Projected ERAs of D-back Starters in 2021

Next season, how will the Diamondback starting pitchers perform? To answer that question, let’s create an equation (you could call it a model).

First, I give credit to a previous work, especially for BABIP and LOB weights, that is shown in a model created in 2015 by Matt Jackson.

Let’s start with assumptions.

• A pitcher’s results are a combination of his skill (which is repeatable) and factors he cannot control.
• FIP is a measure of a pitcher’s skill. Although far from perfect, in general it can predict how well a pitcher will pitch in the future.
• Factors that a pitcher cannot control include BABIP, runners left on base (LOB%), and homers per fly ball (HR/FB). For each pitcher, in general in the next season those factors will move toward league average.
• Pitchers with experience in the Majors are valued. The Diamondbacks will choose their opening day starters from that group.
• Discussion of runners left on base (LOB%) and FIP can be found in Sean Testerman’s article. Discussion of HR/FB from a pitching perspective can be found in this FanGraph article. For more background, you might want to look at Insights on BABIP.

With the heavy lifting done, let’s build our equation.

Instead of using FIP from a single season, the predictor of a pitcher’s next season performance is a weighted average of FIP in the last 2 or 3 seasons. The latest season will be weighted 3, the previous season will be weighted 2, and if the pitcher was in the Majors, two previous seasons ago will be weighted 1.

Three statistics (BABIP, LOB%, and HR/FB%) describe factors that are largely out of a pitcher’s control can nevertheless impact his ERA. How they impacted the ERA can be determined by how much they differed from league average. The ERA predicted by wFIP will be adjusted to subtract out the estimated impacts. Matt Jackson determined the constants for BABIP and LOB in 2015. The added effectiveness of the HR/FB adjustment (including it’s constant) has not yet been determined.

The equation follows, where ERA is predicted earned run average (ERA), wFIP is weighted average of Fielding Independent Pitching (FIP), BABIP is batting average on balls in play, LOB is baserunners left on base, and HR/FB is home runs per fly ball.

Weights for BABIP and LOB - Matt Jackson’s model. League averages - BABIP from Team Rankings, LOB from FanGraphs, HR/FB from Baseball Reference. Overall formula - Makakilo.

Now, let’s predict next season’s ERAs.

The 2020 ERA’s excluded innings pitched from the bullpen. With the caveat that actual ERAs will likely differ from the projected ERAs, perhaps the projected ERA’s will at least indicate whether a pitcher will improve. That being said, let’s look at what the projections say about the players.

Luke Weaver is projected to be an ace. As a starter, his 2.71 projected ERA would be slightly better than his 2019 ERA of 2.94. In the last two years, experts wrote comments that support this projection:

• “I loved Luke Weaver. Excuse me, I love Luke Weaver….Every single one of Weaver’s pitches grades above average by Command+, for one. He still has the plus fastball and changeup combo. And the curve is improving. As he adds velocity to the pitch, it’s getting more whiffs. What’s he missing? Velocity, three pitches with a chance at four, and command of his entire arsenal? Could it be just a few too many pitches down the pipe? The stuff’s in there for him to succeed.” — Eno Sarris, February 2019
• “I thought Weaver’s fastball looked very good in spring. 93-94 much of the time, with a lot of movement, and most importantly he was commanding it. The Changeup is great. Gets a lot of swing and miss with that. Curveball much improved. He was landing it for strikes and inducing weak contact with it. His Slider/Cutter combo, (Slutter ?) is still a work in progress. But if he has the confidence to use the curve at least 10-15% of the time, his FB/Changeup will be that much more effective.” — Jack Sommers, March 2019
• “By the end of the season, Luke Weaver will be the team ace. He has all the tools, and at last, the opportunity. (Plus, Zack Greinke might be traded by then.)” — Will Leich, March 2019
• “By being more efficient with his spin — something the Diamondbacks are elite at teaching — he finally has above-average ride, and that plays up because of the low release point on his drop-and-drive delivery. Now his fastball is getting more whiffs than it has since his rookie year, and he’s pairing it with the most right-on-right changeups of his life.” — Eno Sarris, August 2020
• “This is a man, who when speaking, projects the image of wearing flip-flops and sunglasses with sunscreen on his nose and a rum drink in his hand. Laid back, and relaxed oozes from his voice while he allows his answers to simply form themselves while he is in the process of speaking. It is obvious, even just over the phone that this is a man who truly lives by his comment, ‘...bringing a good fun personality and atmosphere to the people around me and just being laid back, that’s just a huge part of who I am and I think it helps me through the season and really just the keyword - fun.’ ” — James Attwood, December 2018

Alex Young’s 4.08 projected ERA would be about the same as his 2019 ERA of 3.86. Like Luke Weaver, his high ERA in 2020 was an anomaly.

Although his ERA and quality starts showed he was an ace in 2019 and 2020, Zac Gallen is projected to stumble in 2021. I’d love to be shown wrong when his ERA continues to stay under 3, as it was in 2019 and 2020. Perhaps the model does not apply to Zac Gallen.

Taylor Clarke is projected to improve on 2020, which was an improvement on 2019. His 4.99 projected ERA would be slightly better than 5.09 in 2020 and 5.88 in 2019. His ERA could improve more than projected if he could pitch better in the first inning. FanGraph split showed his first inning ERA was 9.00, and his first inning FIP of 12.19 was highest of D-back starters.

Madison Bumgarner’s ERA is projected to improve on 2020, but fall short of the level that he pitched for the Giants. His 5.39 projected ERA is better than his 6.48 ERA in 2020.

Merrill Kelly will return to the rotation after Thoracic outlet surgery. There is some uncertainty whether he will be ready by opening day. His 5.77 projected ERA is worse than 2.59 in 2020 and 4.42 in 2019. Again, I’d love to be shown wrong when he pitches like 2020, when 4 of his 5 starts were quality starts.

Caleb Smith’s projection was based on 9 innings pitched. Because of the small sample size, there is little to no confidence in his projection. In 2020, he had a career best ERA of 3.00. In his last three seasons, his highest ERA was 4.52.