Mike Petriello

Close call: Data shows tight AL Cy outcome makes sense

Verlander came in second to Porcello, 137 points to 132

Close call: Data shows tight AL Cy outcome makes sense

Rick Porcello won the American League Cy Young Award on Wednesday night, and the reaction was, to put it mildly, mixed. Tigers fans were mad, feeling Justin Verlander got robbed because he'd collected 14 first-place votes to Porcello's eight. Kate Upton, Verlander's fiancee, was really mad, particularly at the two Tampa Bay-area writers (one of whom was Bill Chastain of who omitted Verlander entirely. And as's Richard Justice ably explained, there was a strong case to make that the voters simply got this one wrong.

Perhaps they did. But while the feeling of this writer is that Verlander probably had the strongest case, was it really a travesty? Not really. Let's run down a few reasons why this race was always going to be a close one.

Based on advanced stats, Verlander was the likely favorite, but hardly a slam dunk.

You know Wins Above Replacement, right? It's not the end-all be-all, but it is one of the best tools we have at the moment for capturing total value. FanGraphs' version is based on what pitchers can control (strikeouts, walks, home runs), and while that omits quality of batted-ball contact, it does get away from uncontrollable things like defense and bullpens. Let's check out what it said about AL starters.

5.2 WAR (tie: Verlander/Porcello/Chris Sale)
5.1 WAR (Corey Kluber)

Since WAR is an estimate, never really worry too much about tenths of a point, and read this for what it is: a four-way tie, suggesting that if anyone was "robbed," it was probably Sale. How did it get that way? Verlander and Porcello were basically equal in innings pitched (227, 223). Verlander had a huge edge in strikeouts (254; 28.1 percent) over Porcello (189; 21.2 percent), but Porcello walked fewer (3.6 percent to 6.3 percent) and allowed fewer home runs per nine innings (0.93 to 1.19).

This version of WAR doesn't account for ERA, but even if it did, that's essentially a tie, too (3.04 for Verlander, 3.15 for Porcello). It basically comes down to whether you like dominance (Verlander) or limiting damage (Porcello) ... or pitcher wins, where Porcello's 22-4 record surely swayed some votes over Verlander's 16-9, even though that's largely fueled by the best-in-baseball 6.61 runs per start Porcello received as opposed to Verlander's 3.97.

Verlander on health in 2016

Based on what voters value, Verlander was the favorite, but hardly a slam dunk.

We went over this in detail a few days before the announcement, using our Cy Young Prediction Tool, to show that the expected lead Verlander held was so tight that it was basically a tie. The tool we used models voter behavior since 2010, a date chosen because that was the year that the ballot expanded from three names to five, and because it was the year that Felix Hernandez won with a 13-12 record, showing evidence of a change in what voters consider to be meaningful. 

You can read the full methodology behind the tool here or here, but the point is that it's based on weighting traditional stats like innings, strikeouts, ERA and wins, and it's been extremely successful, like when it nailed all 10 names in both leagues in 2015's Cy Young voting -- in the right order.

At the time, the tool showed that Verlander was in the lead with 78 points, just ahead of Porcello's 74 and Kluber's 73, but also that "while Verlander does have a small lead, it's so tight that it's more or less a tie. For example, when Clayton Kershaw won the 2013 National League Cy Young Award in a landslide, taking 29 of 30 first-place votes, the tool had given him 109 points to Adam Wainwright's 82, a large edge. By that perspective, a four-point lead is telling you that there's not much separation, and that this is essentially a three-way tie."

We knew in advance it was going to be close. It ended up being extremely close. That part, at least, should be no surprise. 

Leaving Verlander off the ballot didn't really affect anything.

The most visible part of the voting has been the fact that two voters each left Verlander off their ballots, with one putting Aaron Sanchez on instead and another going with Masahiro Tanaka. It didn't matter. Even a little.

Think about it this way: The stunning part of those choices is that the pitcher who got the most first-place votes by far wasn't considered to be one of the top five pitchers by two voters. But let's say both had put Verlander fifth, as three other voters did. It wouldn't have raised any eyebrows, it would have added two points to his total … and he'd still have lost in total points, 137-134.

Let's go a step further and imagine a ballot in which both voters had put Verlander fourth, as four other voters did (plus the three who had him fifth). No one would have called out such a choice, not with pitchers like Porcello, Kluber, Sale and Zach Britton in the mix. That would have added four points to his total … and he'd still have lost in total points, 137-136. It was a bad look, yet one that had less impact than you'd think.

Mike Petriello is an analyst for and the host of the Statcast podcast. This story was not subject to the approval of Major League Baseball or its clubs.