Jump to content
witesoxfan

FAQ and Forum on Advanced Stats

Recommended Posts

To try and stimulate some constructive conversation both now and moving forward, we've decided to open this up for all the advanced stat "nerds" to both answer questions people have regarding advanced statistics and to pose beliefs and write actual blogs (as in web logs) and opinions they have about advanced stats. We just ask that, if you are going to be writing about something that has not been covered yet in the thread, you define the statistic you are writing about or answering a question about in your own words.

 

One final note: ANY NON-CONSTRUCTIVE POSTS WILL BE DELETED WITHOUT FURTHER WARNING. We will allow a lot of leeway here, but be forewarned, we want this to be a thread about creative thinking and understanding regarding statistics in baseball, and we want it to be just that.

 

Any stats that are defined will be added to this post and it will be updated at any time by a moderator.

Share this post


Link to post
Share on other sites

I'll add this resource to kick things off: The FanGraphs Glossary -- the real gems here are the full explanations of how each component of WAR is calculated for hitters and pitchers, which is also where you'll learn the most about things like wOBA and wRC+

Share this post


Link to post
Share on other sites
QUOTE (KyYlE23 @ Jan 16, 2014 -> 09:51 AM)
Alright stat nerds, tell me why fWAR is better than bWAR. Go

 

To try and make this as short as possible, bWAR tells more about past performance, but fWAR tells more about future performances. It's more of a true talent indicator. Like for pitching, fWAR uses Fielding Independent Pitching numbers to figure in the calculation, whereas bWAR uses runs, and team wins. You get more outliers it seems with bWAR than fWAR. I think the large majority of the Sabr community would say that fWAR is the better number to use overall.

Share this post


Link to post
Share on other sites
QUOTE (Chilihead90 @ Jan 16, 2014 -> 10:13 AM)
To try and make this as short as possible, bWAR tells more about past performance, but fWAR tells more about future performances. It's more of a true talent indicator. Like for pitching, fWAR uses Fielding Independent Pitching numbers to figure in the calculation, whereas bWAR uses runs, and team wins. You get more outliers it seems with bWAR than fWAR. I think the large majority of the Sabr community would say that fWAR is the better number to use overall.

 

you dont have to keep it succinct. Expound all you want, that is why this thread was created

Share this post


Link to post
Share on other sites
QUOTE (Jake @ Jan 16, 2014 -> 09:34 AM)
I'll add this resource to kick things off: The FanGraphs Glossary -- the real gems here are the full explanations of how each component of WAR is calculated for hitters and pitchers, which is also where you'll learn the most about things like wOBA and wRC+

 

I wanted to do this but didn't want to just C&P from there. It really is a great resource.

 

Speaking of this, I actually get really frustrated when people say "bWAR is meaningless."

 

To start, WAR is Wins Above Replacement, which is a general stat used to help determine how many more wins a guy contributes compared to John Doe in AAA. In theory, a true replacement level player will put up a WAR of 0, but you often see these players unqualified for even that, and they end up with WARs in the negatives, meaning they are actually costing their team wins.

 

bWAR is baseballreference's version of WAR, while fWAR is FanGraphs version of WAR. bWAR is typically calculated using traditional counting statistics, while fWAR is calculated using more inferred rate statistics. The idea and difference behind these is that bWAR shows the value of what a player did on the field, including value added by the defense, while fWAR shows the value of what a player basically SHOULD do given neutral circumstances. The numbers may not agree 100% with offensive players, but they are pretty close and the differences are typically negligible.

 

Here is the detailed description - http://www.baseball-reference.com/about/war_explained.shtml

 

The difference primarily lies in pitchers. bWAR uses ERA as it's primary determinant while fWAR uses FIP (which is Fielding Independent Pitching, using just numbers the pitcher puts up to determine basically how well he has pitched and removing the defense from the equation. Compared to ERA, this statistic is much steadier over time because it essentially takes "luck" out of the equation). NEITHER OF THEM IS WRONG. It's really all in what you are looking to determine.

 

To me, these two answer two different but similar questions

 

bWAR - "how good was he?"

fWAR - "how good should he have been?"

