Pittsburgh Pirates: Will Regression Stop Trevor Williams?

(Photo by Scott Taetsch/Getty Images)
(Photo by Scott Taetsch/Getty Images) /
facebooktwitterreddit

Last season, the Pittsburgh Pirates had two starting pitchers inside the top 20 in ERA. One of them was a bit of a surprise.

Pittsburgh Pirates starter Trevor Williams had a breakout performance last season, posting a 3.11 ERA through 31 starts. It was his second full season at the major league level. This also included a near-historic second half in which he posted a 1.38 ERA.

Trevor William‘s success in 2019 had a great deal to do with his 76.6 LOB%, which was well above the league average of 72.8%. William’s ability to limit runners from scoring reached its peak during his second-half run when he managed to strand 90.8% of runners.

Playing around with the data a bit, if we recalculate William’s high LOB% using the league average of 73%, his run total rises to 70.68, which we will round to 71 runs. Accounting for 6 unearned runs (he had 5 on the season), Williams would have produced 65 Earned Runs for a 3.45 ERA.

Trevor also benefited from a low HR/FB rate of 8%, which was also well below the league average of 12.7%. If we further calculate by moving his HR/FB rate back to that 12.7%, his run total jumps up to 80 runs, again holding for 6 unearned runs, we now have an ERA of 3.91.

While pitchers certainly have more control over the number of home runs they give up, the runners they left on base are for the most part out of their control. Great pitchers tend to strand runners, but that’s mostly because they are good. Most evidence shows that pitchers don’t typically have the ability to play the scoreboard.

Furthermore, Trevor Williams also had a high xFIP of 4.54, which weights for the long ball. That was 7th highest among qualified pitchers in 2018. Jeff Sullivan of Fangraphs focused on xFIP in an article this past month:

"I wanted to focus on the guys with the biggest differences between ERA- and xFIP-, like what Miley just put up. In year one, the 50 guys with the biggest differences averaged an ERA- lower than their xFIP- by 41 points. In year two, the difference regressed to two points. In year one, the 25 guys with the biggest differences averaged an ERA- lower than their xFIP- by 48 points. In year two, the difference regressed to…zero points. Just nothing to be found. Historically, xFIP has won out. Historically, regression has happened."

Williams’ ERA- was 79 and his xFIP- was 112, which is a 39 point difference. While that difference isn’t as large of a margin as the average of Sullivan’s study, there is plenty of reason to believe that his ERA should regress. After all, Williams posted a 4.07 ERA in 2017 with a lower xFIP- of 104.

Diving into his pitch type splits, Williams saw significant improvements in two key pitchers: his Slider and Change-up. For his change-up, it was a drop from a 142 wRC+ to 102. For his slider, he saw an improvement from a 103 wRC+ to an outstanding 79 wRC+.

What also stands out above all else for Williams was his performance in the month of September, where he posted a season-high 24.4 K%, 18.3 K-BB%, and 2.04 FIP. He also posted a 3.31 xFIP, so while his season peripherals point to a decline, Williams may have found something during that final stretch that he will carry into 2019.

If Trevor Williams wants to repeat the breakout season he had last year in 2019, he will need to continue his success of leaving runners on base and keeping balls in the ballpark. For a pitcher with his statistical profile, it is hard to believe that he will continue having such success in these areas.

Next. Examining The Pittsburgh Pirates Bullpen Options For 2019. dark

However, if Williams is able to continue having success with his fastball and slider and can continue improving his change-up, 2019 could very well be better than last season. Regardless, if Williams is able to post an ERA in the 3.90s as I had calculated earlier, that could be all the Pirates need out of a 4 starter in what is a very good starting rotation.