Last year, the NL Cy Young Award went unanimously to Miami¡¯s Sandy Alcantara. It¡¯s safe to say: That won¡¯t be happening again. Not for Alcantara, out with an arm injury after a disappointing season, and certainly not unanimously, because this year¡¯s race looks like it might be one of the most difficult to sort out in years.

There are, by our unofficial count, seven different candidates with a serious case to be made with about two weeks to go: Zac Gallen (D-backs), Kodai Senga (Mets), Blake Snell (Padres), Justin Steele (Cubs), Spencer Strider (Braves), Logan Webb (Giants) and Zack Wheeler (Phillies).

(That¡¯s right, Senga. Didn¡¯t think about him, did you? But as MLB.com¡¯s David Adler convincingly wrote, Senga and his 2.95 ERA has a quietly solid case himself, and his season looks a whole lot more like Snell¡¯s than you think it does, though he likely still won¡¯t finish in the top three here.)

No one has separated themselves from the pack. How you value them might just come down to what you value. What makes for a great starting pitcher, anyway?

It¡¯s a question, really, about what the voters will care about. Let¡¯s see how they might pick this group apart.

1. Do you care about not allowing earned runs?

You should, right? That¡¯s the entire point of pitching, isn¡¯t it? Keep runs off the board. Then your vote goes to:

• Snell (2.43 ERA)
• Steele (2.73 ERA)
• Senga (2.95 ERA)

Those three are the only three with an ERA under three ¨C the only ones close, really ¨C and so the case is closed. We have our top three, and maybe just top two if you don¡¯t think Senga has the momentum the others do.

This might actually be what happens at the end of the season, because voters do value ERA. No one has won the Cy with an ERA above 3.00 since 2016, and that year, when Rick Porcello narrowly won with a 22-4 record, might have been the last vote of the ¡°pitcher wins matter¡± era. Wheeler (3.70) and Strider (3.73) would have the highest ERA ever for winners.

Of course, if it was just ERA, then Julio Ur¨ªas (2.16) would have beaten Alcantara last year (2.25), and Hyun-Jin Ryu (2.32) would have beaten Jacob deGrom (2.43) in 2019, and Clayton Kershaw (2.31) would have beaten Max Scherzer (2.51) in 2017, and Kershaw (1.69) would have beaten Scherzer (2.28) in 2016, and Zack Greinke (1.66) would have beaten Jake Arrieta (1.77) in 2015. But none of that ended up happening, did it?

Voters care deeply about ERA, understandably ¨C but not exclusively, because it¡¯s affected by ballpark, team, and defense, as well as not telling you anything about durability. It¡¯s not just the ¡°lowest ERA¡± award. It never has been.

2. Do you care about not allowing runs?

If it seems like there¡¯s a great deal of overlap with Question No. 1 above, there is. But because Steele has allowed 12 unearned runs this year while Snell has allowed only two, there have been plenty of runs on the board against the Cubs that haven¡¯t counted against Steele's ERA. If we just look at runs allowed per nine, then it¡¯s ¡­

• Snell (2.53 RA/9)
• And absolutely no one else

Snell leads here by a huge margin. No one else is under 3.00.

That's because Steele has allowed 3.24 per nine, Senga 3.37, and everyone else is north of that. Now: How did Steele get knocked back nearly 70 points?

Despite a generally strong Cubs defense, there have been a number of miscues behind Steele that caused ensuing runs to be scored as unearned. Steele himself is responsible for two of those unearned runs, when he committed errors attempting to field bunts against the Reds on May 26 and the Pirates on Aug. 24. There was also the bizarre play in Miami on Apr. 30 where the error was charged to catcher Yan Gomes on a throw that just as easily could have been charged to Steele.

But mostly, there's what happened on July 16 against the Red Sox, when Steele allowed five runs in the fourth inning, thanks mostly to Masataka Yoshida¡¯s grand slam. For weeks, those runs were all listed as earned. Later on, the inning-opening ¡°double¡± that went through Nico Hoerner was changed to an error, meaning all five runs went from earned to unearned ¨C which dropped his ERA by 41 points.

Should the Cy come down to how the official scorer decided ¨C and later was overruled ¨C on how to score this play?

We argue it should not. So that hands the award to Snell, right? It might. But ¡­

3. Do you care about pitching innings?

This was a large part ¨C perhaps the largest ¨C part of Alcantara¡¯s unanimous case in 2022. In a world where starters are not expected to work deep into games, the rare pitchers who do so should be valued, right? If you agree with that sentiment, then you¡¯re looking at:

• Webb (201 IP)
• Gallen (192.2)

No one else among our list of contenders is above 175 innings. It¡¯s a similar story if you like pitchers who work deep into games, because if we look at most NL starts of at least 7 innings, then:

• Webb (15)
• Gallen (11)

Wheeler has eight, Strider has six, Steele four and Snell just two. If you care about consistency, then maybe it¡¯s Gallen, who was voted first in each of the first four monthly MLB.com Cy Young polls. If you just care about getting outs, lots of them, then Gallen and Webb lead the Majors.

