NL Walter Johnson Award

This is the hard one.

In 2015, a National League pitcher had a 1.66 ERA, the best in either league since Greg Maddux’s 1.63 in an abbreviated 1995 season. The last ERA that low with as many innings pitched as his 222 2/3 was Dwight Gooden’s 1.53 in 1985.

In 2015, a National League pitcher struck out 301 batters, the most in either league since Randy Johnson and Curt Schilling in 2002. The last time someone struck out that many batters in as few innings as his 232 2/3 was Pedro Martinez in 1999.

In 2015, a third pitcher had a 1.77 ERA, the fourth best ERA since he was born in 1986. His second-half ERA of 0.75 was the best since Freddie Schupp’s 0.71 in 1916.

One of these guys has to be third on my ballot.

Here’s a fun exercise: Let’s pretend these three guys didn’t exist. A five-man ballot might look like this:

1. Max Scherzer (2.79 ERA, 2.77 FIP, 276 K, 3 CG w/1 hit or fewer)
2. Gerrit Cole (2.60 ERA, 2.66 FIP, ace of 97-win team)
3. Madison Bumgarner (2.93 ERA, 287 FIP, 5 HR as a batter)
4. Jacob deGrom (2.54 ERA, 2.70 FIP, co-ace of pennant winner)
5. Matt Harvey (2.71 ERA, 3.05 FIP, co-ace of pennant winner)

Even this list neglects great seasons from Jon Lester, Shelby Miller, Tyson Ross, and four Cardinals starters with ERAs of 3.03 or better. In a typical season, this would seem like a deep field, and that’s without the three transcendent pitchers I’m filibustering before debating. This probably says something about the quality of the National League in 2015, particularly of the eight sub-.500 teams, who were probably the eight worst teams in all of baseball. Regardless, the NL is full of aces right now and three of them stood out from this impressive pack.

Back to the good guys. Unless you stumbled upon this little corner of the internet while trying to replace the level in your toolbox, you probably know that the three men whose exploits are boasted above are Zack Greinke, Clayton Kershaw, and Jake Arrieta, respectively. Every one of them deserves the Cy Young Award and the Walter Johnson Award and the Pedro Martinez Award (there must be one of these, right?) and the award that will someday be named after him because of how well he pitched in 2015.

In my post about the American League Walter Johnson Award, I noted the similarities between Dallas Keuchel and David Price, then parsed their numbers a little further and revealed that Keuchel, at least in my estimation, was the better pitcher. I’ll try to dig similarly deep with these three, but I fear that such analysis won’t reavel answers as much as it will expose biases. These guys dominated differently, and depending on how we measure pitching greatness, reasonable people will come to different conclusions as to who was the best.

Alas, here are two methods to try to separate these three pitchers: First, like I did in the AL, I’ll try to assign credit for certain outcomes among pitchers and fielders using fangraphs’ value metrics. Giving full credit for strikeouts, walks, and home runs, which comprise FIP wins, half credit for balls in play, which comprise BIP wins, and quarter credit for stranding runners, or LOB wins, the leaderboard looks like this:

Kershaw 8.675
Arrieta 8.525
Greinke 7.9
Scherzer 6.8
deGrom 5.525

All three top candidates look better by this method than they do if you simply look at fWAR and assume everything other than strikeouts, walks, and home runs should be attributed to defense and luck. They all had tremendous success on balls in play, from Kershaw’s .281 opponent BABIP (which translates to 1 additional win) to Arrieta’s .246 (2.9 wins) to Greinke’s hard-to-fathom .229 (4 wins, the most since Catfish Hunter’s 4.7 in 1975). All three are renowned for their defense, so at least a bit of that credit has to go to them. LOB wins were neutral to Greinke, though he stranded an impressive 86.5% of baserunners, while both Arrieta (80%) and Kershaw (78.3%) score negative runs by fangraphs’ LOB-wins metric, whether the result of sequencing luck or just randomness.

For what it’s worth, Baseball Reference saw the defense behind the two Dodgers pitchers as basically neutral (.02 runs), while Arrieta’s fielders were a little better (.09). Greinke led the trio, and all of baseball, in Defensive Runs Saved with nine, to Arrieta’s six and Kershaw’s five. These guys are good.

