Why I Targeted Randy Dobnak Back In October

On last October 3rd, I examined how the effects of the Happy Fun Ball could mess with ERA estimator assumptions. I was self-serving in that I wanted to see how the variables in my own ERA estimator (pERA) changed*. Once I had the new constants, I created the valuations, and Randy Dobnak came in with an estimated sub-3.00 ERA ahead of starters such as Carlos Carrasco, Blake Snell, and Shane Bieber. The rankings were there for the public to admire and they were completely ignored throughout draft season.

I probably would have ignored them also if it weren’t for Spencer Turnbull. At the end of the 2018 season, Turnbull had a 6.06 ERA and was on no one’s radar for 2019. But I had his pERA at 2.31 better than both Justin Verlander and Chris Sale. I completely blew off the rankings and paid for it. From the beginning of the season until a shoulder injury in late June, Turnbull had a 2.97 ERA, 9.2 K/9, and 1.29 WHIP. And I had him rostered on no teams.

When it came to this draft season, especially in draft-and-holds, I wasn’t going to make the same mistake. I grabbed Dobnak everywhere. While Dobnak was my deep sleeper, I was also targeting Matt Shoemaker (3.41 pERA), Alex Cobb (3.45 ERA), and Alec Mills (3.51 ERA) based on just a few 2019 starts. The decisions have paid off.

Each week and I run the pERA rankings and Brett Anderson’s pERA was at 3.30 this last week putting him in the company of Lucas Giolito (3.21), Patrick Corbin (3.26), and Luis Castillo (3.35). It took about two seconds on our player pages to find out why: a 2.0 BB/9 and 63% GB%. His strikeout rate doesn’t really matter when a team has to hit three singles in an inning to score.

To dive in a little deeper, here are his pitch results:

Without knowing the numbers are associated with Anderson, they are outstanding. The two pitches he throws nearly 70% of the time generate groundballs at a 70% clip. Any swings-and-misses he generates with them are gravy. The cutter has a respectable 57% GB%. Finally, the slider lowers his overall groundball rate but it’s creating a ton of popups to go along with the 18% SwStr%. pERA acknowledges the advantage of both the high and low groundball rates.

I’m sure most readers are at the “Nice story dude” point and just want this year’s rankings. OK. Here is the full list and the top-30 starters (minimum two starts) so far this season.

The list passes the idiot check with the results so far this season with just a few names sticking out.

Adrian Morejon: It’s been a small sample (4.2 IP) and no one is going to care about him with a 5.79 ERA but he should be added to watch lists. He’s getting a ton of swings-and-misses (15% SwStr%) while keeping the ball on the ground (60% GB%). The only qualified starters posting similar numbers are Luis Castillo (16% SwStr%, 58% GB%) and Pablo Lopez (14% Swstr%, 61% GB%). Morejon has a ways to go to get to be compared to the pair, but it’s a start.

Tommy Milone: Oh the crap I’m going to take from Grey Albright for this ranking. The deal is that Milone is maximizing his value buy simplifying his approach. He’s not walking batters (1.2 BB/9). He’s throwing his elite changeup (21% SwStr%) 39% of the time to go with his  86-mph extreme flyball-inducing four-seamer (14% GB%) that has seen it’s swinging-strike rate nearly double from 3.7% to 6.8%.

He’s likely getting the extra misses with his NAIA fastball by throwing it up in the zone. Previously, I found that a pitcher can double or even triple a fastball’s swings-and-misses by throwing it higher. Here is a comparison of his 2019 and 2020 fastball locations.

I’m not sure Milone could do anything more to remain relevant but he should be a testament to other struggling pitchers. If Milone and his “fastball” can be useful, what the hell is their excuse?

Nick Magevicius: Again, he’s following a simple plan for improvement. First, he’s walking almost no one (1.4 BB/9). He’s gained 2 mph on his fastball (88.3 mph to 90.3 mph) and is posting a reasonable 8% SwStr%. And his overall 39% GB% is hiding a dual-nature batted ball profile. His fastball has a 33% GB% while his slider (80% GB%) and curve (71% GB%) are worm killers.

I’m sure some analysts will mark his 1.02 WHIP and 4.12 ERA for regression, but if he can keep the results going he could be a nice sleeper over the rest of the season.

 

 

* pERA simply looks at each pitch’s swinging strike and groundball rate. Additionally, their walk rate is added at the end. I created the metric because I was wondering why certain pitchers were limiting hard contact with a middling 45% GB%. II found pitchers who had a 20% GB on their fastball and 70% GB% on the non-fastballs. All the batted balls were either easy to catch popups or groundballs.