Two Groups of Aces

Here are two ranked groups of pitchers:

Group A
1. Max Scherzer
2. Chris Sale
3. Corey Kluber
4. Clayton Kershaw
5. Sonny Gray
6. Jake Arrieta
7. Dallas Keuchel
8. David Price
9. Zack Greinke
10. Chris Archer
11. Michael Pineda
12. Jacob deGrom
13. Clay Buchholz
14. AJ Burnett
15. Johnny Cueto
16. Carlos Carrasco
17. Gerrit Cole
18. Lance Lynn
19. Jeff Samardzija
20. Jordan Zimmermann

Group B
1. Zack Greinke
2. Dallas Keuchel
3. Sonny Gray
4. Max Scherzer
5. AJ Burnett
6. Chris Sale
7. David Price
8. Jacob deGrom
9. Hector Santiago
10. Felix Hernandez
11. Yovani Gallardo
12. Jake Arrieta
13. Carlos Martinez
14. Gerrit Cole
15. Shelby Miller
16. Johnny Cueto
17. Kyle Gibson
18. Chris Archer
19. Wei-Yin Chen
20. Ubaldo Jimenez

Which group better represents the best pitchers in Major League Baseball? Group A has the last two years’ Cy Young Award winners (Scherzer, Kluber, and Kershaw twice) in the top four. The guys you may be surprised to see so far up are probably Arrieta, Buchholz, and Carrasco. Group B includes several of the same guys and King Felix. It also has Santiago, Gallardo, Gibson, and Chen, four names I’ve never heard in a best-pitcher conversation. I’d argue pretty strongly that Group A is a close approximation of a Best Pitchers in the Game list. Group B is an amalgamation of superstars, solid pitchers having good years, and a few flukes.

But wait… Group B has 14 guys pitching in this year’s All-Star game, with the game’s starters at the top. Group A has 12 All-Stars, with a guy who missed the cut at #3 and the last replacement from the NL roster at #4. Somebody must think Group B is full of the game’s best pitchers.

If you haven’t already figured this out, Group A is a ranking of the pitchers with the most WAR, per fangraphs, in the first half this year. Group B is a ranking of the pitchers with the best RA9-WAR. Fangraphs’ FIP-based WAR considers only strikeouts, walks, and home runs when evaluating a pitcher’s success over the innings he’s pitched. RA9-WAR is more similar to Baseball-Reference’s version, based on runs allowed while a pitcher is on the mound.

It’s not surprising that the players and managers who pick the All-Star pitchers prefer the guys with the shiny ERAs. In a lot of cases, these guys have the most wins as well, since fewer runs have been given up while they’re on the field. There’s nothing wrong with rewarding low-ERA pitchers. That said, players like Lorenzo Cain and Yadier Molina were selected to the All-Star Game largely because of their role in run prevention, so pitchers won’t be the only players on the field who are skilled at keeping runs off the board. As far as I know, no All-Star voter ever looked for the position players whose teams gave up the fewest runs while they were on the field and anointed them All-Stars without honing in on their specific roles in run prevention. Why, then, do we do just that with pitchers?

I suppose the answer is history. Wins have been around forever and are easy to understand (if not easy to explain). ERA has been along nearly as long and has been considered among the more sophisticated metrics for a long time. FIP, upon which fWAR is based, is rarely shown on a newspaper page or a TV screen. When a player sees that Corey Kluber is 4-10 this year, he assumes Kluber is having a bad year. Kluber’s 3.38 ERA disproves that assumption quickly, but may not suggest greatness. The 2.51 FIP, on the other hand, tells a different story, one in which our protagonist is the same pitcher he was when he won the AL Cy Young last season. Is there any reason to believe Kluber’s ERA tells us more about how he’s pitched than his FIP?

To corroborate my opinion that Group A is a more accurate ranking of the game’s best pitchers, let’s look at each pitcher’s RA9-WAR from last season.

Group A
1. Max Scherzer 5.3
2. Chris Sale 6.2
3. Corey Kluber 7.0
4. Clayton Kershaw 7.9

5. Sonny Gray 4.3
6. Jake Arrieta 4.7
7. Dallas Keuchel 4.9
8. David Price 4.3
9. Zack Greinke 4.4
10. Chris Archer 2.4
11. Michael Pineda 3.1
12. Jacob deGrom 3.5
13. Clay Buchholz -0.5
14. AJ Burnett -0.5
15. Johnny Cueto 7.7
16. Carlos Carrasco 3.3
17. Gerrit Cole 1.7
18. Lance Lynn 4.3
19. Jeff Samardzija 4.1
20. Jordan Zimmermann 4.8

Group B
1. Zack Greinke 4.4
2. Dallas Keuchel 4.9
3. Sonny Gray 4.3
4. Max Scherzer 5.3
5. AJ Burnett -0.5
6. Chris Sale 6.2
7. David Price 4.3
8. Jacob deGrom 3.5
9. Hector Santiago 0.7
10. Felix Hernandez 7.5
11. Yovani Gallardo 2.3

12. Jake Arrieta 4.7
13. Carlos Martinez 0.3
14. Gerrit Cole 1.7
15. Shelby Miller 2.2
16. Johnny Cueto 7.7
17. Kyle Gibson 1.4
18. Chris Archer 2.4
19. Wei-Yin Chen 3.4
20. Ubaldo Jimenez 0.6

A few ways to tackle the question: First, the eight players unique to Group A were worth 34 total RA9-WAR in 2014; the eight unique to Group B were worth 18.4 (all 16 are italicized above). That means, based on the metric at which Group B has excelled this year, the players in Group A but not Group B were almost twice as valuable last year. Of course, we’re dealing with small samples and eliminating 60 percent of each group only makes the samples smaller.

Another method would be to weight the WAR figures, counting the #1 guy 20 times, the #2 guy 19 times, and so on, down to the #20 guy, whose WAR counts just once. Doing this, we get a weighted average WAR of 4.65 for Group A and 3.72 for Group B. That’s a pretty big gap.

It would be specious to try to declare definitively that this data proves that FIP is a better evaluator of pitchers than ERA (or RA9, to be more precise). However, it’s widely acknowledged that FIP has more predictive value, and I would argue that it has more descriptive value as well. Kershaw and Kluber have struck out batters and avoided walks and home runs at elite levels in 2015. Keuchel has fewer strikeouts, more walks, and a likely unsustainable home run rate (.46/9 innings) and Batting Average on Balls in Play (.255). Greinke has fewer strikeouts, an insanely low BABIP (.233, thanks in large part to the Dodgers’ defense), and an unsustainable strand rate (89%). Keuchel and Greinke have been major parts of better run prevention efforts; Kluber and Kershaw have been better pitchers. Scherzer and Sale have been better yet.

Every pitcher in Group B is having a great season. Every pitcher in Group A is pitching extremely well this year too. With due respect to the emerging Carlos Martinez and the resurgent Ubaldo Jimenez, we should all expect bigger things from Group A over the rest of the season. We should probably be watching Kluber and Pineda in the All-Star Game tonight too.

Advertisements
This entry was posted in All-Star Game, Astros, Dodgers, Indians, Nationals, White Sox. Bookmark the permalink.

2 Responses to Two Groups of Aces

  1. Constantine Kostarakis says:

    As always, original analysis, well thought out, and thinking outside the box.
    Keep it coming, it is very much appreciated.

  2. Nice analysis. Now, Clayton Kershaw FIP is at 2.10, leading the majors. Sale still number two, at 2.35.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s