 

The classic example is Javier Vazquez. In his time with the White Sox - '06-'08 - Vazquez put up bWARs of 2.7, 6.2, and 3.1 for a total of 12 bWAR. During those same years, his fWAR was 4.9, 5.0, and 4.9. fWAR says he pitched "very well" in all 3 years, while bWAR says he was only average to above average during 2 years (in either statistic, 2 WAR is average). However, try and tell any avid White Sox fan that Javy Vazquez was very good in all 3 years, and you will get an ear full.

 

The bottom line is this: WAR - both bWAR and fWAR - are not the end all, be all of statistics. The difference between 2.6 WAR and 2.3 WAR in either is negligible. The idea and concept of WAR is to allow a person the opportunity to compare the actual value of a defensive minded shortstop to a home run hitting first basemen to a run of the mill starting pitcher to a lights out closer. It's a statistic that's always evolving and will continue to evolve. However, it does a terrific job in quantifying the value of players, compared to the league average and compared to players all time, which is why it's so prevalent today. Use either one, but know why you are using them and for what you are using them. Just be wary that if you say "Jim Slim has a bWAR of 3 and it's only July, he's incredible," people will dispute your point stating that he's been pretty lucky with leaving runners on base and a BABIP (batting average on balls in play, which we'll get to eventually) of .250 and that he is due for regression (moving back towards his career norms) and has been pitching over his head. They are also probably right.

Edited by witesoxfan

Share this post


Link to post
Share on other sites

In a futuresox thread someone said it is easier for a SS to get to something like 4.0 WAR so it isn't as impressive. Is WAR measured against your replacement infielder or replacement position? Is SS just so much easier to create dramatic improvement because of the balls in play you have a chance to make an impact on?

Share this post


Link to post
Share on other sites
QUOTE (bmags @ Jan 16, 2014 -> 10:22 AM)
In a futuresox thread someone said it is easier for a SS to get to something like 4.0 WAR so it isn't as impressive. Is WAR measured against your replacement infielder or replacement position? Is SS just so much easier to create dramatic improvement because of the balls in play you have a chance to make an impact on?

 

Yes. There is a replacement level jump that is automatically figured into each position within the equation:

 

Catcher: +12.5 runs (all are per 162 defensive games)

First Base: -12.5 runs

Second Base: +2.5 runs

Third Base: +2.5 runs

Shortstop: +7.5 runs

Left Field: -7.5 runs

Center Field: +2.5 runs

Right Field: -7.5 runs

Designated Hitter: -17.5 runs

 

(For the record, a run in this instance (as I recall) is the difference between the number of runs you personally create and the number of runs you allow.)

 

Roughly 10 runs equals 1 WAR.

 

However, the reason it's "easier" for a shortstop to earn a higher WAR compared to catcher is because shortstops simply affect more plays. They are making strides towards better defining catcher's defensive value (specifically in pitch framing), but, simply put, a shortstop getting/not getting to a ball affects the game far more than a catcher allowing a runner to get to 2B because he couldn't throw him out or allowed a passed ball.

 

Alexei Ramirez, even considering how "bad" he was defensively last year, still saved 12 more runs than he allowed defensively and he was worth 3.1 WAR. Alexei Ramirez was and is still a very solid shortstop, and it really, honestly, is one of the best free agent signings in White Sox history.

 

EDIT: Sorry, that chart is found here. That is part 3 in a (seriously) 15 part series FanGraphs has posted on WAR.

 

http://www.fangraphs.com/library/misc/war/

Edited by witesoxfan

Share this post


Link to post
Share on other sites
QUOTE (witesoxfan @ Jan 16, 2014 -> 04:34 PM)
Yes. There is a replacement level jump that is automatically figured into each position within the equation:

 

Catcher: +12.5 runs (all are per 162 defensive games)

First Base: -12.5 runs

Second Base: +2.5 runs

Third Base: +2.5 runs

Shortstop: +7.5 runs

Left Field: -7.5 runs

Center Field: +2.5 runs

Right Field: -7.5 runs

Designated Hitter: -17.5 runs

 

(For the record, a run in this instance (as I recall) is the difference between the number of runs you personally create and the number of runs you allow.)

