April 8, 2016
Players Prefer Presentation
Cole Hamels, and The Win's Long Con
The prevailing narrative surrounding Felix Hernandez is that he has been one of the great tough-luck pitchers of the modern era. King Felix. Felix the Great. Felix the Strong. Felix the Perpetually Let Down. Burdened with great talent and Mariners’ offenses that prominently featured Endy Chavez, the King’s reign would be recognized by many fans for the games during which Felix would pitch seven innings, give up one run, then sit helplessly on the bench as his team failed to do… much of anything.
Opening Day’s 3-2 loss to the Texas Rangers seemed like another chapter in that sad story. The plot was different (one hit, five walks, two unearned runs) and the cast had changed, but the result was a variation on a familiar theme. Felix might have been the only Opening Day starter in the last 100 years to lose after allowing just one hit while throwing six or more innings, a novel “accomplishment” unlike any other he’s had, but he still lost.
It’s one of those cases where the narrative feels right, and we turn to the record books to see if the facts support our instinct. Felix Hernandez certainly has been rather unlucky, or at least in dire need of additional run support in quality performances. He leads all active players in games in which the starter earned a no-decision and earned a Game Score of 70 or more, with seven such games. The Mariners miraculously won four out of those, so the important win was recorded, even if a pitcher win wasn’t. But the great surprise was that the toughest-luck active player since 1985 wasn’t Felix Hernandez. It was his opponent on Opening Day, Cole Hamels.
For one night, Hamels was blessed, but that hasn't been the norm. Since 1985, Hamels leads all active starters in games in which he has earned a Game Score of 60 or more and got a losing decision. The same holds when you up the Game Score to 70. In other words, Cole Hamels has lost more games that he by every right should have won than any active pitcher. With 10 such games, he’s tied for third place with Bret Saberhagen and Pedro Martinez, who had 371 and 409 career regular season starts, respectively. Hamels has started only 307. Randy Johnson, who leads the pack with 15, started 603 games! Greg Maddux comes in second with 12, and he started 740! Poor Cole Hamels!
In fact, when we look at the top 20 active leaders for such losing decisions, we see that while Felix has taken his share, he isn’t even in second place. And while Hamels is in first, nobody is even close to his heels.
This isn’t a perfect measure of tough-luck. To check another, we dusted off that old BP classic Support-Neutral Win-Loss (SNWL) record, developed by Michael Wolverton for this site. You can read his original paper on SNWL here, and his 2004 article on the same here. As he described it, “The idea is to measure starting pitchers' performance on the familiar and understandable W/L scale, but without the distortions of run support and bullpen support that plague the traditional W/L record.” To put it simply, “A starter's SNWL record is his expected (in the statistical sense) W/L record—how many games he would be expected to win and lose given his pitching performances, assuming he had a league average offense and bullpen behind him.”
We haven't calculated SNWL for a while, and we haven't calculated it with runs allowed—as opposed to Fair Run Average, which is now retired—in even longer. But Rob McQuown made it happen for us, going back to 1985 and limiting it to all pitchers with at least 250 starts.
Before we show you the table, know one thing: Support Neutral Wins tend to run a little low. Our group of starters had 12 percent more wins than SNW, collectively; so an individual who had 11 percent more wins than SNW could be deemed, relative to his peers, "unlucky." Meanwhile, a pitcher who had 13 percent more wins than SNW could be deemed fortunate.
The stars of our show are, indeed, less lucky than their average opponent.
We don't talk about Hamels the same way we talk about Felix. Hamels won a World Series, a Series in which he pitched very well and was even the MVP. (Perhaps the ultimate tough-luck move, Felix hasn't yet pitched in the postseason.) He has pitched almost as many seasons as Felix. And while Felix has undoubtedly been better over that stretch, their career numbers aren’t so dissimilar so as to obviously merit Hall of Fame plans for one and nods of “Yeah, he’s really, very good” for the other.
(Baseball-Reference puts them even closer--just five wins of separation.)
When we talk about Hamels, he doesn’t get the future Hall of Famer treatment, at least not with the same consistency—either in the mainstream or around these parts. Hamels' comment in this year's Baseball Prospectus Annual made the direct comparison: “In no way is Hamels in King Felix's class for dominance, nor even in that of Shields in durability, but his fortitude and consistency and near-ace-if-not-quite-ace stuff are underrated assets.” In an NL field dominated by Clayton Kershaw in most years, and rudely featuring guys like former teammate Roy Halladay when it isn’t, Hamels' highest Cy Young finish was fifth in 2011. He was an All-Star in 2007, 2011, and 2012 but hasn’t appeared in the Mid-Summer Classic since. (Felix was a Cy Young in 2010, and came awfully close to another in 2014. He’s been to six All-Star Games.) He hasn’t earned as many dubyas as he ought to have. But when we look at Cole Hamels and compare him to the King, Felix seems to have won just enough more.
But wait, you might say. This is Baseball Prospectus. You, me, and everyone we know are supposed to be smart baseball fans and observers. We don’t rely on pitcher wins. In fact, we actively ignore them. We snark, and smile knowingly. We’re know better, opting for gory math and statistics. We mock Hall of Fame voters who invoke such hokum as pitcher wins. But perhaps we’ve been fooled, at least a bit, by a long con.
Because wwhen we look at the star talismans and bold print signifying All-Star selections, or Cy Young finishes, that litter Baseball-Reference pages, we can’t help but be swayed a bit by the easy, sparkling proxies they become. Wins don’t matter for all voters, and I doubt they are all that matter even to those who do consider them important. But I suspect they matter just enough to just enough voters to be more decisive than we care to admit. And that’s where guys like Hamels’ tough-luck gets tougher. That’s where we suffer the long con. We don’t care about wins in the moment, but those wins work their way into the voting for fun trinkets, and those trinkets dot our perception. The Rangers will likely be a more hospitable home for votes and accolades and relevance than a rebuilding Phillies team was. But when we look back on Hamels, and tally up career achievements, we remain vulnerable to the insidious influence of the win.
Buried under a weird loss for The King is a lesson, wrapped in a Cole Hamels win, dipped in his sadness. Tough-luck reaches out much further than we except. Even very good pitchers are not immune, and even very smart observers can be fooled. And the thing we said we didn’t care about can win out after all.