It’s hard to look at a 1.66-ERA season and claim the pitcher was lucky, but Greinke’s season appears to be a freak convergence of dominant pitching, excellent fielding from the pitcher and his defense, a lot of balls in play finding gloves rather than grass, and a very small percentage of baserunners finding their way home. Kershaw, in contrast, blew hitters away, but on the rare occasions when someone put the ball in play against him, he surrendered the occasional run, whether due to a loss of focus or to the random bunching of hits and outs.

A pitcher’s job is to put his team in position to win, but it’s well documented that wins and losses are the result of far more than a pitcher’s performance. Offense is half the game, and even on the defensive side, fielders deserve a significant portion of the credit for run prevention. One method of measuring dominance is Game Score, a Bill James-designed metric incorporating how long a pitcher lasts in a game, how many hits, walks, and runs he allows, and how many outs come by strikeout. While the hit and run components are fielding-dependent, much of what Game Score measures is the pitcher’s contribution to team success- getting outs (better if by strikeout), saving the bullpen, keeping runners off the bases.

What does Game Score think of our three protagonists? Here are their 2015 averages:

Kershaw 67.97
Arrieta 67.24
Greinke 67.06

That didn’t solve much. Let’s dig deeper.

According to this SABR piece by Jeff Angus from 2007, a team wins 79% of its games when its starter’s Game Score is within two points of 70. A team loses 81% of its games when the starter’s Game Score is within two points of 29. If we consider a Game Score of 70 or more an almost-guaranteed Win, 29 or worse an almost-guaranteed loss, and anything in between a toss-up, here’s how these three pitchers look:

(Guaranteed Wins-Tossups-Guaranteed Losses)
Kershaw 13-21-0
Greinke 12-21-0
Arrieta 17-17-0

Now we might have something. Arrieta practically guaranteed a win for the Cubs in half of his starts, and put them in a reasonable position to win the other half. Neither Kershaw nor Greinke laid an egg all year (each had a nadir of 33), but they both trailed Arrieta by a handful of dominant starts.

Let’s lower the bar just a bit, to 65 for a Win and 35 for a Loss:

Kershaw 18-14-2
Greinke 22-10-1
Arrieta 20-14-0

There’s Mr. Consistent again. Greinke topped 65 in two thirds of his starts and only once failed to break 35. In his worst start, he gave up five runs on ten hits in six innings. In Colorado. Arrieta looks great again, striking out seven is his worst start, a 5 1/3-inning loss at St. Louis in which he gave up five runs (four earned) on nine hits. Kershaw’s two rough starts early in the season are exposed, and he falls just short of the others in dominant outings.

Each of these methods, of course, has its faults. Game Score may assign too much credit for run prevention to the pitcher, just as fWAR almost certainly unfairly absolves the pitcher for BABIP and strand rate. Any hybrid approach to balancing DIPS theory and straight run prevention depends heavily on how much each side is weighted. We know Kershaw was the best in the league at striking guys out and not walking them (he always is). We know Greinke was the best at keeping runs off the board (though Arrieta was really close while pitching his home games in a more hitter-friendly environment). How much we weight those two measures of success is a matter of personal preference.

A vote for Arrieta seems to me to be a cop-out. “I can’t put all my eggs in the run prevention basket because of all the randomness and I can’t care entirely about strikeouts and walks because the other stuff matters too, so I’ll go with the guy in the middle”. But that’s not fair to Arrieta, who had a ridiculously good season by any measure. If Arrieta wins the Walter Johnson or the Cy Young or both, he’ll absolutely deserve it. It will only feel like a travesty because of the two Dodger aces who won’t win.

Maybe all this comes down to is avoiding that travesty. If Jake Arrieta doesn’t win, it will be because another guy had a better ERA or because another guy had a better FIP. If Zack Greinke doesn’t win the award, we’ll look back and wonder how that happened to the guy with the best ERA since Maddux. If Clayton Kershaw doesn’t win, we’ll look back and wonder how that happened to the guy who struck out 300 when we never thought anyone else would do that again.

It’s taken me a day and a half to write this post, and at various points, each guy has had be convinced that he was the best of the three. This is it. No more stalling. No more equivocating. Here’s my ballot.

(reads the whole piece again)

(consults fangraphs one more time)

(calls his mother. cries a little)

1. Jake Arrieta
2. Clayton Kershaw
3. Zack Greinke
4. Max Scherzer
5. Gerrit Cole

What a travesty.

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

1 Response to NL Walter Johnson Award

  1. Pingback: Rookies, Managers, and Relievers | Replacement Level Baseball Blog

Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s