Fixing Miles Mikolas’ HR Rate

Cardinals starter Miles Mikolas was in the news this week when he was pushed back from pitching in a simulated game on Monday. Manager Mike Shildt said the move was “cautionary” and Mikolas was at the facility that day, stretching and doing cardio. On Tuesday, Jeff Jones reported the following:

The delay might just be “cautionary” but the fact that Mikolas is still not throwing to batters is concerning. Mikolas is still dealing with the effects of a forearm injury that traces to the end of the 2019 season. After pitching through discomfort late in 2019, Mikolas spent the winter resting his arm, hoping the time off would allow his body to recover. In February, it became clear that was not working. The club tried a platelet-rich injection to encourage recovery and shut him down for another month.

Then COVID hit and baseball went dark through June. When the club finally resumed workouts in July, Mikolas was still experiencing discomfort. With no reason to risk him during a lost 2020 season, the club sent him under the knife, hoping a corrective procedure would help him stave off Tommy John.

That should have been the end of it, right? Despite a full offseason to recover, Mikolas is still behind. He’s still not throwing to batters. While the Cardinals aren’t publicly pessimistic, the historical trend is not positive.

Even with the addition of Nolan Arenado to their lineup, the Cardinals are a team built around pitching and defense. While they have intriguing rotation depth, their entire pitching staff benefits when the top of the rotation is healthy and performing up to their potential. Having lost Dakota Hudson for the season last fall, and with question marks surrounding Carlos Martinez, the Cardinals need Miles Mikolas to be the stabilizing rotation presence he was in 2018 and 2019.

While acknowledging the unknown about his availability, let’s also acknowledge what he can be for the Cardinals when he’s healthy.

In 2018, the front office brought Mikolas back from Japan on the recommendation of Mike Maddux and their scouting department to compete for a spot in the rotation. He did not disappoint. Mikolas was legitimately one of the best starters in the National League, producing a 4.2 fWAR and a 2.83 ERA in just over 200 innings pitched. He barely walked anyone. While his K numbers were surprisingly low considering his heavy fastball and quality slider, he generated ground balls at a high rate and kept the ball in spacious Busch as well as anyone.

Nearly everyone expected a step backward in 2019. Mikolas’s walk rate stayed excellent and his K rate improved, but familiarity led to consistently harder contact throughout the league. He produced a solid 4.16 ERA and 2.5 fWAR in 184 innings.

What changed between 2018 and 2019 to cause a 1.33 rise in ERA? There honestly wasn’t much. The difference in run prevention seems centered on one specific area. Consider the following key indicators:

BB rate – increased from 3.6 to 4.2%. That’s a change of just .6 percentage points or three walks over a full season.

K rate – increased from 18.1% to 18.8%. That’s a marginal change and it’s in the positive direction. It pretty much balanced out the change in BB rate.

GB/FB ratio – decreased from 1.73 to 1.60, which represents an increase of just 1.1 percentage points in fly balls. That’s about .5 a fly ball per start; really not much.

BABIP Allowed – increased from .279 (below average) to .302 (around average). Ok, that kind of change would lead to more runs. But 1.33 per game?

HR/FB – jumped from 9.2% of fly balls leaving the park to 16.1%. Woah! Bingo!

That leaves a pretty clear picture. Mikolas gave up a marginally higher percentage of fly balls. He had more men on base than in 2018. A much higher percentage of his fly balls left the park. That would lead to a significant difference in earned runs allowed.

HR/FB rate, then, becomes the key to understanding Mikolas’ past and future.

In 2018, Mikolas was superb at keeping fly balls in the range of fielders – both at home in power suppressing Busch stadium and also in the less-friendly confines of the road.

Mikolas’ 2018 HR/FB rate: 9.2% overall
Home – 7.4%
Road – 10.9%

Compare that to what happened the next season:

Mikolas’ 2019 HR/FB rate: 16.1% overall
Home – 11.2%
Road – 21.5%

Nearly one-third of all of Mikolas’ homers allowed in 2019 came on the road – 10 at home vs. 17 on the road (27 total).

