The unfairness of pitcher wins, 2017 edition

My mom being a Nationals fan, she recently asked me why Gio Gonzalez only has eight wins and five losses even though he has started twenty games. This prompted a minutes-long rant on my part about the fragility and arbitrary nature of pitcher “wins.”

The conversation went something like this: Gio’s wins and losses should total twenty, because in baseball you either win the game or you lose the game. Wins + losses = total, right?

Wrong, for some reason.

Then what constitutes a win for a pitcher? According to the MLB glosssary:

“A pitcher receives a win when he is the pitcher of record when his team takes the lead for good ... A starting pitcher must pitch at least five innings (in a traditional game of nine innings or longer) to qualify for the win.

Let’s break that down. For a starter to get the win, he must go at least five innings and either:

1. Leave with a lead which lasts for the remainder of the game; OR
2. Leave after completing a half-inning, and have his team take a lead which lasts in the next half-inning.

Wins misrepresent the talent level of pitchers for two basic reasons: because they can be foiled by a bullpen, or be stymied by an offense.

The Pen

Because the Nationals bullpen was one of (if not) the worst in the league for the first half of this season, sometimes Gio left the game with a lead and the bullpen blew it. Other times he left when the game was tied. Both of these scenarios result in what is called a “no decision,” which explains those seven missing games. In those cases, he doesn’t get the win but is also not responsible for the loss. Or, he could last only four innings, leave with the surefire lead, and still take a no-decision.

Gonzalez has five losses, but in two of those games he only allowed one run, and those runs were not the deciding factor, thanks to some ugly bullpen work behind him. 8-3 is a much better winning percentage than 8-5. And in four of his seven no-decisions, Gonzalez left the game with a lead, which the bullpen proceeded to cough up. If you take both these factors into account, and give Gonzalez a surefire bullpen, a record of 12-3 looks a heck of a lot better than his current 8-5 record. Same pitcher, different wins and losses, just because of the bullpen he gets to pitch in front of.

The Offense

For a second case study in the nonsensical nature of pitcher wins, I want to move on to the blue-haired gift to humanity that is Carlos Martinez. Just how good has Carlos, AKA light of my baseball life, been this season? In 123 ¹⁄₃ innings, he has struck out 9.71 batters per 9 (9th in the NL), limited opponents to 1.09 homers per nine innings (10th in the NL), and earned a FIP of 3.84 (8th in the NL). And if you like more traditional stats, Martinez also has a 1.14 WHIP, good for 8th in the National League!

Carlos is very good and is a guy you can count on to win ballgames. But you might not think so if you looked just at his wins and losses. He is top-ten in basically every stat, yet his record is 6-8. Why? He is liable on any given night to get screwed by the Cardinals’ lethargic offense, most recently in his Friday start against the Cubs. He went a respectable six innings, and kept the damage done by Chicago’s offense to three runs (two earned). You might have heard about the 8th inning in that game:

But even though the Cardinals went on to win (by a lot), and even though Carlos left the game having allowed just three runs (and came away with a quality start), he was on the line for the loss and not the win. In a game where his team outscored the opponent by seven runs, the starter did not get the win. He pitched two-thirds of the game, but because the offense waited until the 8th to get going, the team’s win does not count toward his total. That’s crazy!

Gio Gonzalez falls victim to this as well. He started a game in St. Louis, and walked away with the loss after giving up only one run on two hits over seven innings of work. The Nats couldn’t score a run during those innings. They scored a run in the top of the ninth, which would have resulted in a tie game and a no-decision had Gonzalez still been the pitcher of record. His one run allowed did not decide the game. That honor goes to the home run a relief pitcher surrendered in the 8th.

Due to the structure of pitcher wins, a loss that should have been pinned on the bullpen is attributed to the starter because the offense only scored after he departed the game. Gonzalez gave up just one run, the Nationals scored a run, and he still got tagged with the loss.

Carlos Martinez has six no-decisions and the Cardinals won three of them. One of those starts netted him a game score of 93, the tenth-highest of the entire season! Yet, once again, he did not get credit for the win, because St. Louis didn’t do any hitting while he was still in the game. If you added those wins, they would bump his record up above .500, to 9-8.

A League-Wide Problem

It’s not just Carlos and Gio experiencing this. There are a lot of good, and potentially great, pitchers who simply don’t get the wins they deserve and are left to bear most of the responsibility for losses that may belong to the offense or the bullpen. More and more people are realizing that pitcher wins aren’t informative or fair, but it’s a slow process, when they’re still shown in every box score and broadcast.

There are better tools available! From the FanGraphs library: “Fielding Independent Pitching (FIP) measures what a player’s ERA would look like over a given period of time if the pitcher were to have experienced league average results on balls in play and league average timing.” Basically, FIP takes into account “results a pitcher can control directly: strikeouts, walks, [hits by pitch], and home runs.” Below is a list of the top qualified NL starters by FIP who either have more than four no-decisions or a losing record:

These are fifteen very good pitchers, regardless of the offense and bullpen backing them up. Four of these guys have losing records, meaning that nearly a third of the NL’s most effective pitchers do not have winning records! That’s the sign of a stat that isn’t doing anyone any good.

And can we please talk about what is happening to Jeff Samardzija? He is the sixth-best NL pitcher by FIP, but has only four wins and is stuck with eleven losses! I knew the Giants were bad, but I did not realize how much of an impact the offense and the bullpen were having on Samardzija’s record. Three of his losses were quality starts (in which he went at least six innings and allowed three or fewer earned runs). He has five no-decisions, all of which were starts where he gave up three or fewer runs. The Giants won three of those five games! Again, it’s a product of the offense not scoring while he’s in the game. Including those three games the team won, 7-11 would probably feel a lot better and be closer to a true picture of the caliber pitcher Samardzija is this season. (Still not an accurate picture, though, because Samardzija has been great, and it’s not his fault the rest of the Giants have been so bad.)

To determine who you want on the mound for your team, do not use a pitcher’s win or loss total. A pitcher can give his team every opportunity to win a game, but it’s all up in the air until that twenty-seventh out, and the pitcher can’t do it all. It does not make sense to evaluate a pitcher’s skill based on a game’s outcome when it also depends heavily on offensive production and bullpen efficiency. You cannot adequately judge a pitcher by his number of wins or losses because of the unfair manner in which they are awarded.

Audrey Stark is a Contributor at Beyond the Box Score. You can follow her on Twitter @highstarksunday.