Wednesday, August 12, 2009

Contrasting Beckett's two Stadium starts

(This is all part of an ongoing effort to understand and explain some of the data included in PitchFX charts, an incredible reservoir of information about pitchers and pitching and everything that goes into that part of the game. You can find those charts for every major-league pitcher in every game at

Josh Beckett, like fellow ace Jon Lester, found himself knocked around a little bit to start the season. He had a 7.22 ERA at the end of April and allowed 10 hits in three straight starts, a stretch that included a May 5 start at Yankee Stadium in which he escaped having allowed just three earned runs in six innings.

He then went back to Yankee Stadium on Friday night and pitched seven spectacular innings, holding the Yankees without a run on four hits in a game that eventually went deep into the night before Alex Rodriguez went deep against relief Junichi Tazawa.

When you look at the PitchFX charts, you can see some significant differences between the way Beckett attacked the Yankees on May 5 and the way he attacked the Yankees on Aug. 7:

1. Using the whole plate against lefties
Part of the issue with the new Yankee Stadium is the short porch in right field, the only place in the world where Dustin Pedroia can hit an opposite-field home run. A strong lefty can get a ball in on the fists and, as long as he gets his body turned just a little bit, muscle a ball over that fence.

Check out the way Beckett pitched lefties on May 5 -- keeping in mind that the chart is from the perspective of the catcher...

... and the way Beckett pitched lefties on Aug. 7:

He consciously pitched away from the inside half of the plate against lefties -- and switch-hitters like Mark Teixeira and Nick Swisher, naturally hitting lefthanded against him -- and instead worked the outer half. He didn't give lefties a chance to take the ball over the fence in right field, but he did give them a chance to extend their arms and a chance to dive out over the plate a little bit. If Beckett gives up the inside half of the plate, he gives up quite a bit of his effectiveness.

The next time he pitched at Yankee Stadium, he didn't pitch to the ballpark. He pitched his game. He attacked the inside half of the plate. In doing so, he regained control of every at-bat.

(One thing he also did: He buried his curveball. Rather than throwing his curveball in a spot where it could be hit, he threw it down and in and even in the dirt occasionally. If it's going to be a strikeout pitch, that's what he has to do with it.)

2. Reintroducing the two-seam fastball and changeup
Beckett is dangerous enough with two pitches. When he can throw a third pitch with effectiveness, it makes him all the more lethal.

Check out two more charts, first from May 5...

and, second, from Aug. 7:

It helped that he threw his fastball with a little more velocity, of course. But the key to pitching isn't speed as much as it is differential and timing: The more a pitcher can keep hitters off-balance, the better. Throwing a 95-mile-an-hour four-seam fastball isn't going to help much if your two-seam fastball is 92 or 93 with similar movement. You might as well ditch the two-seamer and just throw four-seamers all night long.

In Beckett's Aug. 7 start at Yankee Stadium, though, he found a way to differentiate his two-seam fastball from his four-seam fastball. Both pitches had the same type of movement but were separated by six or seven miles per hour rather than three or four miles per hour.

On top of that, he threw his two-seamer with consistency. Graphs like the above graphs can be overwhelming, but what's often most telling is how closely grouped the points of data are. A pitcher always strives to throw his pitches with consistency, to have the same velocity and the same movement on every fastball and simply to vary the location and pitch selection to keep the hitter off-balance. If fastballs show different amounts of movement, they're going to be difficult to control and thus throw with pinpoint control.

The chart from May 5 shots Beckett throwing two-seamers and changeups with varying wildly varying amounts of movement. You couldn't draw a circle around that group of data points; you'd have to draw an awkward-looking oval. His changeups and two-seamers from Aug. 7, though, fit neatly into a confined area and indeed could be contained within a pretty small circle.

That's what consistency and pinpoint control looks like.

No comments: