Thanks to Sam Miller, who is as much a part of my daily routine as coffee and jogging and my job*, I wound up reading this article about Brad Lidge, and about how Brad Lidge is still a free agent. There's a lot of somewhat interesting stuff in there, and that's somewhat interesting by my standards. I have high standards. If you're someone who finds a lot more things interesting than I do, then that article could conceivably blow your mind.
* this is all part of my daily routine
One thing that caught my eye was this:
"You'd think it would just be production. But it's not. It's bizarre to me because I still have a very high swing-and-miss percentage."
I touched on that briefly over here. Another thing that caught my eye was this:
"I think velocity is so important for some teams, and the prototype closer throws hard. So if you're not throwing as hard, suddenly you're not a closer. And I don't understand that totally. But it is what it is."
The content of that blockquote immediately followed the content of the first blockquote, so I guess they caught my eye at approximately the same time. Anyway, Lidge is right. The prototypical closer does throw hard. The prototypical closer has a rock music entrance ballad and a fastball in the high-90s he leans on to blow everyone away. The prototypical closer is dominant, in the classic sense of being dominant.
But that made me wonder: how much harder does the average closer throw, really? How much harder did the average closer used to throw, ten years ago?
All of the information we need to investigate comes courtesy of FanGraphs. FanGraphs tracks saves, and FanGraphs tracks fastball velocity. FanGraphs also tracks a million other things that we don't care about (for this). Let's begin with this past season, shall we?
Relievers with 10+ saves: 93.6 mile-per-hour average fastball
All other relievers: 91.8
Relievers with 5+ saves: 93.6
All other relievers: 91.7
I didn't know which cutoff to use. It doesn't really matter. Now let's go back a decade, to the first year for which we have data. What was the situation then? Different? Similar? The exact same?
Relievers with 10+ saves: 92.6 mile-per-hour average fastball
All other relievers: 89.8
Relievers with 5+ saves: 92.2
All other relievers: 89.8
Again, the cutoff doesn't really matter. Let's put it at 10+ saves. In 2002, relievers with at least ten saves had average fastballs 3 percent harder than all other relievers. In 2011, relievers with at least ten saves had average fastballs 2 percent harder than all other relievers. One way of looking at that is that, over the past decade, closers have lost about a third of their fastball advantage. Another way of looking at that is that there's basically no meaningful difference.
Maybe more interesting is that all relievers as a group are throwing harder. In 2002, the average fastball by a reliever was about 90.2 miles per hour. In 2011, it was about 92.0 miles per hour. People have observed this before, with relievers and starters, and it's interesting, if it's accurate. We can't swear by the accuracy of our 2002 data, but it's probably all right.
For the curious, here are pitchers from 2002 who had at least ten saves, and who threw their average fastballs below 90 miles per hour:
The same list, for 2011:
Closers do throw harder, on average. Which makes sense, since pitchers are better when they throw harder, on average. Closers do not throw way harder, on average, and you don't necessarily have to be a flamethrower to be a closer. The image of the prototypical closer is the same that it's been for a long time, but not every closer is the prototypical closer.