This is maybe the largest knock against Snell and Steele, who lag here. Steele missed several weeks in June with an arm injury, and trails Webb by 36 innings. Snell¡¯s strong year has basically been the opposite of Alcantara¡¯s last year; he has pitched exactly five or six innings in 25 of his 30 starts, and he¡¯s faced fewer batters in the seventh and beyond (seven) than decidedly non-Cy starters such as Dustin May, Joey Lucchesi and Ben Lively.

Another way to think about this is that while Snell¡¯s 2.43 ERA (or 2.53 RA/9) is clearly better than Webb¡¯s 3.45 ERA (or 3.54 RA/9), Webb has pitched the equivalent of nearly four additional complete games more work -- innings that might have inflated Snell's runs allowed if he'd had to continue deeper into games to get them. That has to matter too, right?

4. Do you just care about pure filthy dominance?

Then it¡¯s Strider, who earned a jaw-dropping compliment from first-ballot Hall of Famer (and former Brave) John Smoltz, who said on the Sept. 13 broadcast that, ¡°Strider is so far ahead of any one of us that's ever pitched in an Atlanta Braves uniform.¡± Just look at the top of the strikeout rate leaderboard.

• Strider (37.6%)
• And nobody else.

That would be an impressive list if it was just showing you the top of the highest strikeout rate among our NL Cy contenders (no one else is even at 32%). It can be that if you choose to view it that way, but instead it¡¯s actually this: Strider is currently striking out batters at a higher rate than any National League pitcher (minimum 160 innings) in a season ever. He has the most strikeouts through the first 50 starts of a career of anyone in more than a century, and he hasn¡¯t even made 50 starts yet.

He¡¯s also got a home run problem, and his 3.73 ERA would be the highest by a Cy winner ever, and if we¡¯re using unearned runs to change the way we view Steele, then it¡¯s worth pointing out that Strider is at 3.99 runs allowed per nine, which hardly screams ¡°Cy.¡± It¡¯s not hard at all to see why he¡¯s in the conversation, because ¡°pure filthy dominance¡± is a pretty valuable thing; it¡¯s basically the most valuable thing a pitcher can do on a batter-by-batter basis. But he¡¯s probably not going to win, either.

5. Do you care about your Cy winner not being last in the Majors in something?

We focus, understandably, on all the things a pitcher does well. Does it matter in voting to not select a pitcher who is trailing everyone in something important? Because if so, you¡¯re going to want to look at the highest walk rates among qualified starters ¨C or maybe not.

• Snell (13.4%)
• Josiah Gray (11.5%)
• Senga (11.1%)

Gray isn¡¯t in the Cy mix, but he¡¯s there to illustrate that two of our three names walk a whole ton of batters. Snell doesn¡¯t just have the most walks allowed in baseball (93), he has the most by a large margin (Charlie Morton, at 76, is second), and it¡¯s been more than 60 years since a Cy was given to the pitcher who walked the most batters (Early Wynn¡¯s 1959 win was mostly about having the most, well, wins). Snell might end up with one of the highest walk rates from a qualified pitcher in the last four decades.

This isn¡¯t disqualifying, necessarily. It just colors the fact that while Snell does indeed have the second-lowest average against in the NL (.186), his .298 on-base percentage allowed is more in the good-not-elite range, similar to Braxton Garrett, Andrew Abbott ¨C or Senga.

6. Do you just care about the things we think a pitcher can exert control over?

Now we¡¯re getting a little fancier, by considering FIP (Fielding Independent Pitching), which just looks at strikeouts, walks, home runs, and hit by pitches, or Statcast¡¯s xERA, which does all that but also assigns an expected outcome to every batted ball based on the quality of contact, or Baseball Prospectus¡¯s DRA (Deserved Runs Allowed), which does all that and a whole lot else as well, including park adjustments.

FIP

• Strider (2.83)
• Steele (2.91)

xERA

• Strider (3.03)
• Wheeler (3.19)

DRA

• Strider (3.38)
• Webb (3.41)
• Wheeler (3.57)

This says a whole lot of nice things about Strider¡¯s dominance, doesn¡¯t it? It doesn¡¯t tell you anything about how many innings were thrown. It doesn¡¯t tell you anything about how many runs actually crossed the plate. But by the more advanced ways we have to evaluate pitchers and try to strip away the other effects on runs allowed, Strider stands out, and Wheeler shines here too.

