Bottom 20 Starting Pitcher SIERA Laggards

Yesterday, I listed and discussed the top 20 starting pitcher SIERA leaders. Since the entire season represents a small sample size, you should be far more inclined to focus on a pitcher’s SIERA, driven by his underlying skills, than the ratios that actually count in your fantasy leagues, ERA and WHIP. The underlying metrics fueling SIERA stabilize far quicker and account for skills pitchers have more control over. It therefore makes for a significantly better rest of season projection, even though SIERA is meant to be backwards looking, not forward looking. With that said, let’s check in on the bottom 20 pitchers in SIERA and discuss the interesting names. As a reminder, just because a pitcher finds his name here doesn’t mean you should drop him or trade him away, as the pitcher could improve his skills, pushing his SIERA down into more attractive territory.

SIERA Laggards
Name SIERA
Mike Fiers 6.05
Tyler Anderson 5.55
Jon Gray 5.55
Martin Perez 5.53
Dylan Cease 5.40
Ryan Yarbrough 4.91
Garrett Richards 4.89
Johnny Cueto 4.80
Alec Mills 4.62
Brady Singer 4.62
Adam Wainwright 4.55
Chris Bassitt 4.55
Trevor Williams 4.48
Antonio Senzatela 4.48
Mike Minor 4.48
Jose Berrios 4.44
Randy Dobnak 4.43
Dallas Keuchel 4.41
Alex Cobb 4.41
Lance McCullers Jr. 4.40

As a starting pitcher on the Rockies who has had to deal with one of the league’s most hitter friendly environments, it’s no surprise that Jon Gray’s ERAs have represented a roller coaster ride since his first full season in 2016. But his SIERA marks had typically been much more consistent, aside from last year’s jump. This year, his skills have completely collapsed, namely his strikeout rate has shockingly been cut in half. Although his SwStk% remains just barely in double digits, his strikeout rate sits at a lowly 12%. That ranks second worst among qualified pitchers. Who would have thunk that after 30 innings, Gray would be the guy who ranks that low?! One of the obvious possible reasons is a significant decline in fastball velocity, which has dipped two miles per hour from last year. He was always a risky asset as a Rockies starter, but with reduced velocity, I want no part of banking on a rebound.

I thought Dylan Cease was a reasonable buy in a shortened season as owners should be focusing on strikeouts, while the ratios bounce all over the place given small sample randomness. Cease has proven to be a partly correctly call as a target, as his ERA sits more than two runs below his SIERA, but the strikeouts have been weak. Unlike Gray, his fastball velocity is actually up by more than a mile per hour. He has also thrown his better whiff-inducing slider instead of his curveball, but neither of these factors have helped his strikeout rate. I’d bet he improves his skills, bringing down his SIERA, but it won’t be enough to fend off some major ERA regression.

Garrett Richards returned from TJ surgery last season, and he recorded just 8.2 innings. So the short season seemed the perfect chance to roster Richards, who wouldn’t have to worry about an innings limit. Unfortunately, his strikeout rate has collapsed and overall skills poor. He’s no longer a ground ball pitcher and still possesses below average control, so without the strikeouts, he’s a below average pitcher. His fastball velocity remains well below its peak, but he’s throwing his slider at nearly the highest rate of his career, which should be a positive for his strikeout rate, but it hasn’t been enough to offset the weaker fastball.

Chris Bassitt has enjoyed an outwardly strong start to the season, but a deeper look suggests trouble ahead. His skills are mediocre at best and comes with a below average strikeout rate. It’s worth noting that over his career, he has dramatically outperformed his SIERA, but that’s over just 368.2 innings, which is too small a sample to determine whether he’s inherently a SIERA outperformer or has just benefited from a string of good fortune over about two full seasons worth of innings.

After a slow start from a skills perspective, Mike Minor’s SIERA now sits almost exactly the same as last year, but his luck has turned. The entire difference is his HR/FB rate has skyrocketed, while his LOB% has plummeted. Both should move closer to his career norms, resulting in an improved ERA, but even a neutral luck Minor is pretty meh.

Jose Berrios has earned the perception of one of the better young pitchers in baseball and a targeted fantasy asset, but his underlying skills have never been very impressive. With a SIERA well above 4.00 once again, he’ll need to rely on some good fortune like last year to get that ERA below 4.00. I’d rather just own a guy whose skills actually support a sub-4.00 ERA than hope Berrios posts another better than league average HR/FB rate.

Sheesh, is Randy Dobnak playing with fire or what?! With a 1.78 ERA, he’s been a boon to fantasy owners, but a 13.5% strikeout rate means success is unlikely to last much longer. He does offset those lack of whiffs with an extreme grounder rate, so he could rely on double plays to erase all the baserunners he should be allowing. Those baserunners aren’t making it onto the basepaths though thanks to a crazy .226 BABIP, which makes little sense from an extreme ground ball pitcher, since grounders go for hits far more frequently than fly balls.

Lance McCullers Jr. was seemingly another short season beneficiary, as he missed all of 2019 recovering from TJ surgery and the shortened season means there wouldn’t be as large a difference between his innings and the top tier pitchers in innings. Unfortunately, while he has remained healthy, his performance has been quite disappointing, and it’s backed by weak skills. His strikeout rate has fallen below 20% for the first time, but while his fastball velocity is down, it’s only down marginally, and not enough to question how well recovered he actually is. I’ve been a big fan in the past, but it’s anyone’s guess whether he improves his skills the rest of the way given the uncertainty of pitchers’ recoveries from such major surgery.





