Kershaw or Scherzer? A tough choice awaits Joe Maddon for his All-Star starting pitcher

Clayton Kershaw may not be available to start the All-Star Game, but if he is, should he get the nod over Max Scherzer? It's a close call.

The Diamondbacks may disagree, but Clayton Kershaw was not quite in vintage form on Tuesday night during the Dodgers’ 4–3 win. Working deep into counts, he went above 18 pitches in three separate innings, something that he'd done just two other times this season, and averaged a season-high 4.68 pitches per batter. Despite his relative inefficiency, Kershaw held Arizona hitless for 6 1/3 innings before yielding an infield single to Chris Owings, ending his quest for his second career no-hitter and the majors' second no-hitter this season (also against the Diamondbacks, by Edinson Volquez on June 3). His seven innings of two-hit, two-walk shutout ball with 11 strikeouts read as a vintage performance.

The outing was Kershaw's third straight without allowing an earned run, the sixth such streak in his career and his first since last September 14–24. He's put together four completely scoreless outings in a row twice, including from June 18, 2014 (the date he no-hit the Rockies) through July 4 of that year as part of a personal best 41-inning scoreless streak. An unearned run in the fourth inning of his June 29 outing against the Angels means he's only at 10 consecutive scoreless innings this time around, but his three-start binge nonetheless comes just as the All-Star Game is approaching, and right after MLB Network host/resident stathead Brian Kenny stirred the pot by suggesting that Nationals starter Max Scherzer had overtaken the Dodgers' ace southpaw as the game's best pitcher.

Though he's won three Cy Young awards—the only active pitcher among the nine who have done so—Kershaw has never started the All-Star Game. If the schedule holds, things could remain that way, because he's on turn to start this coming Sunday, which would make him ineligible to pitch in the game at Marlins Park two days later. But if the Dodgers—who are attempting to lighten the 29-year-old southpaw's workload with the goal of him pitching deep into October—decide that he needs an extra bit of rest after coming within one pitch of his season high of 118, he could be an option to start for NL manager Joe Maddon. Thus, it's worth a closer look at the two pitchers' performance as well as those of the rest of the NL and AL fields.

Kershaw leads the NL in wins (13), but that 's a function of his receiving a robust 5.6 runs per game of offensive support. His 123 1/3 innings and 18 starts both lead the pack as well, though Scherzer, with 120 2/3 innings thus far, will likely overtake him in his next turn, which comes Friday. And truth be told, the Nationals' righty has a solid edge when it comes to the rest of this year's key stats (league leads in bold):

	Kershaw	Scherzer
W-L	13–2	10–5
ERA	2.19	1.94
GS	18	17
IP	123.1	120.2
H/9	6.4	5.0
HR/9	1.2	0.9
BB/9	1.6	1.9
K/9	10.7	12.2
K/BB	6.64	6.27
WHIP	0.892	0.771
FIP	3.11	2.61
OPS+	57	34
WAR	3.7	4.6

Kershaw is second to Scherzer in ERA, WAR, hits per nine and WHIP, and he's third in FIP (Stephen Strasburg is second at 3.08). Another National, Gio Gonzalez, is third in ERA (2.77) and WAR (3.5), but there's no statistical reason to bump him ahead of the other two, either on the basis of this year's stats or a larger slice of playing time.

As for the question of Scherzer vs. Kershaw as the best pitcher in baseball, I noted earlier this week that when it comes to All-Star rosters and lineups, I favor recognizing each league's true elites. I'd rather take into account larger swatches of performance than just the last three months rather than relying upon hot first half performances that may prove fleeting. After all, when do similarly sizzling second half performances get their due? This stance probably puts me in the minority, particularly now that the All-Star Game's status as an exhibition has been restored by the decoupling of its result from home field advantage in the World Series, but that's my story and I'm sticking to it.

Mid-Year Report: Aaron Judge shines, Cubs flail in homer-happy first half of 2017 season

So has Scherzer, a two-time Cy Young winner (2013 in the AL, last year in the NL) with a pair of no-hitters under his belt, already overtaken Kershaw? Including last year, the Dodgers' ace has a significant edge in ERA (1.92 to 2.60), ERA+ (210 to 163) and FIP (2.39 to 3.02), though he trails in K rate (32.9% to 31.3%), WAR (10.8 to 9.3) and innings (349 to 272 1/3), the last two because of his 2 1/2 month absence due to a herniated disc last summer. Including 2015, Scherzer's first season after signing with the Nationals, Kershaw has the edge in ERA (2.01 to 2.68), FIP (2.21 to 2.92) and K% (32.5% to 32.0%) but trails in WAR (17.9 to 16.8) and innings (577 2/3 to 505). Incorporate 2014, when Kershaw won his third Cy Young, into the discussion and he has the edge in WAR (24.3 to 23.9) as well as the other stats despite throwing fewer innings (798 to 703 1/3), and likewise if we include 2013, when Schezer won his first Cy Young; there the gap is 32.1 to 30.6 in WAR, 1.92 to 2.83 in ERA, 2.17 to 2.87 in FIP, and 30.6% to 30.4% in K rate.

