The Cy Young Races Are Up In the Air

Shohei Ohtani and Ronald Acuña Jr. have been the MVP frontrunners for quite some time. They lead the majors in WAR, and to make things nice and easy, there’s no need to specify which kind; they’re the top dogs here at FanGraphs, as well as Baseball Reference and Baseball Prospectus. MLB.com has conducted three MVP polls throughout the season, asking over 40 writers and analysts to fill out a ballot. Ohtani and Acuña won all three, and in the latest iteration, released on July 13, they were unanimous choices. The NL Rookie of the Year is just as cut and dry, with Corbin Carroll the hands-down favorite. Meanwhile, Gunnar Henderson is pulling ahead in the AL, and Josh Jung’s fractured thumb likely takes him out of the race.

Thankfully, at least one of the major awards will provide a compelling race down the stretch. There is no clear-cut frontrunner for Cy Young in the AL or the NL, and if the season ended today, five or six pitchers could earn first-place votes in either league.

Let’s talk about the NL first. Zac Gallen has won all three MLB.com polls for NL Cy Young, but he’s never been close to a unanimous selection. In the latest one, released July 20, he was one of six pitchers to receive a first-place vote. What’s more, he went through a rough patch from June 27 to August 1, posting a 4.67 ERA and 4.55 FIP in seven starts. His ERA rose from 2.84 to 3.41, and his FIP from 2.58 to 3.19. He certainly didn’t destroy his Cy Young case, but he gave other contenders a chance to catch up. Here’s what the race looks like now:

Wheeler holds the lead in WAR here at FanGraphs. Unfortunately, his 3.71 ERA is a major roadblock in his quest for the Cy. In the last ten years, no NL pitcher with an ERA that high has gotten so much as a single vote. Since 2013, three AL pitchers have earned down-ballot votes with an ERA approaching 4.00, but only one — Collin McHugh, who received all of five points in 2015 — earned votes with a higher ERA and a lower ERA+ than Wheeler. It doesn’t help Wheeler’s case that he only recently got his ERA under 4.00. He’s been an ace, but after three straight seasons with an ERA under 3.00, his performance this year has flown under the radar.

Snell is the mirror image of Wheeler. He has a sparkling ERA, and the incredible run he’s been on lately has drummed up plenty of media attention, but his underlying numbers aren’t nearly as impressive. He ranks eighth in WAR and 20th in K/BB rate, with an xERA in the 60th percentile.

Strider, Webb, and Steele all fall somewhere in between, with lower ERAs than Wheeler and better underlying numbers than Snell. Strider leads the league in FIP and xFIP, thanks in large part to his ridiculous 14.44 K/9. Webb has thrown nearly 20 more innings than Strider with a lower ERA, and he still ranks among the top five in FIP, xFIP, xERA, and SIERA. Steele is the only qualified starter besides Snell with an ERA under 3.00, and the advanced metrics like him significantly more. He ranks top five in xERA and FIP, and his HR/9 is the second best in the league.

Finally, Burnes is lurking in the background. I don’t see much of a case for him personally, but he was an All-Star this summer, and he’s a former Cy Young winner; name recognition alone could earn him some support. More to the point, MLB.com gives him the fifth-best odds to take home the hardware, as of August 3. I don’t claim to know much about sports betting, but clearly plenty of people think he belongs in the conversation. He ranks 11th in ERA and 12th in WAR right now, but I suppose if anyone could force his way into the race with a dominant performance down the stretch, it’s Corbin Burnes.

Now, here’s the American League picture:

Cole won the first AL poll, beating out Shane McClanahan and Gray. McClanahan took over first place the following month, with Eovaldi in second and Cole in third. Cole was back on top in the latest iteration, but McClanahan was still right behind him. Eovaldi, Valdez, and Gausman also earned first-place consideration. While Cole leads the league in ERA and innings, he lags behind in most advanced metrics, ranking fifth in xERA, sixth in FIP, seventh in K%-BB%, eighth in xFIP, and ninth in SIERA. In other words, he might be the favorite for the average voter, but some might be inclined to leave him off the ballot altogether.

Eovaldi has similar numbers to Cole — his ERA is a few ticks higher, his FIP is a few ticks lower — but has made four fewer starts due to a forearm strain. His Cy Young case depends heavily on how soon he returns to the field. Gausman is the WAR leader in the AL and has a much more respectable ERA than Wheeler. He also leads the league in FIP, xFIP, and SIERA. Valdez has a similar candidacy; his ERA estimators are slightly worse, but his ERA is slightly better. He doesn’t rack up strikeouts, but he averages half an inning more per start and leads the league in groundball rate. His no-hitter should also work in his favor, as a tiebreaker if nothing else.