Mike Podhorzer is the 2015 Fantasy Sports Writers Association Baseball Writer of the Year. He produces player projections using his own forecasting system and is the author of the eBook Projecting X 2.0: How to Forecast Baseball Player Performance, which teaches you how to project players yourself. His projections helped him win the inaugural 2013 Tout Wars mixed draft league. Follow Mike on Twitter @MikePodhorzer and contact him via email.

9 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
weekendatbidens
3 years ago

SEIRA seems to discount lesser-stuff pitchers’ success. Does SIERA miss these instances because the success isn’t classified under what has been labeled as successful metrics or is it just luck in favor of these pitchers?

Anon
3 years ago

This has become one of my issues with the estimators – they can vary by a LOT. SIERA thinks Cueto and Richards should be around 4.85 or so and Mills at 4.62. But Statcast’s xERA (using Statcast data) thinks they all should be at 4.15. That’s a big difference.

And most vary by more than a little – here are some big ones:
McCullers – SIERA 4.40, Statcast 6.07
Cease 5.40 vs 6.04
Trevor Willilams is the other way – 4.48 vs 3.66

Most guys vary by at least 0.20 and if the difference for everyone was that small it wouldn’t be a big deal but when that is the minimum difference, I guess I don’t know which numbers to think are more accurate. Statcast is based on better data as far as I’m concerned, but their xwOBA needs some additional (and frankly, very obvious) tweaks to really be accurate and xERA leans on xwOBA.

DDD
3 years ago

Yeah, there is definitely an issue. On the opposite side, Matthew Boyd (with an 8.48 ERA) has a 4.29 SIERA. He sports a 1.74 WHIP, 2.51 HR/9, and a 315 BAA. I realize he has 10.36 K/9 and a 374 BABIP, but… really? C’mon Man!!! Even his FIP is at 5.84 and it has its own issues.

DDD
3 years ago
Reply to  Mike Podhorzer

Mike,

First, allow me to say that I greatly value and appreciate your work. And, while I hope this doesn’t come across negatively, I would like to defend my position while realizing that Boyd is only one individual and I’m guessing there are many examples to the contrary. You’re right, I don’t fully understand how SIERA is calculated. And, to be clear, I’m not advocating the use of ERA as a predictive stat (it’s greatly impacted by too many variables). Nonetheless, below is a breakdown by year of ERA, FIP, xFIP and SIERA. I’m not including 2015 (his rookie year – SSS) and 2020 (wacky and SSS).

2016 2017 dif 2018 dif 2019 dif

ERA 4.53 5.27 0.74 4.39 0.88 4.56 0.17
FIP 4.75 4.51 0.52 4.45 0.12 4.32 0.11
xFIP 4.74 5.01 0.53 4.72 0.62 3.88 0.16
SIERA 4.38 4.94 0.89 4.31 0.55 3.61 0.25
“dif” means the difference between the (loosely termed) projected ERA and actual ERA for that year.

Using these stats as forecasters (predictive), FIP was the most accurate of ERA every year and SIERA was the worst 2 out of 3 years (not including ERA as a forecaster). I think the primary reason FIP was more predictive is because it does use the individual pitcher’s HR/9 in it’s formula rather than league average (xFIP) or not all (SIERA) and HR/9 are relatively predictable for each individual (i.e. a flyball pitcher is more likely to give up more HR/9, especially with the 2019-2020 baseball). I also think H/9 is relatively predictable and it is not in any of these formulas (amounts to WHIP since some semblance of BB/9 is already in the formulas). Another consideration is that a good groundball pitcher with less than league average K/9 will appear worse under these stats. Obviously, if a pitcher makes a significant change in his pitching, it can have a huge impact on his success and throw all predictive measures out the window. That’s about it. Feel free to shoot holes in it and thanks for indulging me.

LenFuego
3 years ago
Reply to  Mike Podhorzer

I don’t think you understand how “skill” is defined. Strikeout rates, walk rates and batted ball distributions are not skills; they are statistical measurements that are the *result* of applying skills in game situations.

Throwing a ball at a high speed is a skill. Throwing a ball accurately is a skill. Throwing a ball with a lot of spin is a skill. Throwing a ball with absolutely no spin is a skill. Throwing a ball with a spin that makes it drop precipitously is a skill. Throwing different types of pitches (e.g., fastballs and sliders) with motions that look identical to each other from the batter’s perspective is a skill. OTOH, strikeout rates, walk rates and batted ball distributions are not skills; again, they are statistical measurements that result from applying those types of skills in game situations.

So when you talk about SIERA accounting for a pitcher’s “underlying skills” and cite those three things as those skills, unfortunately it really makes you sound as if you do not know what you are talking about.

Lanidrac
3 years ago
Reply to  Mike Podhorzer

FIP’s main problem is that it doesn’t account for batted ball distribution like SIERA and xERA do. It’s long been shown that the old idea that pitchers have little to no control over their BABIP is a myth. As such, it’s still a decent prediction metric (although still inferior to the other two) but much worse than even standard ERA in analyzing what has already occurred.