You’ve heard the stats by now. On a rainy Wednesday night, Roy Halladay debuted for the 2013 season. And, well, it wasn’t great. Despite nine of the ten outs he made being strike outs, Halladay was roughed up for five runs in just 3.1 innings. Any hitter who didn’t strike out got on base 90 percent of the time against Halladay. Yeah, not good.
The reaction is, which has become almost comical at this point, is the real story here. People who follow the Phillies were already worried about Halladay’s first start before the game was even played, and not just because Cole Hammels got beat on opening day. They said Halladay had a “rough spring” and “lost his cutter” at one point. And this after battling a sore shoulder for most of 2012.
Then, of course, the start didn’t go well, and everyone freaked out. ESPN had a scout analyze Halladay’s outing (with the scout calling Halladay a pitcher who looked like “a journeyman” against the Braves), breathlessly searching for answers for the disaster in Atlanta.
Fangraphs, a site you should always turn to if you’re looking to avoid the have-a-take-fest on ESPN, broke Halladay’s start down beautifully, even pointing out that his arm slot is ever so slightly different from past seasons. The GIF-ery alone in that post makes it worth a read.
But all that analysis and hyperbole still, to borrow from Fangraphs’ title, doesn’t answer the main question: what the hell happened?
The answer, of course, is Halladay had a bad start. But the interesting part is how this one start easily exposed the flaws in both the talking heads’ approach to sports (not that hard to do) and the stat-heads’ approach (a little more difficult).
The talking heads get paid to freak out, so a bad start by a pitcher who has spent the better part of the last ten seasons as a lab-created death machine for hitters is fodder enough to make it through at least one news cycle. Especially since people were freaking out about Halladay’s bad spring before. So when Halladay had a bad start, they got to do their jobs despite this only being one bad start which, of course, Halladay has had before. Just not that often.
Then the stat-heads chimed in with fantastic analysis like the Fangraphs piece I linked earlier. But the problem with reacting to one start, even if that reaction is “well here’s what happened and some peculiarities that came with it” is that they’re both feeding into the talking head-created need to “have a take,” and they’re also ignoring the most important two words in baseball statistical analysis: sample size.
There just isn’t very much analysis that can be done about one start. As handy as Pitch f/x, zone percentage, and even handy gifs showing Halladay’s delivery are, they don’t show any kind of trend or pattern. Just one start. And those talking heads the sabermetrics community has a weird relationship with can just use things like that Fangraphs post as ammo for their staged freakouts.
I myself am either snarking up the whole thing by pointing out the quasi-lame issues involved here or am playing into the talking-head-fest by continuing to talk about Halladay’s start.
And, again, it’s just one start. For there to be any real, usable data about Halladay in 2013, he has to get through at least four more. And that’s IF he doesn’t get hurt or isn’t hurt already. It’s also possible he’s just bad against the Braves.
But “Roy Halladay pitched badly! Everbody panic!” is probably the most ridiculous angle one could possibly have.