Thus it's fair to say that the answer to the question of whether Scherzer has overtaken Kershaw rests in where you draw the line and how you handle the rate stat/counting stat split decision—in other words, how much you penalize Kershaw for his injury.

On a 2017-only basis, Scherzer has been the better pitcher of the pair due to the Dodger ace’s occasional gopher woes, and particularly given the timing of Kershaw's injury, it's perfectly reasonable to anoint Mad Max the NL starter in the All-Star Game. As to the larger question, Kershaw has the better rate stats across longer time horizons, and while his absence hurts him when it comes to counting stats, the gap narrows or even disappears if you go by more than numbers from the last year and a half.

Gary Sanchez on Logan Morrison's Home Run Derby diss: "Not my fault he didn't get selected"

As for this year's AL starter, going by first half 2017 stats alone, the choice would appear to boil down to the Red Sox's Chris Sale and the Royals' Jason Vargas. The latter, who made just 12 starts in 2015-16 due to Tommy John surgery, leads the league in WAR (4.4), ERA (2.22) and wins (12), though he's whiffed just 6.6 batters per nine in 101 1/3 innings; he's seventh in FIP at 3.49. Prior this year, Vargas was regarded as nothing more than a solid, back-rotation journeyman, the owner of a career 67-70 record with a 4.18 ERA (94 ERA+) for five teams since 2005 (the Marlins, Mets, Mariners and Angels before his current stop), with just our seasons qualifying or the ERA title.  The 34-year-old lefty hasn't cracked 90 MPH with his fastball since the surgery, but a lower arm slot and a slight move towards the first base side has turned his changeup into a dominant pitch; batters are hitting just .143 with a .213 slugging percentage against it this year, compared to .313/.458 against his four-seam fastball, .290/.500 against his sinker and .232/.290 against his curve.

By comparison, Sale is a perennial Cy Young contender, though he's never finished higher than third. Traded from the White Sox to the Red Sox in December, he's been dominant thus far, leading the league in strikeouts per nine (12.4), hits per nine (6.4), FIP (2.02) and innings (120 2/3) while running second in wins (11, yawn), ERA (2.61) and WAR (3.9). He's whiffed more than twice as many hitters as Vargas (166 to 74) and earlier this year, reeled off his second streak of eight straight starts with at least 10 strikeouts, matching the single-season record he already shared with Pedro Martinez.

The Indians' Corey Kluber, who is currently second with 11.7 strikeouts per nine, third with 3.4 WAR and fourth with a 2.85 ERA, belongs in the discussion, particularly when considering stats beyond this year. Since the start of 2016, he has a 9.8 to 8.8 edge over Sale in WAR, and a minuscule edge in ERA (3.06 to 3.08) that becomes much wider when adjusting for park (154 ERA+ to 137) in 44 fewer innings (347 1/3 to 303 1/3); Sale has the slight edge in FIP (2.96 to 3.03, again negated by park adjustment) and K rate (29.0% to 28.4%).

Tigers righty Michael Fulmer, the 2016 AL Rookie of the year, is worth mention for his ERA (3.12) and ranking in WAR (7.8, third among AL starters in that span); the latter is held back by his not debuting until April 29, 2016, but he's whiffed just 19.7% of batters he's faced. Fellow Tiger Justin Verlander, who to these eyes should have garnered his second Cy last year, is next among AL starters with 7.3 WAR, but has been rocked for a 4.96 ERA this year, bumping his two-season mark to 3.62 and burying any chance for an All-Star start. It's a two-horse race between Kluber and Sale when including 2015 stats (14.0 to 12.1 in WAR) or dialing back to 2014 (21.4 to 18.7).

Based on 2017 numbers alone, I’d go with Sale over Vargas to start the All-Star Game on the basis of how much more work he does himself by missing bats. But for a choice that incorporates multiple years of performance in recognition of being the league’s best, I’d make the call for Kluber. I’d be very surprised, however, if Indians and AL All-Star pilot Terry Francona calls his number, because it would look too much like favoritism. Still, it would be a perfectly defensible choice.