 

Roughly 10 runs equals 1 WAR.

 

However, the reason it's "easier" for a shortstop to earn a higher WAR compared to catcher is because shortstops simply affect more plays. They are making strides towards better defining catcher's defensive value (specifically in pitch framing), but, simply put, a shortstop getting/not getting to a ball affects the game far more than a catcher allowing a runner to get to 2B because he couldn't throw him out or allowed a passed ball.

 

Alexei Ramirez, even considering how "bad" he was defensively last year, still saved 12 more runs than he allowed defensively and he was worth 3.1 WAR. Alexei Ramirez was and is still a very solid shortstop, and it really, honestly, is one of the best free agent signings in White Sox history.

 

Makes sense. So even though it's position adjusted, the variance between great and good is much larger at shortstop because they have a chance at more plays.

Share this post


Link to post
Share on other sites
QUOTE (bmags @ Jan 16, 2014 -> 10:36 AM)
Makes sense. So even though it's position adjusted, the variance between great and good is much larger at shortstop because they have a chance at more plays.

 

I don't necessarily think that's true. I think you are going to see the differences in variances no matter what position it is, but what you will see is that it's easier to accrue very high values up the middle than it is on the corners. This is why people have been screaming for Trout to win the MVP the last two years over Cabrera, and why statheads hate the idea of the Angels using Trout in LF.

 

The best two examples I can think of regarding maximizing value in players are Neil Walker and Jed Gyorko. Walker was a catcher coming up but was bad defensively, so they moved him to 3B. His bat simply wasn't good enough to stick there. So they moved him to 2B where he has actually been solid defensively and great - comparatively speaking - offensively, and he has turned into one of Pittsburgh's most valuable players. Gyorko was and still really is a 3B, but the Padres have Headley there - so they moved Gyorko to 2B, and even though he is not good defensively, he creates positive value there because he is creating surplus value there compared to other players in general around the league.

Share this post


Link to post
Share on other sites
QUOTE (witesoxfan @ Jan 16, 2014 -> 10:34 AM)
Yes. There is a replacement level jump that is automatically figured into each position within the equation:

 

Catcher: +12.5 runs (all are per 162 defensive games)

First Base: -12.5 runs

Second Base: +2.5 runs

Third Base: +2.5 runs

Shortstop: +7.5 runs

Left Field: -7.5 runs

Center Field: +2.5 runs

Right Field: -7.5 runs

Designated Hitter: -17.5 runs

 

(For the record, a run in this instance (as I recall) is the difference between the number of runs you personally create and the number of runs you allow.)

 

Roughly 10 runs equals 1 WAR.

 

However, the reason it's "easier" for a shortstop to earn a higher WAR compared to catcher is because shortstops simply affect more plays. They are making strides towards better defining catcher's defensive value (specifically in pitch framing), but, simply put, a shortstop getting/not getting to a ball affects the game far more than a catcher allowing a runner to get to 2B because he couldn't throw him out or allowed a passed ball.

 

Alexei Ramirez, even considering how "bad" he was defensively last year, still saved 12 more runs than he allowed defensively and he was worth 3.1 WAR. Alexei Ramirez was and is still a very solid shortstop, and it really, honestly, is one of the best free agent signings in White Sox history.

 

EDIT: Sorry, that chart is found here. That is part 3 in a (seriously) 15 part series FanGraphs has posted on WAR.

 

http://www.fangraphs.com/library/misc/war/

 

So a catcher essentially has a 2.5 WAR head start over a 1st baseman? (Assuming 162 games, which obviously catchers dont play)

Share this post


Link to post
Share on other sites
QUOTE (LittleHurt05 @ Jan 16, 2014 -> 10:49 AM)
So a catcher essentially has a 2.5 WAR head start over a 1st baseman? (Assuming 162 games, which obviously catchers dont play)

 

Yes, which is an indication that it is much easier to find a guy to play and hit as a 1B than it is a C.

