ERA vs. xERA, 2022

[Frobby]

As we think about next year’s staff, it’s useful to consider which pitchers had an ERA that differed significantly from their xERA.   Here’s a list of everyone still with the team who threw 40+ innings, ERA first and then xERA.

Lyles 4.42/4.94

Kremer 3.23/4.46

Bradish 4.89/4.49

Wells 4.25/3.78

Watkins 4.70/5.07

Voth 4.34/4.40 (includes time with the Nats)

Bautista 2.19/2.66

Tate 3.05/3.15

Perez 1.41/3.62

Baker 3.49/3.22

Krehbiel 3.90/4.08

Akin 3.20/3.62

Zimmermann 5.99/6.53

Seems like Kremer and Perez are the two who are most obviously due for a big regression.  10 of 13 had an ERA below their xERA, but I think the LF wall and above average defense had something to do with it.

[Jim'sKid26]

“If we, as an analytics community, don't hold up errors as the end-all-be-all of defensive analysis, then why should that be the primary factor in judging whether or not a pitcher's runs were influenced by poor defense?”

https://www.beyondtheboxscore.com/2013/11/13/5080514/the-end-of-era-ra9-sabermetrics

[Frobby]

I find it interesting that Kremer’s xERA is so much worse than his actual.  I can’t say that I got the feeling that Kremer was incredibly lucky when I watched him pitch.   Part of the reason, I think, is sequencing.  xERA won’t take into account the fact that a pitcher did well in RISP situations.   Kremer had a .756 OPSA with the bases empty, but only .626 with runners on base or in RISP situations.   Remember that as a whole, the league pitchers do worse in RISP situations than non-RISP.   So, I’d say Kremer’s low ERA compared to xERA may be partially due to good outcomes on batted balls (.312 wOBA compared to .326 xwOBA), but the fact that he pitched well when runners were on base had more to do with the ERA discrepancy.   Whether Kremer can repeat the feat of pitching better with runners on remains to be seen.   

I’ll also throw out that Kremer’s FIP of 3.80 was a lot closer to his actual ERA than his xERA.   He limited the homers and walks.  
 

[Jim'sKid26]

16 hours ago, Frobby said:

I find it interesting that Kremer’s xERA is so much worse than his actual.  I can’t say that I got the feeling that Kremer was incredibly lucky when I watched him pitch.   Part of the reason, I think, is sequencing.  xERA won’t take into account the fact that a pitcher did well in RISP situations.   Kremer had a .756 OPSA with the bases empty, but only .626 with runners on base or in RISP situations.   Remember that as a whole, the league pitchers do worse in RISP situations than non-RISP.   So, I’d say Kremer’s low ERA compared to xERA may be partially due to good outcomes on batted balls (.312 wOBA compared to .326 xwOBA), but the fact that he pitched well when runners were on base had more to do with the ERA discrepancy.   Whether Kremer can repeat the feat of pitching better with runners on remains to be seen.   

I’ll also throw out that Kremer’s FIP of 3.80 was a lot closer to his actual ERA than his xERA.   He limited the homers and walks.  
 

How relevant is ERA/+ERA for relievers? Most of these guys don't throw nearly as many innings as starters, especially good starters. I just wonder if there are better ways to predict future success for relievers than ERA or the metrics based off of ERA. For example, I like to look at K/9 for relievers. Couple that with BB/9 and you have a pretty good sense for a reliever. But I don't know how well rate stats reflect future success.

[Frobby]

4 minutes ago, Jim'sKid26 said:

How relevant is ERA/+ERA for relievers? Most of these guys don't throw nearly as many innings as starters, especially good starters. I just wonder if there are better ways to predict future success for relievers than ERA or the metrics based off of ERA. For example, I like to look at K/9 for relievers. Couple that with BB/9 and you have a pretty good sense for a reliever. But I don't know how well rate stats reflect future success.

I think there’s a lot more variance in ERA for relievers, partly because the total sample size is smaller, partly because the situations in which they enter and leave the game can vary so much.  Coming into an inning with two outs is very different from coming in at the start of an inning.  And, relievers probably bequeath a higher percentage of their baserunners to the next pitcher, at which point the outcome of whether those runners score is beyond their control.   

[Ohfan67]

11 minutes ago, Jim'sKid26 said:

How relevant is ERA/+ERA for relievers? Most of these guys don't throw nearly as many innings as starters, especially good starters. I just wonder if there are better ways to predict future success for relievers than ERA or the metrics based off of ERA. For example, I like to look at K/9 for relievers. Couple that with BB/9 and you have a pretty good sense for a reliever. But I don't know how well rate stats reflect future success.

ERA and ERA+ are not great predictors of future success for relievers or starters. That's why people have started to focus so much more on k/9 and other metrics that are significantly more correlated with future success. 

[btdart20]

I think that's why teams are looking at 'stuff' more than 'stats' for relivers.  Elias has a bunch of waiver claim notches in the belt looking at one pitch.  If a guy has 'plus stuff' for one pitch, then there's a chance they can develop a pitching plan for him.  Velo sure, but also spin rate, horizontal/vertical movement, tunneling, etc...  We took a few guys with some control issues and basically said "throw it to the middle of the plate and trust your stuff to either miss bats or make weak contact (and that LF wall) to limit damage."  For some guys who don't have pinpoint control, I think that's made a big difference.

[Jim'sKid26]

33 minutes ago, Ohfan67 said:

ERA and ERA+ are not great predictors of future success for relievers or starters. That's why people have started to focus so much more on k/9 and other metrics that are significantly more correlated with future success. 

I wish it were that simple.

