2020 Review: Starting Pitcher SIERA Underperformers

Of the three ERA estimators available on FanGraphs (SIERA, xFIP, FIP), SIERA is the best at predicting future ERA, even though it was designed as a backwards-looking metric, like the other two. If you’re still using xFIP or FIP in your pitcher analysis, then stop, and immediately switch to SIERA. In a short 60 game season, focusing on SIERA, rather than ERA, is even more important when forecasting a pitcher’s future performance.

The underlying skills that drive SIERA stabilize more quickly and the metric isn’t influenced by the gyrations of the three “luck metrics” — BABIP, LOB%, and HR/FB — which don’t have enough time to settle around the pitcher’s true talent level. ERA is heavily influenced by how a pitcher performs in those three metrics, but there’s far too much randomness involved to place significant weight on them, even over a full 162 game season. Remember though, even SIERA isn’t perfect because there are pitchers who consistently underperform or outperform due to some skill or lack thereof that has been a challenge to identify.

So let’s review the pitchers who underperformed their SIERA marks most this season (minimum 40 innings pitched). I’ll identify which of the three luck metrics fueled that underperformance and discuss whether there’s a chance the pitcher underperforms again in 2021 or reverts closer to his SIERA (I’ll only discuss the fantasy relevant names).

SIERA Underperformers
Name BABIP LOB% HR/FB ERA SIERA Diff
Derek Holland 0.288 62.5% 26.1% 6.86 4.13 2.73
Matthew Boyd 0.308 65.8% 19.7% 6.71 4.60 2.11
Luke Weaver 0.349 63.2% 12.8% 6.58 4.52 2.06
Anibal Sanchez 0.347 67.5% 15.7% 6.62 5.03 1.59
Tanner Roark 0.329 70.3% 20.3% 6.80 5.36 1.44
Jordan Lyles 0.285 57.2% 15.0% 7.02 5.62 1.40
Mike Minor 0.269 62.9% 15.7% 5.56 4.20 1.36
Trevor Williams 0.313 70.0% 24.2% 6.18 4.84 1.34
Jordan Montgomery 0.320 65.0% 16.7% 5.11 3.84 1.27
Austin Voth 0.297 72.4% 17.9% 6.34 5.09 1.25
Frankie Montas 0.329 68.7% 17.5% 5.60 4.40 1.20
Rick Porcello 0.373 59.5% 7.6% 5.64 4.45 1.19
Madison Bumgarner 0.266 73.2% 22.0% 6.48 5.31 1.17
Robbie Ray 0.323 73.0% 18.6% 6.62 5.49 1.13
Josh Lindblom 린드블럼 0.316 66.5% 12.2% 5.16 4.09 1.07
Tyler Glasnow 0.281 78.6% 23.4% 4.08 3.04 1.04
Jack Flaherty 0.281 68.8% 23.1% 4.91 3.89 1.02
League Average 0.290 72.0% 15.5% 4.46 4.44 0.02
Significant luck metric underperformance highlighted in red

Woah, that’s a lot of red! The first thing you might notice is that this group of pitchers wasn’t very good in 2020, even with neutral luck. Of the 17 pitchers, only three posted a SIERA below 4.00, which six posted SIERA marks above 5.00. It’s pretty amazing that even with SIERA marks above 5.00, they still underperformed their ERA marks by significant margins!

Matthew Boyd
Metric(s) Underperformed: BABIP + LOB% + HR/FB
2021 Rebound? Yes, to a degree. Boyd hit the trifecta in 2020, vastly underperforming the league averages in all three luck metrics, which he hadn’t done since his small sample debut in 2015. Boyd’s issues with the long ball that manifested in 2019 became even worse this year, but HR/FB rate takes a whopping 400 fly balls to stabilize, whereas Boyd only allowed 76 this season. It’s tempting to include his elevated 2019 mark and believe he’ll be an annual underperformer, but then you’re completely ignoring his 2016-2018 seasons when his HR/FB rates hovered between 10.6% and 12.9%, not far off from the league average during those years. It is certainly possibly he has lost some of his command, but that doesn’t justify any HR/FB rate. Perhaps he doesn’t get the mark down to league average, but I would be shocked if it doesn’t improve meaningfully.