 

To expound on this, in 521 PAs, Justin Smoak hit .238/.334/.412/.746 and was worth 0.2 WAR (he was poor defensively, but not moreso than other 1B around the league). In 475 PAs, Miguel Montero hit .230/.318/.344/.662 and was worth 0.9 WAR (he was solid defensively). Smoak was obviously the far superior hitter, but because he played a position where it is easier to find good offense, his value is actually lower because he was not as good as his contemporaries.

Edited by witesoxfan

Share this post


Link to post
Share on other sites
QUOTE (raBBit @ Jan 16, 2014 -> 10:52 AM)
That's actually a very interesting debate. If Simmons had a wRC+ of 105 but was the best defensive player in the majors where does he rank for MVP voters?

 

He's probably barely top 10, if that, but I'd have him in my top 3-5, depending on the field. Jean Segura put up 7.1 offensive runs this year with a wRC+ of 107. (wRC+ is a weighted statistic meant to help determine overall offensive value, all things considered, and then adjust it to the league average. More reading here). Segura's defensive runs were 5.4. If you assume the same wRC+ with Simmons' defensive value, you are talking about a 6+ WAR player. That is incredible.

Share this post


Link to post
Share on other sites
QUOTE (witesoxfan @ Jan 16, 2014 -> 05:45 PM)
I don't necessarily think that's true. I think you are going to see the differences in variances no matter what position it is, but what you will see is that it's easier to accrue very high values up the middle than it is on the corners. This is why people have been screaming for Trout to win the MVP the last two years over Cabrera, and why statheads hate the idea of the Angels using Trout in LF.

 

I guess I'm confused about this part of it. You know that a CF should have a higher WAR than a 1b, which makes total sense to me. But I'm getting tripped up I guess because it seems to me it's not really weighting shortstops against each other correctly if it's pretty easy for any SS to get to 2.0 WAR. And maybe it isn't and that's just my presumption I haven't looked at where all shortstops fell last year.

Share this post


Link to post
Share on other sites
QUOTE (bmags @ Jan 16, 2014 -> 11:05 AM)
I guess I'm confused about this part of it. You know that a CF should have a higher WAR than a 1b, which makes total sense to me. But I'm getting tripped up I guess because it seems to me it's not really weighting shortstops against each other correctly if it's pretty easy for any SS to get to 2.0 WAR. And maybe it isn't and that's just my presumption I haven't looked at where all shortstops fell last year.

 

I think what you are actually seeing is a trend towards what we saw in the 80s where good defensive shortstops were abundant but good offensive shortstops were on very short supply. Cal Ripken kind of broke that mold back then, but we're back. This is why, if Leury Garcia is as good defensively as advertised, I've had no problem dealing Alexei. Alcides Escobar hit .234/.259/.300 (yes, that's a .559 OPS) with a wRC+ of 49 and, because of his defense and the positional adjustment, he was still worth 1.1 WAR entirely because of how good he was defensively. If he hits .250/.300/.325, he's probably a 2 WAR player.

 

I also think, to some extent, you are also assuming just a bit too much how easy it is. Last year, there were 15 full time shortstops who had WARs of 2 or more. Considering there can only be 30 full time shortstops, and the WAR of an average player is 2, that distribution actually works out pretty perfectly.

 

http://www.fangraphs.com/leaders.aspx?pos=...=&players=0

Share this post


Link to post
Share on other sites
QUOTE (Chilihead90 @ Jan 16, 2014 -> 10:13 AM)
To try and make this as short as possible, bWAR tells more about past performance, but fWAR tells more about future performances. It's more of a true talent indicator. Like for pitching, fWAR uses Fielding Independent Pitching numbers to figure in the calculation, whereas bWAR uses runs, and team wins. You get more outliers it seems with bWAR than fWAR. I think the large majority of the Sabr community would say that fWAR is the better number to use overall.

 

bWAR uses team wins? What do you exactly mean by that?

Share this post


Link to post
Share on other sites
QUOTE (witesoxfan @ Jan 16, 2014 -> 05:16 PM)
I think what you are actually seeing is a trend towards what we saw in the 80s where good defensive shortstops were abundant but good offensive shortstops were on very short supply. Cal Ripken kind of broke that mold back then, but we're back. This is why, if Leury Garcia is as good defensively as advertised, I've had no problem dealing Alexei. Alcides Escobar hit .234/.259/.300 (yes, that's a .559 OPS) with a wRC+ of 49 and, because of his defense and the positional adjustment, he was still worth 1.1 WAR entirely because of how good he was defensively. If he hits .250/.300/.325, he's probably a 2 WAR player.