Doesn’t it seem odd that a pitcher who was so incredibly good at limiting HRs one season would become so exceptionally bad at it the next season? And that badness would be limited almost exclusively to the road?

After all, Mikolas’ 11.2 HR/FB% in Busch in 2019 is significantly higher than his 2018 rate, but it’s still very good. That change feels more like regression to the mean than poor performance. HR/FB rate is somewhat controllable by both hitters and pitchers, but there’s also more than a little luck involved as well. No matter what a pitcher does, a certain percentage of balls hit into the air will leave the ballpark. It’s just physics.

That reality led me on a little statistical adventure. Knowing that Busch Stadium suppresses homers, I wanted to see what a typical HR/FB rate was for Cardinals starters. My theory was that knowing the history of the HR/FB rates among Cards starters should help me determine whether I could project a lower HR rate for Mikolas going forward just based on historical trends and not on any significant change in performance. In other words, could Mikolas perform the same as he did in 2019 and still see a significant decline in HR rate and therefore, ERA, FIP, and fWAR?

I ran the HR/FB numbers for Cardinals starters from 2000 to 2020. These are the relevant returns for the last five seasons:

2020 - 15.9%
2019 – 16.4%
2018 – 10.3%
2017 – 14.9%
2016 – 13.0%

Before that, the Cardinals’ HR/FB rates were shockingly low. From 2010 to 2015, Cardinals starters did not have an HR/FB rate above 10%.

Before we start making too much of those 2010-2015 numbers, remember that the playing environment throughout the league has changed. Homeruns are up across the board. At the same time, are they THAT much up? In 2018, the Cardinals starters had a 10.3% HR/FB rate. In 2019, that had jumped to 16.4%. That feels excessive.

2020 was an odd season but it is also informative. The starter’s HR/FB rates were lower in 2020 even though the Cardinals were forced to use 10 different pitchers to start games and several of them – including Martinez (26%), Flaherty (23%), Ponce (19.5%), and Woodford (100%) – allowed extremely high HR/FB rates. Those four, forced to pitch under the most unusual of circumstances, represent 38% of the starts made by Cardinals starters last season.

We know individual performances can skew numbers significantly. In 2019 not only was Mikolas higher than expected, but Michael Wacha and Dakota Hudson were both over 20%. Those two probably account for the lion’s share of the increase witnessed in 2019.

We’re looking for trends here. 2019 and 2020 both saw significantly higher than normal HR/FB rates, tied directly to unusual circumstances and a few outlier performances. With the baseball potentially being less-juiced in 2021, isn’t it plausible that the Cardinals starters could see an overall downtick in HR/FB rate next season?

How would Mikolas benefit from this? Below is a simple sliding scale that leaves everything else about Mikolas’ 2019 season alone and re-calculates his ERA based on a declining HR/FB rate. This scale encompasses the highs and lows of his actual performance, ranging from the 16% he allowed in ’19 to the 9% he allowed in ’18.

16% = 4.16
15% = 4.01
13% = 3.82
11% = 3.52
9% = 3.28

That should reveal just how much impact HR/FB rate had on Mikolas’ regression in 2019. With BBs and Ks relatively stable, the normalization of his BABIP (i.e. more and harder contact) accounted for about a half run of ERA – 2.83 to 3.28 at the same HR/FB rate. That passes the math smell test. The rest of his regression is tied directly to his unusually high home run rate.

Where will Mikolas land this season? If he can hold steady with his K and BB rates, then the Cardinals should probably expect improved performance on home runs just by normalization to historical models. How much improvement is a matter of debate. I do not feel comfortable taking Mikolas down to the 9% HR/FB rate of 2018. I would comfortably land him in the 13% range.

If he does that, with all else about the same as normal, then Mikolas would produce an ERA in the 3.80 range, with an fWAR over 3.0 assuming 180+ IPs.

IF he can resume throwing soon and doesn’t suffer any more setbacks. That’s a big if. The Cardinals need it to happen or they’ll be lacking production in their rotation that they likely can’t make up.