Starting pitching was thought to be the Astros weakest link in 2017. But that narrative may change if the rotation continues to pitch well.
Dallas Keuchel, for all intents and purposes, had a rough 2016 season. He is also at the center of the Astros rotation debate. And it was because of his 2016 season leads some to doubt that he could return to his best form.
There is merit behind that thought as Keuchel relies on pitching low in the zone while not possessing overwhelming velocity. Last season, we witnessed his velocity drop and the strike zone not as generous to Keuchel as in years past. Combine all these factors to a bum shoulder, and it easy to see why he struggled.
Fast forward to 2017, and we are starting to see the Keuchel of old return. Sure, it has been only just 14 innings of proof. The results through two starts have been solid thus far.
Courtesy of Baseball Savant, you can see the progression of last night’s game pitch-by-pitch. If you scroll through it all, you will notice that Keuchel’s most thrown pitches have a velocity clocked anywhere from 88 MPH to just a hair over 90 MPH. This appears to be closer to normal. However, upon further inspection, there is a bit more to think about.
For example, Keuchel’s average four-seam fastball velocity in 2015 ranged anywhere from 89.90 MPH to 90.85 MPH per Brooks Baseball. His average sinker velocity had a range of 89.65 MPH to 90.95 MPH. In 2017, his average velocity for both pitches thus far is 89.47 MPH and 89.48 MPH. So a little down, but not by much.
On the other hand, Keuchel’s average velocity across the board was lower in April of last year compared to this season. I will concede that there is only two starts to his credit in 2017. This has to be considered a positive step forward, though, for the bearded left-hander if the numbers continue to hold up in each start. But there is a caveat to the increased velocity.
In terms of velocity measurements, I caution that plenty of readings this season are coming off a bit higher than previous years.
This is due to the fact that Major League Baseball has begun using Statcast data instead of PITCHf/x data. Per USA Today via the Associated Press, there is a reason behind this change in velocity readings.
"The key difference is that PITCHf/x calculates velocity at a set point — usually 50 or 55 feet from the back of home plate — while Statcast measures velocity directly out of the pitcher’s hand."
So if you happen to see a trend of a general increase of velocity across bored in baseball, you now know why. Here is another interesting tidbit from the USA Today article in regards to the average increase of velocity.
"Because of that difference, Statcast readings are faster than PITCHf/x by about 0.6 mph on average, according to MLBAM senior data architect Tom Tango."
Per Brooks Baseball’s website, they have used PITCHf/x data from MLBAM and Sportvision. Since the transition from PITCHf/x to Statcast data has taken place for Major League Baseball, I assume that Brooks Baseball is doing the same. Dave Cameron of Fangraphs has an article detailing this change from PITCHf/x to Statcast that is worth your time.
If that is the case, then we will need more starts and data before we can draw a fair comparison on Keuchel’s velocity. Like, how do we do compare previous years data to this year’s accurately? In terms of recording a lower velocity, perhaps Keuchel’s arm is still recovering some of that lost velocity slowly but surely. Regardless, the first two starts have generated positive results.
Only time will tell. And like last season, Keuchel is the focal point of the starting rotation. This detail will not change anytime soon.
**Statistics courtesy of MLB.com, Brooks Baseball, and Baseball Savant**