After two rough years in which he was used as a starting pitcher, Archie Bradley was moved to the bullpen in 2017.

It was the best possible decision at the time.

After struggling mightily with his control in 2015 and 2016, Bradley calmed himself and turned into a star reliever. The 2017 campaign saw the pitcher put up a 1.73 ERA with 79 strikeouts in 73 innings, good enough for a 273 ERA+ and 20th place finish in NL MVP voting, tied with teammate Zack Greinke.

But it’s only been downhill from there.

Bradley was still good in 2018 but less so, a performance that could have been expected given the sudden improvement the year before. While strikeout numbers remained close to the same, Bradley got hit a lot harder in 2018. His home runs per nine innings rose to 1.1 from 0.5, and his ERA ballooned to 3.64.

This year, things have gotten even worse. Bradley’s role has been tinkered with as he’s posted a 4.71 ERA in 27 games. The control issues have returned as well, as Bradley is walking 5.3 batters per nine innings compared to 2.5 in 2018. 5.3 sits closer to his numbers in the 2015-16 seasons, as those were 5.6 and 4.3, respectively.

As Kellan Olson detailed here, and as manager Torey Lovullo pointed out, Bradley’s fastball command has gotten away from him. The four-seamer, which Bradley has thrown 64.94% of the time, isn’t nearly as effective in 2019.

Data from Brooks Baseball suggests a couple points.

First, Bradley has changed the vertical release point of the pitch.

This chart shows where Bradley has released the four-seamer from his hand since the start of his very successful 2017 season. Then, Bradley was typically throwing the pitch between 5.9 and 6.05 feet above the ground. This season, it’s been elevated.

Bradley elevated it last year and didn’t suffer, but the elevation wasn’t nearly as significant. The vertical release point of the pitch sat just below the 6.1 foot mark last season.

Besides just general batting statistics, Brooks Baseball also supplies us with “zone profiles” that show batting statistics within areas of the strike zone.

With Bradley being a right-hander releasing his pitches from approximately six feet above the dirt, consider where a four-seam fastball, a pitch usually thrown with little horizontal or vertical movement, should land within or around the strike zone. Those mini-zones typically include the middle-left area of the plate.

According to Brooks Baseball’s data, Bradley’s seen batters sky-rocket their averages in those mini-zones in 2019 compared to 2017, which includes increases from .345 to .417, .308 to .546 and .059 to .600.

While a low 2019 sample size could be to blame for the differences, the season is approaching its halfway point, and Bradley has not improved substantially as months have gone on.

Another possible explanation for Bradley’s decline could simply be the reliance on the four-seamer, which has averaged 96.06 MPH this season. Though fast, a lack of movement and the constant throwing of the pitch could lead batters to catch up with it, similar to what has had happened to other fastball-reliant pitchers like the Yankees’ Chad Green or the Brewers’ Freddy Peralta.

Bradley’s decline has been a stunning development for the D-backs. Perhaps a look at his mechanics and the numbers could change that.