Pitcher wins and losses are stupid stats. Let’s start right there. Cards on the table.
That’s not a new stance, of course, and I’m not alone with this opinion. Far from it. It’s just on my mind at the moment after watching a relatively low-scoring slate of games in the big leagues Thursday. And it’s time to make a pledge.
We’ll get to that in a moment. First, let’s look at the recent absurdity of this archaic statistic. Up in Toronto on Thursday, Chris Sale pitched eight brilliant shutout innings against the Blue Jays.
Boston’s newest lefty ace struck out 13 and allowed only four hits. Of his 102 pitches, 80 were strikes. That’s all kinds of ridiculous. According to this bit of research on Baseball-Reference’s Play Index, he’s the first pitcher in the history of baseball to throw at least 80 strikes in 102 or fewer pitches while striking out 13 hitters.
Yeah, maybe that’s a bit of an obscure stat. But it’s just a way to quantify what Sale did in his third start with the Red Sox. He was exceptionally exceptional, in a historical sense.
Boston scored a run in the top of the ninth to snag a 1-0 lead, meaning Sale was in line to get the individual pitcher win, too. But closer Craig Kimbrel, who had been brilliant himself this season, gave up a home run to Kendrys Morales in the bottom of the ninth that tied the game. Boston rallied with three runs in the top of the 10th, meaning Kimbrel “earned” the individual pitcher win.
That makes absolutely no sense, of course.
If we were starting from scratch, there’s zero chance any rational baseball fan would look at Sale’s effort and Kimbrel’s effort and decide that Kimbrel “deserved” the individual pitcher victory.
Right? It would never happen.
So let’s stop talking about pitcher wins and losses, OK? Please? Let’s be better. Baseball statisticians have spent the past few decades giving us better options, and the tide has been turning. Advanced statistics are more a part of the baseball conversation now than ever before. But we haven't wiped out the individual W/L quite yet.
And, look, I’m not under any delusions that these blights of baseball statistics are completely going away. They’ve long been ingrained in baseball history — the 20-win season and 300-win career have been benchmarks forever — because in the early days of the sport, they were considered the best way to evaluate a pitcher’s contribution to a team’s victory. Those days are long gone, though.
Even if you’re not a fan of advanced metrics such as WAR, FIP or DRA (you should be, but whatever), there are better traditional ways to evaluate a pitcher’s value to a team, such as ERA (though that’s flawed, too). Individual pitcher wins/losses might literally be the worst of all options.
Because even if you refuse to accept those newfangled stats, you have to admit there are just too many other things that happen during a baseball game to rationally award a decision to one individual player, starting with this: If a pitcher’s team doesn’t score, he won’t get a “W” for the day. In 1967, for example, Gaylord Perry threw 16 shutout innings against the Reds, but finished with a no-decision because his Giants didn’t win until the 21st inning. Four times in his career, Nolan Ryan threw at least nine full innings, allowed zero earned runs, struck out at least 12 and got a no-decision (including a 15-strike out, 0-walk 10-inning gem in 1990).
And then there’s this one: In 2000, Hall of Famer Pedro Martinez struck out 17 in nine innings, but was saddled with the “L” because the Rays strung together two hits and a stolen base with two outs in the eighth inning. Tampa Bay won that one, 1-0.
But enough of the historical extremes. Let’s just look at the most recent season, using B-R’s invaluable Play Index to show the everyday lunacy of using pitcher wins and losses. In 2016, there were 195 starts in which a pitcher threw at least seven full innings, struck out at least seven and allowed zero or one earned runs and zero or one walks. We can all agree that’s a pretty exceptional start, right?
Of those 195 starts, 40 times that starting pitcher either received a no-decision (31 times) or a loss (nine times). On the other hand, on 29 occasions last year, a starting pitcher threw six or fewer innings and allowed at least five earned runs and still wound up with a win in the stat column.
So, if we’re only using pitcher wins/losses as the guide, on 69 occasions last year, a stat line of 6 IP, 5 ER (or worse) was a more valuable pitching line than 7 IP, 1 ER (or better).
Let’s look at Thursday’s action, the catalyst for this little rant.
In the day’s 11 games, nine starting pitchers lasted at least six full innings and allowed zero or one runs (Sale, Marco Estrada, Ervin Santana, Lance McCullers, Wade Miley, Scott Feldman, Danny Duffy, Andrew Cashner and Clayton Richard). They combined to throw 61 2/3 innings and allow only four earned runs. That’s a combined ERA of 0.78 for the day, which is brilliant pitching, right?
Of those nine, only two received wins (McCullers and Richard); the other seven were left with no-decisions. You know which two starting pitchers earned wins Thursday? Philadelphia's Aaron Nola (five innings, four earned runs) and Milwaukee's Zach Davies (5 1/3 innings, four earned runs).
Nola and Davies had mediocre (at best) performances, but their teams scored a bunch of runs while they were on the mound, so they got wins. Meanwhile, Sale faced Estrada, Miley faced Feldman and Duffy faced Cashner, so none of those pitchers were given wins despite outstanding individual efforts. Santana's Twins led 2-1 when he finished his day, but the bullpen wasted that lead, so he didn't get a win, either.
That’s just not right. Statistics should be rational. They should be based in some type of logic. The problem isn’t that pitcher wins/losses are determined by a flawed set of standards; it’s that pitcher wins/losses shouldn’t exist, period.
Again, I get that pitcher wins/losses aren’t disappearing from official baseball records anytime soon (or ever, probably). That’s OK. We don’t have to wait, though. We can be better, right now. All of us.
Being a rational baseball fan, I recognize the inherent fallacy of awarding wins and losses to individual pitchers. I embrace the more intelligent ways to evaluate the contributions of pitchers, and I pledge to raise the level of discourse in my daily baseball conversation, at my job, in my home, at the ballpark and online. I recognize that ingrained habits are hard to break, but I also recognize the value in helping my fellow baseball fans appreciate the game on a deeper level. If I do reference pitcher wins/losses, I will always include a mention of the flawed nature of this archaic statistic. Finally, I pledge to eliminate pitcher wins/losses from my fantasy league's scoring system. I will be the change I hope to see in others.
I’ve long tried to avoid using pitcher wins/losses in my writing at Sporting News and in my conversations with friends and colleagues. I’m sure there have been occasions when I’ve been lazy in spouting out statistics and used it from time to time, but no longer. I’ve taken the pledge.
Join me, won’t you?