Boyd’s BABIP was surprisingly high considering his high FB%, but he also failed to induce a double digit pop-up rate for the first time. His BABIP has bounced around, so it’ll be heavily dependent on the quality of his outfield defense. Finally, LOB% is mostly the product of BABIP and HR/FB rate, so if both move closer to his career average, then it will naturally move higher.

Luke Weaver
Metric(s) Underperformed: BABIP + LOB%
2021 Rebound? Wow, no Major League quality pitcher owns a true talent .349 BABIP. However, Weaver’s batted ball profile has been all over the place since his 2016 debut, so it’s difficult to get a sense of what we should expect next year. You would expect more pop-ups given all the fly balls, but you also figure considering all those flies, it’s shocking he allowed such a high BABIP. He’s a tough one to forecast, as although he owns a .328 career BABIP, it comes in just 349.1 innings, which is just two seasons. BABIP takes 2,000 balls in play to stabilize, and he has only even faced 1,524 batters. Like I said for Boyd, when Weaver’s BABIP returns to a more normal level, his LOB% will rise. I’m just not entirely sure what that level might be given the gyrations in his batted ball distribution.

Mike Minor
Metric(s) Underperformed: LOB%
2021 Rebound? Yes. Everything here was in line with his history, with the exception of that suppressed LOB%. That’s really surprising considering his low .269 BABIP, which means opposing offenses must have strung together hits very often. Interestingly, while the 15.7% HR/FB rate was right around the league average, this was by far Minor’s highest career rate. He had never even posted a mark of 13%, let alone 15%+. Minor just signed with the Royals, which is a great park for fly ball pitchers, and will greatly increase his chances of pushing that HR/FB rate closer to his career average.

Jordan Montgomery
Metric(s) Underperformed: BABIP + LOB%
2021 Rebound? Yup. While his stamina and ability to even last long enough in a game to earn a win is questionable at the moment, this was an impressive short season after recovering from TJ surgery during the majority of 2019. Sure, his BABIP was probably inflated thanks to a high 25.6% LD%, but he wouldn’t last in the Majors if he maintained a LD% that high to begin with. So assume that BABIP declines, which will raise his LOB%. He’s an obvious AL-Only sleeper.

Frankie Montas
Metric(s) Underperformed: BABIP + LOB% + HR/FB
2021 Rebound? You betcha, but don’t expect a rebound to 2019 surface results! After a breakout 2019, Montas was a major disappointment in 2020. The funny thing is Montas dramatically outperformed his SIERA marks in both 2018 and 2019, then dramatically underperformed this year. This is how it’s supposed to work folks! That’s precisely why you need to look at SIERA, not ERA, when evaluating pitchers. This is especially true over small samples, as Montas hasn’t recorded a 100 innings pitched season. Montas allows too many line drives, which could be the culprit behind his high BABIP (besides potentially poor luck), while he suddenly suffered from gopheritis for the first time, excluding his ridiculous 2017 season in which he allowed a 26.3% HR/FB rate over just 32 innings. It’s true that Montas’ underlying skills faded this year after surging in 2019, but he didn’t deserve results this poor. He’s still armed with a mid-to-high 90s fastball and elite splitter, so don’t give up on him.

