The Cy Young Award and the Wins Barrier

With October approaching (I am as surprised as anyone by that fact), awards season is starting to shape up, which means the writers on these fine digital pages are doing their best to break down who should win and who should not. As I quite fortunately find myself to be one of those writers, I have some thoughts, especially as they relate to the Cy Young. Today, my hope is that you will join me on a journey of sorts into some preconceived notions of Cy Young benchmarks and barriers.

First, an introduction to our candidates. In the American League, it looks as if anyone will have a hard time beating David Price, as Sonny Gray’s most recent starts — one of which was the shortest of his young career — have been sub-par, and Dallas Keuchel also had one of the worst starts of his career recently, coughing up nine earned runs to the Texas Rangers. Chris Sale and Chris Archer also merit some consideration, as well.

In the National League, it’s unsurprisingly a battle between permanent fixture Clayton Kershaw, teammate Zack Greinke, and relative newcomer Jake Arrieta. That one is arguably tougher to call, though batted-ball mastermind Tony Blengino tried his hand at it last week.

All of this leads to an abstract thought I had recently. Let’s say we find ourselves with two leading candidates for one league’s Cy Young, both possessing the exact same ERA and WHIP. Maybe one pitcher’s fielding-independent statistics are better (something that potentially had a hand in Corey Kluber’s Cy Young win last season), but the main difference between the two pitchers is a pretty standard measure of “success”: wins. One pitcher plays for a bad team — as is the case with Sonny Gray, for example. One of them plays for an exciting, playoff-bound team — as is the case with Dallas Keuchel.

Whether we like it or not, wins have had a big say in who claims the Cy Young since the award’s creation in 1956. Though that may be changing, the question remains: what has it taken to overcome a lower win total?

For this particular article, we’re going to try to answer that question by looking at individual pitcher win totals, ERA, and WHIP. These are all very traditional measures of success, but they’re probably still the most often-used statistics for judging a player during award season by the writers who have the privilege to vote. We might want to use FIP to decide who wins the Cy, but I’m assuming that isn’t what’s happening in the majority of these voters’ cases, and my interest is in trying to understand their rationale for picking a winner. There are a lot of things that go into winning the Cy Young, so we’re at least going to discuss other factors as the article unfolds.

First, let’s talk about the pool of players. I’ve pulled the past 15 years of Cy Young winners and second-place finishers, calculating their ERA- and WHIP-. For ease, I’ve simply combined the two and created an aggregated stat, ERA/WHIP-. Then I went through and looked for times when pitchers won the award with fewer wins, and how big a gap in ERA/WHIP- there was between that pitcher and the runner-up (who had a higher win total). Then I averaged those instances, weighting each one with the share of the Cy Young vote they received. That last point is key, as there were a range of different confidence levels (in terms of vote share) for winners with a lower win total.

There were 11 different examples of Cy Young winners with lower win totals during the past 15 years, excluding relievers (who I took out of the sample). Here is the entire breakdown of those years. (Note, once again, that ERA/WHIP- represents the average of ERA- and WHIP-.)

Once we weight the average of the all of the Wins/ERAWHIP- above, we arrive at right around six points of ERA/WHIP- per each win (6.1, to be exact). Basically, for every win that a pitcher has lacked between themselves and their main counterpart in past Cy Young races, they’ve had to be around six points better in both ERA- and WHIP- to win the award.

The main issue with this, of course, is that ERA and WHIP don’t exactly tell us the entire story of a pitcher’s performance. I tried overlaying this system onto the current Cy Young races, and it just doesn’t work very well at forecasting — pitchers with low win totals would have to post impossibly low ERAs and WHIPs for the entire season compared to league average.

However, that might be the story: for instance, Corey Kluber has, by his peripheral and fielding independent numbers, been almost the pitcher he was last year — but his win total and ERA more or less eliminate him from the Cy Young race. Not only are the prospective winners of the award races bound to the constraints of largely traditional statistics; every potential candidate is, as well. To even be in the running for a Cy Young, you have to have a certain number of wins and a certain ERA. That’s just the way it is.

According to the system above, the ERA- and WHIP- numbers that certain pitchers (like Sonny Gray and Chris Archer) would have to produce to get over the difference in wins is almost impossible, but that’s close to accurate — their candidacies are knocked so far down by their lack of wins and/or ERA that it makes a serious run at the award improbable at best.

Then, finally, there is a simple and powerful driver of award decisions in baseball that can’t be quantified: narrative. The story behind the season is often as important as the season itself, and the words we use to describe them exist in the void between analytics and on-field performance.

There’s no getting around that, and it’s not an inherently bad thing; the fact that the void exists is part of the continuing allure of the game. Whether you believe that the grey area between the story and the numbers should or should not be quantified, we all enjoy certain aspects of the unquantifiable.

For that reason, David Price will probably win the AL Cy Young, and that’s great. WAR tells us that he probably should, his ERA and WHIP tell us that he probably should, and most importantly, the story does too. Maybe someday everyone will decide to start using xFIP, FIP, and WAR to decide who should win the Cy Young. Maybe not. Until that potential day, we can see in the numbers that winners and candidates have very real barriers to winning, whether it’s their fault or not.