We talk a lot about hits and luck generally, but not enough about hit sequencing.
Let's assume that the best barometer of a pitcher's ERA is his overall batting average allowed (the larger sample) and not the relatively meager sample of the batting average allowed by a pitcher when there are runners in scoring position. Yet the smaller sample in this case is likely going to have a greater influence on ERA. Is this bad luck? Is it bad "clutch" pitching (assuming such a thing even exists)? Or is it the lack of a skill (pitching out of the stretch vs. out of the windup)? I'll investigate these questions in a bit, but first let's note the pitchers who are way better with runners in scoring position than overall and those who are way worse (through Sunday).
Here are the positive outliers, meaning way tougher/luckier/more clutch with runners in scoring position than they are overall:
The samples do overlap, meaning that the pitchers' overall batting average allowed includes their average with runners in scoring position, too. The average ERA here is 3.30, outstanding. That's no surprise. The question is whether this is a skill (and thus likely to last) or just random luck (unsustainable).
We have a lot of data on Beckett. Has he shown the clutch gene before? Prior to this year, .245 with runners in scoring position vs. .245 overall. That seems incredible, but I double-checked, and it's true.
What about Darvish? He's allowed a .190 average with RISP prior to this year vs. 207 overall. Maybe he got a lot more clutch this year or made an adjustment with his delivery out of the stretch, though it's still hard to believe he'd be better out of the stretch than the windup. Still, it's possible. But how likely is it?
Hamels is a another guy we have a lot of data on and for his career prior to this season, he's .233 with RISP vs. .237 overall. See the above points with Darvish. Rinse, repeat. I'm picking these guys at random, people.
Let's try Wainwright. He was .306 with RISP last year. For his career, not including this year, .242. Overall average allowed for his career: .247. These last three guys do seem to be a little clutch, yes. But very little.
I'm inclined to fade all these clutch aces going forward. And don't say that I'm saying to cut them. I'm only saying the future ERA of pitchers who are benefiting from this effect seems likely to be significantly higher going forward. And don't wonder too much about Estrada. Every hitter he faces is in scoring position because he's Mr. Home Run Derby.
Now here are the 21 unluckiest (or maybe unclutch) pitchers measured this way (or maybe guys who just can't pitch out of the stretch). But you pitch a lot out of the stretch with just a runner on first, too.
Maybe Quintana and Wheeler could be elite in ERA going forward if it's just bad luck on batting average allowed with runners in scoring position.
But what gets me is the average ERA of this group is barely worse (3.60) than the average of the group in which it's the reverse (way higher average overall than with RISP). And how does Niese have a sub-3.00 ERA with a .300-plus average allowed with RISP?
Maybe this can be selectively applied to a class of disappointing pitchers as a reason for their struggles. Look at Verlander and Dickey, whose ERAs do seem hurt by this effect, especially Verlander's. The Tigers righty has a .238 career average allowed with RISP, including .211, .200 and .219 from 2011-13. And we do have splits with no one on base in which Verlander is .258 this year, higher than his career average from 2005-13 (.227) but way better than with runners on (.308). Is Verlander's problem having the extra gas you need to maintain stuff/velocity out of the stretch? I haven't heard that anywhere. But it looks plausible. Of course, Verlander is a problem just begging for a solution and this seems too convenient to me.
If you want to know what's happened to Dickey since his Cy Young 2012, here you go. Average allowed with RISP by year: .177, .280 and this year's .329. Compare that to his average allowed overall: .226, .242, .248 (this year), barely any effect. Why would a knuckleballer struggle so much out of the stretch? I know he counts on his harder knuckler, but it seems pretty random to me.