Leaders in kwERA

Patrick McDermott

Beyond the Box Score presents this year's ERA leaders, based solely on their strikeouts and walks.

When I evaluate a pitcher's skill and/or attempt to project his future performance I always start by first looking at his strikeout and walk numbers.

One of the main reasons I'm such a fan of using Ks and BBs is that when they're combined correctly they become a pretty incredible predictor of future performance (runs, ERA, etc.)

Sabermetrician Tom Tango is known mainly for his book (The Book) and subsequent website, as well as, for being the mind behind All-Star metrics such as, wOBA and FIP.

An ERA estimator, which many of you may not know about, that I'm a big fan of that can also be credited to Tango. It is known as kwERA.

(*editor's note, kwERA was actually the brain child of GuyM, Tango merely helped him develop the metric and continues to promote today).

kwERA is ERA based solely on strikeouts and walks; thus it's similar to FIP, yet it not only ignores all balls in play, but home runs, as well.

Interestingly, kwERA is even simpler than FIP when it comes to Ks and BBs, as they're weighted equally. The formula for this statistic is:

kwERA = 5.40 - (12*((K-BB)/PA)))

Although kwERA is ultra simplistic, more simple than even ERA itself, it is one of the strongest indicators of future ERA performance that we have.

kwERA is merely Ks minus BBs divided by batters faced, scaled to look like ERA. I personally think kwERA is a great tool, unfortunately there are no websites that put this statistic on their player pages or leaderboards.

The wonderful leaderboards at FanGraphs come (sort of) close, by providing us with K% (K/PA) and BB% (BB/PA) for each pitcher, these numbers can then be converted into kwERA fairly easily.

The readers of Beyond the Box Score are in luck, though, as I have compiled a 2012 kwERA leaderboard in this google doc for everyone to use.

A few notes about this leaderboard:

  • The minimum number of innings pitched in 2012 was 100 for individuals to qualify for this leaderboard
  • I modified the kwERA equation to include intentional walks and hit by pitches

For those who don't want to look at the entire leaderboard, I'll discuss some of the differences between the 2012 ERA leaders and 2012 kwERA leaders.

Top 10 in ERA and kwERA

Pitcher

ERA

Pitcher

kwERA

1. Kris Medlen

1.57

1. Stephen Strasburg

2.71

2. Clayton Kershaw

2.53

2. Max Scherzer

2.83

3. David Price

2.56

3. Cliff Lee

2.86

4. Justin Verlander

2.64

4. Marco Estrada

2.97

5. R.A. Dickey

2.73

5. Cole Hamels

3.12

6. Johnny Cueto

2.78

6. Kris Medlen

3.16

7. Matt Cain

2.79

7. Clayton Kershaw

3.18

8. Jered Weaver

2.81

8. Justin Verlander

3.19

9. Kyle Lohse

2.86

9. R.A. Dickey

3.21

10. Gio Gonzalez

2.89

10. Chris Sale

3.21

There isn't too much of a difference between the 2012 leaders in each category.

Medlen, Dickey, Kershaw and Verlander all appear on both lists.

Strasburg, Lee, Hamels and Sale all had ERAs below 3.17; thus, their ERAs agree with kwERA that they were elite in 2012.

The inclusion of Estrada and Scherzer among the kwERA leaders may surprise some, but Estrada was surprisingly good last year and Scherzer struggled through one of the unluckier first halves in baseball which crushed his ERA, despite putting up incredible K and BB numbers.

Price, Cain and Gonzalez all finished in the top-25 for kwERA; thus, even though they have shown the ability to outpitch their peripherals in year's past their K and BB performance was on par, or nearly as good as their ERAs would indicate.

As for Lohse, Weaver and Cueto I'll discuss their kwERA-ERA gaps, later in this piece.

This leaderboard allows us to look at large gaps between kwERA and ERA in an attempt to project which pitchers who could regress either positively (higher ERA than kwERA) or negatively (higher kwERA than ERA) in 2013.

Pitchers who could improve in 2013

Below I listed pitchers (min. 100 IP) who had a kwERA that was at least a full run below their ERA in 2012:

Pitcher

ERA

kwERA

Gap

Jake Arrieta

6.20

3.65

2.55

Philip Humber

6.44

4.41

2.03

Josh Tomlin

6.36

4.57

1.79

Luke Hochevar

5.73

4.30

1.43

Jeff Francis

5.58

4.17

1.41

Chris Volstad

6.31

4.95

1.36

Joe Blanton

4.71

3.39

1.32

Francisco Liriano

5.34

4.04

1.30

Tim Lincecum

5.18

3.95

1.23

Johan Santana

4.85

3.64

1.21

Freddy Garcia

5.20

4.01

1.19

Hector Noesi

5.82

4.66

1.16

Scott Feldman

5.09

3.94

1.15

Ivan Nova

5.02

3.95

1.07

Tommy Hunter

5.45

4.39

1.06

J.A. Happ

4.79

3.73

1.06

Randy Wolf

5.65

4.61

1.04

Felix Doubront

4.86

3.86

1.00

This list may make me look a little crazy for putting stock in kwERA. As much as I've studied batting average on balls in play and defensive independent pitching statistics, sometimes I still have trouble when arguing for lists like this one.