Of course, Strider has that 3.90 ERA. FIP predicts next season¡¯s performance better than ERA does. But we¡¯re not predicting next season here, are we? We¡¯re rewarding this season. On the other hand, ERA isn¡¯t really ¡°what happened,¡± because it excludes unearned runs ¨C as well as relying upon varying levels of ballpark effect and defensive support ¨C and some of our candidates have allowed so many unearned runs that it considerably affects their ERA.

There¡¯s not a right answer on this one. Some voters won¡¯t care about this at all. Others will value it heavily, or at least as a tiebreaker.

7. Do you care about the more advanced metrics that try to put them all together?

Like Wins Above Replacement, which is essentially a marriage of production and playing time. Innings count here, and so do productive innings.

Except there¡¯s three different kinds of WAR.

Baseball-Reference WAR, which mostly functions on runs allowed, says:

• Snell 5.3
• Webb 5.1
• Senga / Steele / Gallen 4.1
• Wheeler 3.8
• Strider 3.4

FanGraphs WAR, which mostly cares about FIP (strikeouts, homers, walks, HBP) and says:

• Wheeler 5.4
• Strider 5.1
• Gallen 4.6
• Steele 4.5
• Webb 4.4
• Snell 3.6
• Senga 3.4

Baseball Prospectus¡¯s WARP, which works off their DRA, says

• Webb 5.0
• Strider 4.5
• Wheeler 4.3
• Gallen 4.1
• Steele 3.5
• Senga 2.3
• Snell 2.2

And now, we¡¯re starting to get a headache.

Snell is the best, a view also echoed by Statcast's pitching runs. Unless Wheeler is. Unless Webb is.

That these three lists are so, so different tells you everything about the different ways there are to value a pitcher ¨C and how voters are going to have an impossible time choosing.

8. What have voters actually valued?

Ultimately, human beings will be making the choices here, and if we can look back, we might be able to look forward, right? We know, based on recent voting, that pitcher wins are hardly considered at all. We know that ERA still matters, but so does WAR. So: What¡¯s happened recently?

MLB.com¡¯s Tom Tango has created a simple predictor model, and the inputs are nothing fancier than innings, earned runs, strikeouts, and wins, weighted by what has correlated to voting success. From 2006 through 2020, it had projected every single Cy winner as being first or second headed into the award.

Using an updated version of the predictor, it says:

• 1st: Snell
• 2nd: Steele
• 3rd: Strider
• 4th: Senga
• 5th: Gallen

Which points to the fact that in voting terms, strikeouts are cool, innings only matter somewhat, and the difference between runs allowed and earned runs allowed probably doesn¡¯t get taken into account by as much as we¡¯d like it to.

9. Will it be the closest race ever?

Maybe! This is actually a harder question to answer than you¡¯d think, because the rules have changed a number of times over the years. For the first 11 years the Cy was handed out, there was just one for the entire Majors, rather than one per league, and for decades, voters picked only their top three, before it was expanded to top five in 2010. That all means that it's difficult to answer this on an apples-to-apples basis.

What we can say is that there has never been an NL Cy race in which six different pitchers got first-place votes. Five different first place votes has happened a number of times, most recently in 2012 (but also 2000, 1987, and 1977). Over in the AL, the absolutely wild 1970 race ended with ¡­ seven different pitchers getting a first. That¡¯s probably not going to happen here.

But we can say that the closest award in total point margin came just two years ago, when Corbin Burnes topped Wheeler by a mere 10 points, in what was an extremely modern Cy discussion, as Burnes dominated in every advanced and rate metric, yet threw 46 1/3 fewer innings. Perhaps this will end up being one of the small handful of times where the winner didn¡¯t even receive the most first-place votes, as happened in 1998 (Tom Glavine) and 2009 (Tim Lincecum).

10. So who is going to win?

In a race this tight, without any standout performances, every single remaining start is going to carry a great deal of importance, as we saw on Friday night, when Steele allowing six earned runs to Arizona changed the face of some of these leaderboards. It seems somewhat likely to us that many voters will look at Snell's run prevention skills -- he's allowed three earned runs or fewer in 21 consecutive starts -- and not really worry about any of the rest of it. He's likely the mild front-runner so long as his remaining starts go well.

But Strider will get some support, too, just due to the strikeouts. Webb might post a surprising case as well, if we're truly to the point where voters will ignore the 10-12 record a poor San Francisco offense has saddled him with, just because all of the innings have helped him climb the WAR leaderboards.?

If you thought the Most Valuable Player award race between Mookie Betts and Ronald Acuna Jr. was going to be a difficult question, then at least in that case there¡¯s only two choices. (Sorry, Matt Olson and Freddie Freeman.) Here, there might be seven. What are voters going to do when ballots are two in two weeks? You might be glad you don't have to answer that question.