Gray is a strong candidate as well, with the fifth-best ERA and second-best FIP among qualified starters. But he’ll fall off the pace if he can’t maintain his minuscule 4.9% HR/FB, and his 31st-percentile hard-hit rate suggests that might be a problem. On the flip side, Eflin and López are Statcast darlings. They’re both long shots in this conversation, especially López and his 4.01 ERA, but they deserve some recognition. Eflin leads the league in xERA, and López ranks second. They also rank second and third, respectively, in Baseball Prospectus WARP. López will need a miracle to decrease his ERA enough to get back in the race, but Eflin is an interesting dark horse to follow. He ranks third in FIP and xFIP and fourth in WAR, and his 3.46 ERA isn’t so high as to disqualify him from consideration.

Finally, I had to mention Bautista. The Cy Young hasn’t gone to a reliever in a long time, and rightfully so, but what Bautista is doing right now is exceptional. He’s given up five earned runs in 48 games. He has struck out more than half the batters he’s faced. He ranks fourth among all AL pitchers in bWAR, and he’s within shouting distance of second place. Since the BBWAA introduced the current voting system (ranked-choice ballot of five), at least one reliever has earned at least one vote in 22 of 26 races. In a year with no clear favorite among starting pitchers, Bautista could earn even more support. It’s doubtful he’ll win, but we could see the best reliever finish since Zack Britton in 2016 (fourth place with five first-place votes).

McClanahan was a top choice in those MLB.com polls, but he never had the advanced metrics to back up his candidacy, and he has fallen off the pace with a 7.36 ERA and 4.86 FIP in his last six starts. It doesn’t help that he’s now on the injured list with forearm tightness. Marcus Stroman is in a similar position in the NL; he got off to a hot start but has a 9.00 ERA in his last seven starts, and now he’s on the IL with a hip injury. Luis Castillo is another name whose Cy Young odds are surprisingly high, but unlike Burnes in the NL race, I just couldn’t come up with any argument to support his candidacy.

So, with all that said, who are the most likely winners in a contest with no clear frontrunner? If history is any indication, as much as it pains me to say it, it’s probably whoever leads the league in ERA. Right now, that’s Cole and Snell. I could write a treatise on the flaws of ERA for evaluating pitchers, and as FanGraphs readers, you probably don’t need me to. Nevertheless, no individual statistic is more predictive of a Cy Young victory. Over the last five seasons, seven of the ten Cy Young winners have led their league in ERA. The other three all ranked second, and none was more than two-tenths of a run behind the leader:

It gets a little murkier the farther back you go, but keep in mind that the voting criteria are always changing. From 2013 to ’16, wins were still a meaningful factor. That’s why Rick Porcello (22–4 in 2016) is the only winner with an ERA over 3.00; similarly, that’s why Scherzer (20–7 in 2016) is the only winner with an ERA ranked worse than fifth:

Results from further down the ballot also point to the major role that ERA plays in Cy Young voting; there’s hardly another way to explain Urías and Alek Manoah finishing in third place last season. And while I’d never argue that WAR should be the be-all, end-all in awards discussions, it is telling that ERA is a better predictor of winning the Cy Young than all three versions of WAR:

To be fair, the pitchers with the lowest ERAs right now won’t necessarily be atop the leaderboards at the end of the season. In theory, the guys with better underlying numbers can hope to close the gap. That said, considering the size of their leads, it’s hard to imagine that either Cole/Eovaldi and Snell/Steele won’t finish on top. Here’s how the candidates might stack up at the end of the year, using Depth Charts projections, FIP, and SIERA to stand in for their rest-of-season ERA, using Depth Charts playing time estimates:

Thus, it’s looking like Cole will finally win his Cy Young; the longer Eovaldi spends on the injured list, the better his chances become. It will be interesting to see, though, if one of the less traditional candidates can beat Cole’s low ERA and famous name.

I’m even more curious to see how things shake out in the NL. Considering their ERAs below 3.00, Steele and Snell might be the frontrunners. At the same time, Gallen, Strider, and Webb have cultivated ace reputations over the last calendar year, and that can play a meaningful role, too. Anecdotally speaking, those three seem to have a better shot than Snell or Steele. Thus, it’s possible we see an NL Cy Young with an ERA in the threes for the first time since Brandon Webb in 2006. That may not sound so exciting, but for an ERA cynic like myself, it’s welcome news indeed.