them, you lashed out at everyone LIKE A BIG JERK. Here's another article you can read. http://www.fangraphs.com/blogs/granderson-and-left-field/

 

The CF-playing-a-corner guy doesn’t magically lose some range when shifting to a corner. His physical abilities are the same, and his ability to cover ground to his left and right remains the same. The only difference is that instead of covering two gaps, he’s now covering one gap and one line. (...) The total difference in defensive performance between the two alignments is simply the drop in value in right-center balls caught minus the rise in value in left field line balls caught. We’re talking about a marginal difference on just a handful of balls in play over a full season. We’re talking about fractions of a run.

[nextyear]

Very interesting. Apparently some questions aren't answered on this thread... What is the purpose of it again??

 

from fangraphs - "In reality, while plate appearances will obviously dovetail pretty closely with defensive chances, it’s not an exact match."

 

Looking at basic range factors there are about 20-25% more plays in center then left...

 

For an average defender it shouldn't matter if they play left of center

 

An above defender should be more valuable in center (because they are getting more chances)

 

A below average defender (Dexter Fowler, as he looked after 2014) should do less damage in left (less chance to screw up)

 

Sorry for using to many words -- the counter point is this

 

YO BRO 2 WAR LEFT FIELDER = 2 WAR CENTER FIELDER

 

The great minds have spoken. This is the YO-BRO thead... show your genius in 8 words or less.... other wise do not speak

 

I think I share you frustration; I understand it but I don't accept all of it (but I do accept most of it).

 

I like most of the stats on fangraphs but not WAR or FIP. Anyone who says that they don't accept stats such as WAR will be insulted on this forum, however, they are in good company since the best known Sabermetrics person, Bill James, who developed many of the stats and has written books on the subject, is NOT impressed by the WAR stat; here is what he says - http://mlb.nbcsports.com/2014/10/29/bill-james-not-a-statistical-guru-not-impressed-by-war/

 

The purpose of Sabermetrics is to try to compare baseball players from different teams, different time periods and different positions all on the same level. This is what Bill James tried to do with Sabermetrics; unfortunately some current Sabermetrics enthusiasts are taking it to previously unimagined extremes.

 

What I personally would like to see in the field of Sabermetrics is to simplify the calculations. People who have mathematical, statistical and baseball knowledge can figure out calculations that are just as informative but which can be calculated by anyone; it is just a matter of the right people getting involved. In fact the right people will likely be able to come up with simpler calculations that are better.

 

The WAR stat is not scripture. If people really want to know how much a certain player contributes to a team winning then there are better methods such as examining team wins directly (the W in WAR) instead of using a convoluted equation. One method is factorial analysis that can separate out the effect of one player among 9 players on the field.

[KingKat]

I think I share you frustration; I understand it but I don't accept all of it (but I do accept most of it).

 

I like most of the stats on fangraphs but not WAR or FIP. Anyone who says that they don't accept stats such as WAR will be insulted on this forum, however, they are in good company since the best known Sabermetrics person, Bill James, who developed many of the stats and has written books on the subject, is NOT impressed by the WAR stat; here is what he says - http://mlb.nbcsports.com/2014/10/29/bill-james-not-a-statistical-guru-not-impressed-by-war/

 

The purpose of Sabermetrics is to try to compare baseball players from different teams, different time periods and different positions all on the same level. This is what Bill James tried to do with Sabermetrics; unfortunately some current Sabermetrics enthusiasts are taking it to previously unimagined extremes.

 

What I personally would like to see in the field of Sabermetrics is to simplify the calculations. People who have mathematical, statistical and baseball knowledge can figure out calculations that are just as informative but which can be calculated by anyone; it is just a matter of the right people getting involved. In fact the right people will likely be able to come up with simpler calculations that are better.

 

The WAR stat is not scripture. If people really want to know how much a certain player contributes to a team winning then there are better methods such as examining team wins directly (the W in WAR) instead of using a convoluted equation. One method is factorial analysis that can separate out the effect of one player among 9 players on the field.

 

Bill James is probably not wrong that there are ways that the calculations can be improved but nothing he or you says is going to chance the fact that the correlation between WAR and wins on the field is uncannily strong. The proof is in the proof as they say.

 

http://disciplesofuecker.com/wp-content/uploads/2013/12/Vineet.png

http://disciplesofuecker.com/15418/15418

[nextyear]

Bill James is probably not wrong that there are ways that the calculations can be improved but nothing he or you says is going to chance the fact that the correlation between WAR and wins on the field is uncannily strong. The proof is in the proof as they say.

 

http://disciplesofuecker.com/wp-content/uploads/2013/12/Vineet.png

http://disciplesofuecker.com/15418/15418

 

 

I am not trying to be argumentative, but this link shows a different picture - http://stats.seandolinar.com/predicting-baseball-wins-with-war/. I posted this on page 7 of this thread along with their plot.

 

In any case, by taking team WAR it filters out (averages out) the large variations (or noise) that would be evident if factorial analysis was used to compare individual player WAR with real win contribution.

[KingKat]

I am not trying to be argumentative, but this link shows a different picture - http://stats.seandolinar.com/predicting-baseball-wins-with-war/. I posted this on page 7 of this thread along with their plot.

 

In any case, by taking team WAR it filters out (averages out) the large variations (or noise) that would be evident if factorial analysis was used to compare individual player WAR with real win contribution.

 

That's not particularly argumentative. There's obviously a lot of noise in individual WAR but the fact that it largely seem to average out and correspond to actual wins is a good sign. Means that for the most part it's going in the right direction even if it's always going to be a work in progress. I think most people who use WAR understand its limitations. If you look at the MVP debate this year, I haven't seen too many people argue tooth and nail that it should be Trout over Donaldson because of WAR. It seems like most people agree that the difference between the two could easily be attributed to noise and shouldn't be considered significant.

[Cyborg]

Sounds like you just wan WPA + defense.