It's becoming common thinking, in baseball-writing circles anyway, that wins are a useless stat. Last year, Dayn Perry wrote a column headlined THIS PITCHING STAT IS MEANINGLESS, citing Andy Hawkins’ famous no-hitter loss and that time when Woody Williams gave up nine earned runs in five innings for the win.
Are those examples of times when wins and losses didn’t reflect the pitcher’s performance? Of course they are. They’re great examples. They also don’t prove the rule.
Do home runs count any less when they’re in a park with short fences, or hit off a September call-up? Do stolen bases have asterisks when the catcher takes too long with the throw? Do we consider strength of schedule when calculating a pitcher’s ERA? Of course not. Because over the course of a 162-game season, statistical anomalies have a way of evening out.
We are enamored by perfect games, but they’re statistically silly. Pedro Martinez pitched nine perfect innings before giving up a double in the 10th. Armando Galarraga has his bid for perfection ruined by a bad call. Harvey Haddix took a perfect game into the 13th, before an error destroyed it. Terry Muholland and Jonathan Sanchez lost perfect games to errors. So do perfect games not matter because we can find outlying examples?
Baseball is full of statistics that rely on specific situations. Like RBI. In 2007, Alfonso Soriano slugged .560, hitting 33 home runs and 42 doubles, but drove in just 70 runs. Soriano’s .299 batting average and 97 runs scored helps the Cubs finish first that year. His slugging percentage was good enough for top ten in the National League, as were his home runs, total bases, and extra base hits. Did his lack of RBI make his season any less impressive? Yes. Because RBI is a statistical category that measures clutch hitting, an important part of baseball.
Clutch hitting is almost as important as clutch pitching, and clutch pitching is really what wins measure. They measure a pitcher’s ability to pitch well enough for his team to get the victory. And that is what’s at stake, isn’t it?
This next NFL season, Peyton Manning and Tom Brady might both pass Brett Favre for the all-time NFL wins record. Can a quarterback win with a terrible offense surrounding him, and a terrible defense coughing up more points than he can score? No more so than a pitcher can win with a no-hitter when his team gives up three crucial errors. But if you think that wins don’t matter, look at the winningest quarterbacks of all-time. After Favre, Manning, and Brady come Elway, Marino, Tarkenton, Unitas, and Montana.
Of course, there is the argument of how ridiculous it is when a pitcher does something like, say, blow a four-run lead in the ninth and somehow picks up the win. You know, hypothetically. That can be fixed by making it impossible to get the win unless you pitch five innings – whether you start the game or not. Not every game needs a winning pitcher. If runs can score without RBI, pitchers can end a game without a win.
ERA, WHIP, and quality starts (when that stat is finally universally defined) are clearly more important than wins. And while Ws don’t matter much for a reliever, wins and losses do tell the story of how clutch each starting pitcher was. And over a 162-game season, the five-inning, nine-run wins tend to cancel out the nine-inning, one-run losses.
Are wins the most important statistic for a starting pitcher? Of course not. But to say they’re meaningless is swinging the pendulum too far in the other direction.