Madison Bumgarner
Metric(s) Underperformed: HR/FB
2021 Rebound? Nah. After departing San Francisco and their dreamy pitcher’s paradise, many were predicting gloom for Bumgarner. But did anyone expect this much gloom?! It wasn’t only a crazy 22% HR/FB rate, but his skills completely collapsed, as his SIERA jumped over 5.00 for the first time. Plus, his strikeout plummeted to just 15.8%. All this came along with a fastball that lost major velocity, settling in at just 88.6 MPH, a whopping 3.1 MPH drop from 2019. Given the bizarre season, there’s no way to know now whether this was the cause of the velocity loss or if age and an arm that has a ton of innings on it is to blame. Perhaps both? It’s difficult to be optimistic as he had always posted ERA marks well below his SIERA and it’s almost surely due to a combination of the Giants’ home park and their defense, both of which Bumgarner won’t benefit from again.

Robbie Ray
Metric(s) Underperformed: BABIP + HR/FB
2021 Rebound? Nopers. Despite becoming an extreme flyballer, Ray allowed his highest BABIP since 2016. Too many line drives and too few pop-ups are certainly to blame, but perhaps the insane walk rate also suggests he simply couldn’t locate his pitches at all and they ended up in more hittable spots. That’s complete speculation based on no actual research, but it makes sense logically. While he appears here as a HR/FB rate underperformer, he has been in the mid teens and higher (as high as 20% in 2019) every season since 2016, so this appears to be an inherent flaw in his pitching ability. I wouldn’t expect much improvement in these luck metrics, but really, fantasy owners only care about Ray’s future walk rate at this point. If that returns to the low double digits, or better yet, drops below 10%, and his strikeout rate at least stays this high, or even jumps back above 30%, then he should be back to earning some positive fantasy value.

Tyler Glasnow
Metric(s) Underperformed: HR/FB
2021 Rebound? Of course. Glasnow is one of the few truly really good pitchers on this list that you wouldn’t have expected to find. He was elite once again this season and even pushed his strikeout rate up another notch to an insane 38.2%. But home runs were a major issue, after he allowed just an 8.5% HR/FB rate in 2019. That gopheritis carried over into the postseason, which could give fantasy owners pause and make him undervalued in 2021 drafts. Since HR/FB rate was the only problem, and he only allowed a little more than 10% of the minimum number of fly balls needed for the metric to stabilize, there’s little reason to ding him for it. While I worry about his stamina and how many innings he could actually get through in a season, his actual skills should not be questioned.

Jack Flaherty
Metric(s) Underperformed: LOB% + HR/FB
2021 Rebound? Even in 40.1 innings, who would have predicted that Flaherty would end the short season with a 4.91 ERA?! Remember this was just nine starts, and over such a small sample of games, those luck metrics often run wild. His LOB% was undoubtedly suppressed by all the homers, but there’s little reason to think that will continue. Take your 2021 discount and leave your draft a happy fantasy owner.





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.

3 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Joe Wilkeymember
3 years ago

Boyd also had an above average SwStk% with a below average K%, so his SIERA should probably be even better. He’s not going to ever be really great, but he will be decent in deep leagues, especially playing in that stadium. Both him and Weaver had relatively large wOBA-xwOBA differences too on balls in play.

A horrifying stat for Bumgarner: his wOBA allowed for balls in play was .418, 13% worse than league average. A more horrifying stat for Bumgarner: his xwOBA allowed for balls in play was .436, 20% worse than league average, 4% worse than his actual, and sixth worst in the league out of 166 pitchers with 500+ pitches. He’s done.

That K% for Glasnow is almost assuredly going to come down in 2021, his swinging strike rate wasn’t that great. Also, his xwOBA for balls in play was more in line with his 2017 and 2018, with 2019 being the outlier. I’ve been a big Glasnow fan for a while, but I’m not sure I want to pay the freight on him in 2021.

Flaherty is an interesting case. I feel like both his K% and BB% are too low, and he plays in a sneaky pitchers’ park. He outperformed his xwOBA on contact by the biggest margin of the pitchers discussed (, but I’ve always felt like xwOBA does not capture Busch Stadium well, especially for fly balls. If you can get him at a discount, that’s great, but I’m not sure the discount is going to be big enough for me to be in.