 

I also think, to some extent, you are also assuming just a bit too much how easy it is. Last year, there were 15 full time shortstops who had WARs of 2 or more. Considering there can only be 30 full time shortstops, and the WAR of an average player is 2, that distribution actually works out pretty perfectly.

 

http://www.fangraphs.com/leaders.aspx?pos=...=&players=0

 

You're last paragraph nailed it. In the original thread (which I had not looked up anything beyond that), someone said Leury would easily be a 2.0 WAR player, which is where this is all coming from.

 

Share this post


Link to post
Share on other sites
QUOTE (bmags @ Jan 16, 2014 -> 11:19 AM)
You're last paragraph nailed it. In the original thread (which I had not looked up anything beyond that), someone said Leury would easily be a 2.0 WAR player, which is where this is all coming from.

 

Yeah, easily isn't exactly right (even if it was me who said it). He has to hit a little bit. If he hits .250/.300/.350, which I certainly think is doable but is absolutely not a given, he's probably a 2 WAR player because of how good he is supposedly is defensively. It was the same thing with Eduardo Escobar when he was here.

Share this post


Link to post
Share on other sites

Another important difference between bWAR and fWAR - defense. bWAR uses Total Zone while FanGraphs uses UZR EDIT: Baseball-Reference switched to DRS from TZ, look to next post for information on DRS

 

http://www.fangraphs.com/blogs/the-fangraphs-uzr-primer/

 

With UZR the amount of credit that the fielder receives on each play, positive (if he makes an out) or negative (if he allows a hit or an ROE), depends on how often that pa,rticular kind of batted ball, in terms of its location, speed and several other factors, is fielded by an average fielder at the same position, measured over a time span of several years, in addition to whether the batted was a hit, out, or error (or FC).

 

...

 

With UZR, if a fielder makes an out, and the UZR engine estimates that it was a difficult ball to field (and turn into an out) by an average fielder at that position, then the fielder will get more credit than if the UZR engine determined that it was an easy ball to field. Likewise, if a batted ball drops for a hit, a fielder will get more negative credit if UZR determined that it was an easy ball to field (for that fielding position) and less negative credit if it was a difficult ball to field. If a fielder makes an error, UZR automatically assumes that it was a relatively easy ball to field, since that is presumably the definition of an error in the first place, so there is no need to incorporate the speed and location of the batted ball and other factors that can influence how difficult a batted ball is to field. In other words, in UZR, errors are treated as balls that are normally fielded by that fielder and that fielder only (the one who made the error), 95% of the time, or whatever the average error rate is for that position and that type of ball.

 

...

 

How many UZR opportunities do you need for UZR to be reliable? There isn’t any magic number. If I asked you how many AB you need before a player’s BA becomes reliable, you would likely answer, “I don’t know. The more the merrier I guess.” That is true with UZR and with all metrics. Of course, for some metrics, you need more or less data than for other metrics for an equivalent reliability. It depends on the sampling error and the spread in underlying talent, and other things that are inherent in that metric. Most of you are familiar with OPS, on base percentage plus slugging average. That is a very reliable metric even after one season of performance, or around 600 PA. In fact, the year-to-year correlation of OPS for full-time players, somewhat of a proxy for reliability, is almost .7. UZR, in contrast, depending on the position, has a year-to-year correlation of around .5. So a year of OPS data is roughly equivalent to a year and half to two years of UZR.

 

...

 

One problem that comes up with any metric when you combine years in order to increase sample size and thus reliability, is that a player’s true talent may change from one year to the next, such that you are in some sense adding apples to oranges. We generally handle that by giving more weight to recent years and less weight to more distant years. So keep that in mind when you are looking at multi-year UZR’s.

 

...

