Changes in Archer's pitch shapes

Non-believers have often labeled Chris Archer as a "two-pitch pitcher". They aren't completely wrong, as Archer features a decent fastball and an excellent slider but has struggled to find a consistent third option. He has thrown a changeup occasionally, but has been throwing it less often each year since 2012 and it hasn't proven to be a reliable option. While Archer has been able to use his slider effectively against hitters from both sides of the plate -- holding hitters to a .188 BAA in 2014, it would helpful if he could develop his changeup to help neutralize lefties.

In his opening day start, Archer performed better than his box score stats suggest. Most of the damage the Orioles generated came from a few mistake pitches, and other than those, Archer looked decent.

Looking at his PITCHf/x data from opening day, some of his pitches had different movement, or shapes, than they had during the 2014 season. While it may seem foolish to look at one start and evaluate pitches, shapes and velocities stabilize very quickly -- in as little as three pitches. Below is a chart comparing the differences in movement during Archer's opening day start.

Key

• Green: four-seam fastball
• Black: two-seam fastball/sinker
• Blue: changeup
• Red: slider
• Squares: Opening Day
• Diamonds: 2014 season

There are a few things that immediately stand out. First, Archer gained some serious rise on his fastballs, going from 10.21 inches to 11.95 inches on his four-seam, and 7.8 to 10.87 on his sinker/two-seam. While more rise could lead to more whiffs, running a simple regression of sinker components against groundball rate results in a negative correlation between rise and ground balls. Logically, this makes sense. More rise on a pitch would cause hitters to have a higher probability of making contact on the bottom half of the ball and hitting the ball in the air. Less rise would increase the chance of hitting down on the ball and putting the ball on the ground.

Both of his fastballs generated groundballs at a rate close to league average in 2014, but in 2013, both were significantly above that rate -- his sinker went for a groundball on 62.5% of the balls in play. The decrease in groundballs led to an increase in fly balls, which isn't always a bad thing, but can lead to home run troubles. This is certainly something to pay attention to -- the change in movement could get him more whiffs, but could be detrimental as well.

In addition to his fastballs, Archer's changeup moved differently than it had in 2014. Generally, more movement on a changeup is better, with a particular emphasis on vertical movement, according a study conducted by Harry Pavlidis at Baseball Prospectus. While Archer's vertical movement decreased, it was by only half of an inch, and a change this small is negligible. However, the increase of horizontal movement by close to two inches is encouraging, and could lead to better results with the pitch.

From Archer's first start, the overall changes in movement for his arsenal were a net positive. But, some of the changes didn't hold up in his second start of the season on April 11.

Here is the same scatterplot from before, with the addition of the movement from Saturday:

Circles represent pitches from his second start.

Some thoughts on his second start:

• The increase in vertical movement on his fastball and sinker disappeared in his second start. However, he still had more rise than he did in 2014, and it'll be interesting to see if he regains the rise or continues to trend toward his last start.

• Archer's slider gained almost three inches of drop, moving back toward its 2014 shape. It was much more effective on Saturday, as it drew 38.8% whiffs, compared to only 18.1% on Monday. This is most likely due to the vertical movement, as more drop causes hitters to chase more often.

• Perhaps most importantly, Archer had significant differences in his changeup movement. The shape reverted back toward 2014, featuring close to 6.5 inches of rise and run. While the decrease in rise is beneficial in generating groundballs, the drop in overall movement isn't encouraging. The changeup shape in his first start gave hope that he may have found the elusive third pitch, but a shape like that of 2014 is a little uninspiring.

• Furthermore, Archer had more velocity on his changeup. While more velocity on a fastball is beneficial, a harder changeup isn't always good. Pavlidis' work suggests that the velocity difference between a pitcher's fastball is positively correlated with groundballs, but negatively correlated with whiffs. Because Archer's fastball didn't speed up like his changeup did, the gap between these two pitches decreased. So far, Archer's changeup has generated groundballs and whiffs at a below average rate. If this new velocity on his changeup sticks and he gets more groundballs, then it may help him be more effective against lefties. However, it does "cap" some of his growth on the pitch, because a changeup that fast with Archer's shape is unlikely to become elite.

Archer is slated to start on Thursday against the Blue Jays. During that start, it'll be important to watch his changeup velocity and see if the faster changeup is here to stay. Furthermore, it'll be important to see if the rise on his fastball stays close to 2014, or jumps back up to the 11 inches of rise he had on opening day.

I think that the differences in his changeup shape have the potential to be most impactful. If he can hold onto the shape changes he had on Opening Day, this could be a major step forward for Archer's development. If not, he still has the stuff to be a solid starter at the top of the rotation.

PITCHf/x data is from Brooksbaseball.net.