This New York Times blog post by Dan Rosenheck argues that Tim Lincecum’s two-hit shutout Thursday night “was actually both more impressive and more valuable” than Roy Halladay’s no-hitter Wednesday night.
More valuable because Lincecum had to throw a shutout for his team to win 1-0, while Doc had a three-run margin for error in a 4-0 victory. That’s true without being useful. A pitcher has no control, as a pitcher, over how many runs his team scores, though Halladay did have a key RBI single.
But more impressive? Then why is everyone, and I mean everyone, and I mean I bet even Lincecum if you asked him, more impressed by Halladay’s performance than by Lincecum’s? Is the whole world stupid?
No. Rosenheck argues that Lincecum pitched better because he struck out 14 while Halladay only struck out eight. “Extensive research into the subject shows that the vast majority of pitchers wind up giving up hits on about 30 percent of balls in play over the course of their careers,” Rosenheck writes. “As a result, the only ways for most pitchers to reduce the number of hits they allow are to avoid surrendering home runs and to get more strikeouts, so batters never put the ball in play to begin with.”
Defense-independent pitching statistics, which is what Rosenheck is writing about, tell us that variations on a .300 average on balls in play are mostly due to luck, and by Rosenheck’s figuring, “with normal luck, a pitcher with Halladay’s eight strikeouts, one walk, and zero home runs allowed in 28 batters faced would give up an average of 1.55 earned runs per nine innings, while one with Lincecum’s 14 strikeouts, 1 walk, and 0 home runs allowed in 30 batters faced would surrender just 0.37.”
That’s all fine, and looking at a career, or even a season, it’s a useful way to evaluate a pitcher. A whole season’s worth of innings reduces the role of variance — luck — quite a bit, and calculations like Rosenheck’s help us figure out by how much.
But we’re talking about one game here. In a single game, the bad news is that variance can play a huge role, but the good news is that we can see and remember the whole game. And if you saw Halladay’s no-hitter you know that there wasn’t a whole lot of luck involved in the 19 non-strikeout outs he got.
There was one reasonably hard-hit ball, a line drive, not a screamer but a well-hit ball, to shallow right field that Jayson Werth caught. There was one nice play, Jimmy Rollins ranging into the hole at shortstop. Either of those balls could have been hits, yes, but Rollins’ play wasn’t spectacular by any means, and if that ball had gone for a hit, it would have been a good example of luck — a batter getting a hit on a routine but well-placed bouncer.
Lincecum didn’t need a whole lot of flashy leather to get his 13 non-K outs either, but let’s not forget that he gave up two ringing doubles. Those shots, by Omar Infante and Brian McCann, didn’t just scoot humbly between infielders. They were well hit. Halladay didn’t give up anything like those two.
We’re comparing Michelangelo and Rembrandt here. It’s silly. Lincecum was fantastic, Halladay was fantastic. But to use a metric that shows Lincecum was better than Halladay is to find the limits of that metric. Comparing two single games in which almost nothing happened other than outs might be too much to ask of defense independent pitching stats. Sometimes we just have to trust our eyes, and with how impressed we are.