March 20, 2013

Pebble Hunting

Watching Jered Weaver

by Sam Miller

One of my plans for the 2013 season is to try to watch baseball like a scout would. Buy a thousand index cards, jot down notes on each player in each game, throw some numbers on the tools, and file them away. I figure the practice will give me more discipline, help me focus on the process in each play, and leave stronger impressions on my mind about what I watched.

Of course, I don’t have nearly the knowledge or technique to actually scout, but ideally this will be a start toward picking up things that I'd like to pick up. For example, it might have been nice to have noticed this:

After Jered Weaver pitched three innings against Italy in his second exhibition start of the spring, he said that he's finally reverting to the lower arm angle he favored when he was younger. Weaver said he'd gotten away from stretching and had changed gradually to a more over-the-top delivery in recent seasons.

"I am more of a three-quarter guy and the last couple years I went more over the top to go around the tightness," he said. "Now I can go back to my normal arm slot."

"As far as the naked eye, you aren't going to see (the change)," Butcher said, "but his arm feels a lot better to him."

I’ve been watching Jered Weaver closely since 2008, and I didn’t notice this at all. I’d have sworn he was already and always more of a three-quarter guy. I once looked at his pitching motion over the years looking for clues related to another topic, and I missed any change in his arm slot.

And yet, it’s just like he says, a small but persistent change. With the help of PITCHf/x, and looking at his vertical release point in each of the past five seasons, and adjusting (thanks, Max Marchi!) for the changing calibration of the PITCHf/x cameras, we see a clear trend upward. On all fastballs thrown at home in each of those seasons, here are the median vertical release points:

• 2008: 6.070
• 2009: 6.115
• 2010: 6.118
• 2011: 6.139
• 2012: 6.181

Possibly—presumably?—Weaver’s arm slot would be even lower if we had data for 2007, and 2006, and 2005. Is the change visible to the naked eye? Here’s Weaver in his first major-league appearance, in 2006; and in his last (to date) major-league appearance, in 2012: 

See a difference? You might. I might. He's more upright in 2012, and looks like he's got more drive—longer stride, more follow-through—but I'm not sure on the arm angle. But maybe frame-by-frame:

Not yet

Maybe

Oh wow, there it is. For exactly one frame, the difference is stark. 

Here is each of those images lined up with the same frame for one of Weaver's spring training pitches this month. 

2012 vs. 2013:

2006 vs. 2013:

So we have concluded: 

• Jered Weaver wasn't lying. His arm slot has moved higher, steadily, over the past seven or eight years. 
• Jered Weaver wasn't lying. His arm slot is back down to where it was when he was a younger, more flexible man.
• Mike Butcher wasn't lying. It's a fairly small change, hard to pick up with the naked eye. 

Weaver's arm slot is especially important because, thanks to his long frame and across-the-body motion, he has one of the more famous arm slots. Jeremy Greenhouse wrote about his extreme throwing angle in early 2011—when it was more extreme than it was in 2012—to explain why his fastball plays up. Jeff Sullivan suggested, convincingly, that Weaver's arm slot might play especially well against the hitter's eye in Anaheim. So being aware of Jered Weaver's arm slot is probably more important than being aware of, say, Mike Pelfrey's arm slot. 

Being aware of it is not the same as knowing what it means. Noticing Weaver's arm slot would have raised questions that aren't necessarily answerable, such as "is this good or bad?" Those questions are hardly ever easy. But, for instance, there is some indication that Weaver's arm slot affects his velocity. Dividing his 2012 fastballs* into five release-point tiers—the lowest fifth, the second-lowest fifth, etc. —will have you certain there's a trend here:

Release point	Two-seam velocity	Four-seam velocity
Lowest 20 percent	88.4	89.2
Second-lowest	88	88.6
Third-lowest	87.6	88.7
Fourth-lowest	87	87.8
Highest	87.3	87.6

Except that, as Max Marchi showed last year when writing about Barry Zito, arm slot and velocity might share a confounding variable: fatigue. The later into a game, Max showed, the lower the arm slot. That's not what we're talking about with Weaver—in fact, the opposite is true; the lower his arm slot, the higher his velocity. But the same principle might be true, especially if, as Weaver says, his arm slot gets higher when his shoulder is weaker. 

There is also some indication that Weaver's arm slot affects his horizontal movement. The movement he got on his two-seamer at each arm slot tier: 

Release point	Two-seam movement
Lowest 20 percent	5.41
Second-lowest	5.27
Third-lowest	4.98
Fourth-lowest	4.79
Highest	4.85

but the difference completely disappears if you limit it to two-seamers thrown in the first three innings, when Weaver is presumably not as tired. So, again, the causation may be coming from a different direction altogether.

And, logically, the arm slot could affect his ability to take advantage of the hitter's backdrop in Anaheim. I went out to the park in the middle of the day Tuesday to see what, exactly, it looks like from the batter's box. From the right-handed hitter's box, the view against Weaver is unexceptional: 

but this is the view from the left-handed batter's box, and note that the camera doesn't do justice to the white glare of the mid-day sun off the rockpile: 

If Weaver really is able to throw from an extreme-enough angle to release his pitches out of the rockpile background, then lowering his arm slot would be an advantage. I guess. What am I, Jack Maloof? 

The point isn't that any of these effects are or they aren't. The link between process and outcome is hard to prove. Weaver's arm-raising coincided with his best years, ERA-wise; it also coincided with his declining strikeout rate, strikeout-wise. It coincided with much better results against lefties, but there are plenty of explanations for that—some discussed here last summer—that might be more compelling than this one. Come to think of it, the subtle shift could be used to narrativize (waiting ... waiting... and there's the squiggly red line under narrativize) almost any outcome: 

• If Jered Weaver gets better now it's ... because a lower arm slot gives him more movement and a more disorienting release point. Thanks a lot, arm slot!
• If Jered Weaver gets worse now it's ...  because lefties pick up the lower arm slot better, as lefties tend to do. Thanks a loooooooot (sarcastic tone), arm slot.
• If Jered Weaver stays basically the same, it's ... because his arm slot has changed by only a little bit. 

So I suppose we should all just keep watching. Closely. 

*In all cases, fastballs were limited to home ballpark within a single year to avoid substantial calibration issues. Thanks to Ben Lindbergh and Max Marchi for research assistance.

Sam Miller is an author of Baseball Prospectus. Follow @SamMillerBB
Click here to see Sam's other articles. You can contact Sam by clicking here