“What can be concluded from this piece is how much defensive performance and luck can alter a pitcher’s ERA, and what statistics should be used to predict future performance for pitchers. On average defensive performance and luck account provide about half a run in variation of a starting pitcher’s ERA, and about one run in variation of a relief pitcher’s ERA. Additionally, the statistics that are most effective in predicting future pitching performance are xFIP and SIERA.”

A Brief Analysis of Predictive Pitching Metrics | Community Blog (fangraphs.com)

[NCRaven]

If Perez and Kremer pitch to their xERA’s, they’re still a decent reliever and back of the rotation starter.

[Tony-OH]

21 hours ago, Jim'sKid26 said:

“If we, as an analytics community, don't hold up errors as the end-all-be-all of defensive analysis, then why should that be the primary factor in judging whether or not a pitcher's runs were influenced by poor defense?”

https://www.beyondtheboxscore.com/2013/11/13/5080514/the-end-of-era-ra9-sabermetrics

 

37 minutes ago, Jim'sKid26 said:

I wish it were that simple.

“What can be concluded from this piece is how much defensive performance and luck can alter a pitcher’s ERA, and what statistics should be used to predict future performance for pitchers. On average defensive performance and luck account provide about half a run in variation of a starting pitcher’s ERA, and about one run in variation of a relief pitcher’s ERA. Additionally, the statistics that are most effective in predicting future pitching performance are xFIP and SIERA.”

A Brief Analysis of Predictive Pitching Metrics | Community Blog (fangraphs.com)

Both of the "studies" are old (2013 and 2018). I'm not a huge "predictor" of future success guy when it comes to most stats because there are so many factors that go into these things.

I would like to see a study done on whether a player expected stats in statcast are a predictor for anything in season or for the next season. I like the "x"  stats, but I just don't know how much of a true predictor they are. 

 

[Frobby]

45 minutes ago, Jim'sKid26 said:

I wish it were that simple.

“What can be concluded from this piece is how much defensive performance and luck can alter a pitcher’s ERA, and what statistics should be used to predict future performance for pitchers. On average defensive performance and luck account provide about half a run in variation of a starting pitcher’s ERA, and about one run in variation of a relief pitcher’s ERA. Additionally, the statistics that are most effective in predicting future pitching performance are xFIP and SIERA.”

A Brief Analysis of Predictive Pitching Metrics | Community Blog (fangraphs.com)

I just want to point out that the quoted post is pretty dated (2018).   I think xFIP was created long before the Statcast era.   I don’t know if xFIP outperforms xERA as a predictive measure.  

[Jim'sKid26]

7 minutes ago, Tony-OH said:

 

Both of the "studies" are old (2013 and 2018). I'm not a huge "predictor" of future success guy when it comes to most stats because there are so many factors that go into these things.

I would like to see a study done on whether a player expected stats in statcast are a predictor for anything in season or for the next season. I like the "x"  stats, but I just don't know how much of a true predictor they are. 

 

I freely admit that I am not an expert in this stuff. Just trying to understand how it works and whether there is stuff available to novices like myself to improve my enjoyment of the game. These "fancy stats" can lead you down some serious rabbit holes. 

As an aside, it's rather intriguing to me that a "study" from 2018 would be considered old. I guess I better go change my teeth and my Depends....

[Frobby]

1 minute ago, Jim'sKid26 said:

I freely admit that I am not an expert in this stuff. Just trying to understand how it works and whether there is stuff available to novices like myself to improve my enjoyment of the game. These "fancy stats" can lead you down some serious rabbit holes. 

As an aside, it's rather intriguing to me that a "study" from 2018 would be considered old. I guess I better go change my teeth and my Depends....

For example, xERA was rolled out publicly in 2020.   As I understand it, xFIP looks at fly ball rate and makes general assumptions about how often fly balls become homers, to eliminate the “luck” involved in HR variance.  But xERA is distinguishing batted balls based on exit velocity and launch angle, and so is a more sophisticated tool for distinguishing high quality and low quality contact.  Instinctively, xERA should be more predictive than xFIP, but I don’t know if that’s been studied or proven.  

[Tony-OH]

6 minutes ago, Jim'sKid26 said:

I freely admit that I am not an expert in this stuff. Just trying to understand how it works and whether there is stuff available to novices like myself to improve my enjoyment of the game. These "fancy stats" can lead you down some serious rabbit holes. 

As an aside, it's rather intriguing to me that a "study" from 2018 would be considered old. I guess I better go change my teeth and my Depends....

Well the study did not look into any of the statcast data to include expected stats, EV, barrel %, etc. Personally, I'm more intrigued by the information collected from statcast then anything done baed off pure statistics from the year before.

In case you are not familiar, "Statcast is a state-of-the-art tracking technology, capable of measuring previously unquantifiable aspects of the game. Set up in all 30 Major League ballparks, Statcast collects data using a series of high-resolution optical cameras along with radar equipment. The technology precisely tracks the location and movements of the ball and every player on the field, resulting in an unparalleled amount of information covering everything from the pitcher to the batter to baserunners and defensive players. Visit MLB.com's glossary for more information."

[Ohfan67]

1 hour ago, Jim'sKid26 said:

I wish it were that simple.

“What can be concluded from this piece is how much defensive performance and luck can alter a pitcher’s ERA, and what statistics should be used to predict future performance for pitchers. On average defensive performance and luck account provide about half a run in variation of a starting pitcher’s ERA, and about one run in variation of a relief pitcher’s ERA. Additionally, the statistics that are most effective in predicting future pitching performance are xFIP and SIERA.”

A Brief Analysis of Predictive Pitching Metrics | Community Blog (fangraphs.com)

You wish what were that simple? I wrote that ERA and ERA+ were not great predictors of future success and that's why there's been a move to other metrics. Those other metrics are better correlated with future success. The relationship between those other metrics and future success is not perfect (the correlation coefficient is not 1). They are metrics correlated with future success, not perfect crystal balls.