Is Curt Schilling toast?

Though he was a key part of this year’s World Series winner, allowing just one run in 5.1 innings in Game 2, it isn’t clear whether or not Curt Schilling has anything left in the tank. The Red Sox just signed him to a one-year, $8 million deal, but even at that below-market price, questions abound about whether he will be worth it.

After coming within one out of a no-hitter against the Athletics on June 7, Schilling came down with shoulder tendinitis, and went on the disabled list after two poor mid-June starts. He didn’t come back until August, and when he did, Schilling looked like he’d lost some speed off his fastball. For a power pitcher, that’s a pretty big deal.

Schilling looked particularly bad in the ALCS, allowing 15 hits and 7 runs in just 11.2 innings over two starts. Actually, Schilling really only had one bad start in the postseason—Game 2 of the ALCS—but after that, there was a certain apprehension among fans every time Schilling started.

The question is whether or not that worry was warranted. Has Schilling really lost it, or can he still get by? If we look at performance measures, it certainly doesn’t appear that he has. Here is Schilling’s performance in the regular season before and after the injury:

	Wins	Losses	IP	ERA	K/9	BB/9	HR/9	
Before	6	4	94.3	4.20	6.78	1.81	1.15	
After	3	4	56.7	3.34	4.76	0.63	1.43

In terms of ERA, Schilling was actually better after the injury than he was before. The worry, however, is that his strikeouts were way down in the second half, and no one, not even Schilling, can walk 0.6 batters a game. If Schilling isn’t going to strike out even five batters a game, he probably won’t be very successful.

But of course, the post-injury sample size isn’t very big. You can’t draw any conclusions from under 60 innings, at least not from basic statistics. But what if we take a look at some of the Pitch f/x data that has become available this season? We might at least find out if Schilling really did lose something off his fastball after the injury and what effect that might have had on his performance.

Now, unfortunately, the Pitch f/x system recorded just 209 of Schilling’s pre-injury pitches, so we’re going to be working off some pretty small sample sizes here, but with that caveat applied to all the analysis that follows, let’s see what we can find anyways.

Pre-injury, Schilling’s average fastball was thrown at 91.1 miles-per-hour—not the same Schilling that struck out 316 hitters in 2002, but still pretty good. After the injury, Schilling’s average fastball measured 89.9 miles per hour—definitely slower, but we can still question the significance of that difference.

Let’s try to understand the significance by looking at the outcomes of Schilling’s fastballs before and after the injury:

	1B	2B	3B	HR	Balls	Swinging    Looking	Foul
Before	5.6	0.0	0.0	0.0	28.2	8.5	    29.6	9.9
After	3.9	0.4	0.0	1.1	29.0	5.0	    27.2	16.5

Note: All numbers per 100 fastballs.

The sample sizes are tiny, but Schilling got a lot less swinging strikes and a lot more foul balls after the injury, which definitely makes it look like he didn’t really have the ability to blow his fastball by people anymore. Nonetheless, he didn’t allow more balls in-play and the average outcome of Schilling’s fastballs post-injury was actually more successful than before. That’s likely a sample size issue, but perhaps all the foul balls aren’t the end of the world.

But maybe that’s not where the worst effects of Schilling’s slower fastball lie. Even if the fastball is still good enough to get by, perhaps it makes Schilling’s secondary pitches easier to hit because of the slower speed. So let’s take a look at his changeup. Again, all the same small sample caveats apply.

Schilling’s changeup did not slow down post-injury like his fastball. Actually, it was a tick faster, 83.2 miles-per-hour versus 82.9 before the tendinitis. That difference is so small, though, we might as well ignore it. The question is whether or not the difference between an 8.2 mile-per-hour fastball/changeup differential and a 6.7 differential is significant. Can hitters sit on Schilling’s changeup now, rendering his strikeout pitch ineffective?

The results indicate that they can.

	1B	2B	3B	HR	Balls	Swinging    Looking	Foul
Before	4.5	1.1	0.0	1.1	28.4	8.0	    18.2	22.7
After	6.4	4.0	0.0	2.4	29.6	7.2	    10.4	23.2

Note: All numbers per 100 changeups.

Schilling got a lot fewer changeups past batters after coming back from his shoulder injury, and a lot more of his changeups got crushed. He went from giving up 11.1 total bases per 100 changeups before the tendinitis to 24 after. That’s huge! Whether or not all of that can be attributed to sample size, I don’t know, but it definitely isn’t a good sign.

Overall, however, I am not too worried about Schilling coming into 2008. It’s not like his fastball dropped down to the mid-80s—Schilling was still throwing about 90 miles-per-hour on average, and he could still get into the mid-90s when he needed to (actually, Schilling’s fastest pitch recorded by Pitch f/x was a 96 mile-per-hour fastball thrown to Derek Jeter in mid-September).

It is possible that Schilling won’t be as good a pitcher in 2007 as we have come to expect, but he should still be a solid number two or three starter on a team that has two or three other top-of-the rotation kind of guys. With an off-season to strengthen and rehabilitate his shoulder and $3 million in contract incentives to keep his weight down, Schilling should come back stronger last year than he was at the end of this season.

A comparative study on an unwritten rule of baseball.

I think that the negative impressions about Schilling’s ability to still pitch stem from that one playoff game against Cleveland, as he otherwise pitched very well in the postseason, and one game just doesn’t mean anything.

Schilling should still post an ERA in the lower 4s next season, and if he stays healthy, he is a prime candidate to win 15 games, further solidifying his Hall of Fame case.

References & Resources
A big thanks to Josh Kalk for helping me out with the Pitch f/x data, and to Sean Foreman for the fantastic resource that is Baseball Reference.

Print This Post

Comments are closed.