 

Now, even though, as I said, to some extent with UZR we are measuring whether a fielder caught a certain “type” (speed, location, etc.) of ball or not, and that measurement is unambiguous, just because a player has a plus UZR does not mean that he necessarily played good defense – the same for a negative UZR. The analogy with BA is, just because a player had a .334 BA does not mean that he hit the ball well. It is entirely possible the only reason that he hit .334 was because he got a lot of bloops and bleeders and most of his hard hit balls dropped for a hit. But, because we can verify that a player did indeed hit .334, we say that a player’s BA is a good record of what actually happened. In fact, we would be better off if we didn’t record his batting performance by using his BA. We would be better off if we made adjustments to that BA based on how often his softly hit balls happened to fall for a hit and how often his hard hit balls were caught, as compared to the averages for those kinds of balls. If we did that, we would be better able to predict that player’s future BA, and we would have a better handle on his true batting talent, wouldn’t we? So we might actually say that so-and-so had a “virtual BA” of .285, even though he had an actual BA of .334, if lots of those .334 hits were lucky ones. And that .285 would likely be closer to the player’s true talent BA and he would be more likely to hit .285 next year than .334, since we don’t expect his good fortune to continue, if indeed it was good fortune and not some skill that our player had.

 

That is exactly what we are doing with UZR! UZR tries to record a player’s likely true talent and estimate his future performance based on the nuances of the batted ball and the player’s response to those nuances. It is not trying to capture exactly what happens on the field according to some arbitrary categories, like most of the offensive metrics (which make no distinction between a lucky ground ball bleeder through the “5-hole” or a clean, line drive base hit to the outfield), even the advanced ones like wOBA or linear weights.

Edited by Jake

Share this post


Link to post
Share on other sites

for comparison, DRS:

 

http://www.fangraphs.com/library/defense/drs/

 

“…as I understand it, the numbers determines (using film study and computer comparisons) how many more or fewer successful plays a defensive player will make than league average. For instance, if a shortstop makes a play that only 24% of shortstops make, he will get .76 of a point (1 full point minus .24). If a shortstop BLOWS a play that 82% of shortstops make, then you subtract .82 of a point. And at the end, you add it all up and get a plus/minus.”

 

...

 

● DRS uses Baseball Info Solutions (BIS) data in calculating its results. It’s important to note that this data is compiled by human scorers, which means that it likely includes some human error. Until FIELDF/x data gets released to the public, we are never going to have wholly accurate defensive data; human error is impossible to avoid when recording fielding locations by hand, no matter how meticulous the scorers. That said, BIS data is still the best, most accurate defensive data available at this time, so just be careful not to overstate claims of a player’s defensive prowess based solely on defensive stats.

 

● DRS is comparable to UZR in terms of methodology (e.g. the use of “zones” for evaluating defensive success rates) and results. There are some slight differences between the two systems (see below), so DRS and UZR will occasionally disagree on how to rate certain players, but they agree more often than they disagree. The differences between the two systems are smaller than they seem at first glance:

 

Both systems have the same goal- estimate a player’s defensive worth in units of “runs”, and both rely on hit location and type data from Baseball Info Solutions. The differences lie in the various adjustments and calculations that are made.

 

For example, Defensive Runs Saved uses a rolling one-year basis for the Plus/Minus system, while Lichtman uses several years of data to determine each play’s difficulty level. Defensive Runs Saved also includes components to measure pitcher and catcher defense.

 

For those wondering, fWAR uses DRS for catchers.

 

A few other notes: DRS does not do adjustments for handedness of batter (and thus fielder position), ballpark (affects outfielders)

Share this post


Link to post
Share on other sites

Just so you guys don't think I'm all about WAR...

 

WAR does have its limitations. WAR falls short sometimes because it doesn't tell you what a player actually did, just how valuable they were. You have to consider the context to make any sense of it.

 

If I said Joe Blow had a WAR of 5.3 and Greg Leg had a WAR of 4.8, you can't tell me anything about either player.

Share this post


Link to post
Share on other sites
QUOTE (witesoxfan @ Jan 16, 2014 -> 02:23 PM)
If I said Joe Blow had a WAR of 5.3 and Greg Leg had a WAR of 4.8, you can't tell me anything about either player.

Their names both rhyme.

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

  • Recently Browsing   0 members

    No registered users viewing this page.

×