It was an "abysmal" 2012 for a bunch on this list including, Humber, Arrieta, Noesi, Wolf, Lincecum, Tomlin, Volstad and Francis. Thus, to say they'll be a full run or more better in 2013, leads to a good deal of cognitive dissonance for me.

For most of these pitchers, all kwERA is really saying is that their ERAs can't possibly that much worse than the MLB average (4.19 ERA was average for starters) again in 2012, and they're due for some regression to the mean.

The more interesting names on this list are the ones who are projected (by kwERA) to better than average in 2013, after having an ERA higher than the league average in 2012.

Arrieta, a starter for the Orioles has by far the largest gap among the pitchers on this leaderboard. His 3.65 kwERA ranks 31st ahead of names such as Roy Halladay, Matt Garza and Anibal Sanchez, while his actual 6.20 ERA ranks 139th out of the 142 pitchers on the leaderboard.

In 2012, both Arrieta's strikeout rate and his walk rate were better than league average (leading to his above average kwERA), but his batting average on balls in play (.320), home run rate (1.26 per 9) and dreadful strand rate (57.3%) caused his actual ERA to ballon over two and a half runs above what his strikeouts and walks would indicate.

Blanton, Liriano and Feldman were all free agents this off-season.

So far, Feldman and Blanton have been signed to deals, which could indicate that the teams who brought them think they'll regress positively in 2013. Liriano could as well, although he has yet to be signed.

The story of Lincecum's 2012 campaign, at this point, has all but been beaten to death. Questions about Lincecum's release point, velocity and struggles out of the stretch caused many to believe that there was something more than just bad luck that was causing Lincecum's bad results. He even posted the worst strikeout rate and walk rate of his career. Although his strikeout numbers were still very high and offset his below-average walk rate, which resulted in a low kwERA.

In 2011, Nova posted a very pedestrian K-BB%. He was the poster boy for regression his 3.70 ERA was due for serious regression in 2012.

Magically enough, hIs ERA did regress by over a run and a half.

Ironically though, his K-rate improved a great deal, while his walk rate also improved. Thus, leading to a very different projection for Nova in 2013, as he now may be due for positive regression, based on his 2012 peripherals.

Pitchers who could decline in 2013

Below I listed pitchers (min. 100 IP) who had a kwERA that was at least a full run above their ERA in 2012:

Name

ERA

kwERA

Gap

Kris Medlen

1.57

3.16

1.59

Jeremy Hellickson

3.1

4.36

1.26

Ross Detwiler

3.4

4.56

1.16

Kyle Lohse

2.86

3.98

1.12

Matt Harrison

3.29

4.40

1.11

Jered Weaver

2.81

3.89

1.08

Johnny Cueto

2.78

3.85

1.07

Miguel Gonzalez

3.25

4.32

1.07

Tim Hudson

3.62

4.64

1.02

Brandon McCarthy

3.24

4.24

1.00

This list is much shorter than the list of pitchers who could improve by a fair margin.

I think everyone who follows baseball understands that Medlen's ERA is due for regression, only because his 2012 ERA was unbelievably low. Despite significant regression, his kwERA still indicates that his ERA should be one of the best in the game next season.

If you follow sabermetrics at all you easily could have guessed that the Rays' starter Jeremy Hellickson would be on this list. If you follow the author of this piece you also would know the Hellickson is in fact my #Unicorn; thus, I cannot possibly discuss matters of his Ks/BBs and ERA in just one paragraph.

It's become rather apparent that both Hudson and Cueto have the magical ability to always outperform their peripherals, thus their ERAs probably won't regress by a full run, in 2013.

The Angels' ace, Weaver, also may have these abilities, although issues with his future have been brought up at this very site.

Lohse is currently tabbed as the best free agent starter left on the market. I've already gone into great detail on why the team who signs Lohse may not be too happy with the results they see from going forward, and kwERA seems to agree.

All in all, kwERA is a simple tool that I think is rather interesting; thus, I hope the readers are happy that there is now a place to find a leaderbaord for this metric, at least for 2012.

All data comes courtesy of FanGraphs

You can follow Glenn on twitter @Glenn_DuPaul

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. You should read them.

Join Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker