RADIO
The three best predictors of earned run average turn out to be strikeouts, walk, and home runs allowed. These are the three components of most advanced ERA metrics (like FIP, xFIP, SIERA, etc.). Adding on stats into these ERA prediction models can help accuracy a bit more, but most of the heavy lifting is done by these three categories.
For this reason, most of the pitching analysis you see from me will be concentrated on these three stats. In this analysis, I want to really zero in on home runs allowed.
It turns out that HR/9 is a decent predictor of ERA all by itself, as evidenced here by this scatter plot
Featured Promo: Get any full-season MLB and DFS Premium Pass for 50% off. Exclusive access to our Team Sync platform, Premium articles, daily Matchup Rating projections, 15 lineup tools, DFS cheat sheets, Research Stations, Lineup Optimizers and much more! Sign Up Now!
Importance of Home Run Rate
What you see here is every pitcher season from 2015-2021 (100 innings pitched minimum) plotted by the pitcher's ERA and HR/9. The trend line shows a positive correlation between the two variables, albeit far from a super-strong one. As a pitcher's HR/9 comes up, typically so does the ERA. The outliers are plentiful, which is to be expected.
The biggest "positive" outlier seems to be Mitch Keller's 2021 season, where he did a great job limiting homers (0.898 HR/9) but got pasted with a 6.17 ERA anyway. An extreme example from the other side of the line would be Tarik Skubal last season. He posted a decent ERA (4.34) with a miserable home run rate (2.11 HR/9). This fact alone will result in wide differentials between the ERA and FIP marks for these two pitchers.
That is not to say that Keller and Skubal should be expected to be similar pitchers this year, far from it. Skubal was much better in strikeouts and walks than Keller, which are two incredibly important numbers to consider. I am just setting the stage here to show that home runs are a big deal, and something we should focus on closely.
The Components of Home Run Rate
We can't just look at total home runs allowed here, as that is influenced a ton by how many batters a pitcher faces. Home runs allowed per nine (seen above) is probably the most commonly used statistic to level the playing field. I like to also use PA/HR, which is how many plate appearances it takes to hit one home run, on average. The league average for this number in 2021 was right around 30 (meaning that one homer was hit every 30 plate appearances). The worst pitchers will be below 20, and the best will be above 40, we will be using that metric moving forward so keep that in mind.
To hit a home run, a hitter must do (at least) these three things:
- Hit a ball into play (related stat: K%)
- Hit that ball into the air (related stat: FB%)
- Hit that ball quite hard (related stat: average exit velocity on fly balls)
These three pitcher statistics, therefore, have a significant impact on the home run rate. Let's take two opposite examples.
| K% | FB% | FB Velo | PA/HR | |
| Corbin Burnes | 35.6% | 22.2% | 88.9 | 93.9 |
| Mike Foltynewicz | 16.6% | 28.8% | 96.6 | 16.7 |
These two pitchers were at the poles of PA/HR in 2021 for pitchers that faced at least 400 batters. You can see how stark the differences are in every category there. For every 100 batters they faced, Burnes was striking out 19 more batters than Foltynewicz. For every 100 balls put in play, Burnes was allowing 6.6 fewer of them to be hit in the air, and when those balls were put in the air, he was shaving 7.7 miles per hour off the average velocity. It was no wonder that Burnes had a much, much better home run rate (although admittedly these two results are extreme and unlikely to be repeated by these guys in the future).
While we can narrow down home run rates to these components, this is not the whole story. Other things play in as well, like the strength of schedule, ballpark factors, and most notably - random chance. So what we want to do now, for 2022 fantasy purposes, is to find which pitchers fell on the two different sides of that random chance. Who should we look at as "lucky" and "unlucky" in terms of giving up home runs in 2021? Let's get to it.
Good Luck Pitchers
I have created an interactive scatter plot looking at these factors on Tableau. You can check that out here. I recommend you view it on a tablet or a computer as the plots aren't very conducive to a smartphone screen.
Everything we are talking about is represented in this plot. The y-axis is PA/HR, and the x-axis is FB%. Lower is better for both of these marks, so I sorted the axes in that way. The color of the dot represents the average FB velocity (more green is lower/better and more red is higher/worse), and the size of the dot represents strikeout rate (the bigger the dot, the highest the K%).
The "good luck" pitchers would be guys that had bad combinations of K% (smaller dots), FB% (to the left on the x-axis), and FB Velo (red dots), and yet performed well or average in PA/HR (higher on the y-axis). Here are the names that most stand out.
| Pitcher | PA/HR | FB% | FB Velo | K% |
| Mitch Keller | 46.9 | 25% | 92.7 | 19.6% |
| Frankie Montas | 38.9 | 25% | 93.7 | 26.6% |
| Dylan Cease | 35.4 | 31% | 92.7 | 31.9% |
| Luis Garcia | 36.8 | 29% | 92.2 | 26.2% |
| Trevor Rogers | 91.3 | 42% | 89.8 | 28.6% |
| Anthony DeSclafani | 35.5 | 45% | 92.9 | 22.5% |
| Freddy Peralta | 41.3 | 32% | 91.0 | 33.6% |
We saw Keller's name at the beginning here when we highlighted that his ERA was brutally bad while he posted a very nice HR/9. Upon further investigation, it appears that he should have posted a much worse home run rate given that he struck very few hitters out while giving up an average amount of fly balls and a very high exit velocity on fly balls. If you were hoping for the post-hype breakout from Keller after seeing the ERA and FIP differential, you may need to pump the brakes.
The rest of the names, save DeSclafani, are all very much fantasy relevant with ADPs inside the top-150. Let's take each of these names individually.
Frankie Montas - The big ballpark he played in helps and he doesn't have a high FB% (the average is right in the mid-twenties). The thing that stuck out here is the average strikeout rate and the high exit velocity. I would definitely expect some regression here for Montas unless he can raise that strikeout rate a bit.
Dylan Cease - He has always been a guy to give up a bunch of fly balls, and he did so again in 2021. The league average exit velocity on fly balls was 92.2, so he's not much above that mark there. The 32% K% certainly helps things out here, it's tough to give up homers when you struck out one of every three batters you face. The worry would be that he strikes out fewer hitters next year without improving on his fly-ball rate. I think both of those things are pretty likely given that a 32% K% is just tough to do, and he's long had trouble getting ground-balls. I'd bump him down a couple of slots here.
Luis Garcia - A middling strikeout rate with an above-average fly-ball rate, and yet a really great PA/HR allowed. He throws six different pitches, which helps keep hitters out of balance, so maybe that helps - but regardless I would expect some more homers next year.
Trevor Rogers - He was great when on the mound with a near 29% strikeout rate and the underlying metrics to back it up. However, the 91 PA/HR is just impossible, especially when you see the high fly-ball rate. This is another guy that is very deceptive as a lefty and plays in a park that suppresses homers, but things are not going to go this well for him in 2022 in terms of homers allowed.
Freddy Peralta - The K-rate is legit here, but giving up fly balls in that small ballpark makes it tough to post a PA/HR like this. I wouldn't expect a number above 40 again for Peralta, which could damage his ERA, at least slightly.
Bad Luck Pitchers
Now we're going to reverse course here and look for the pitchers that had bad PA/HR rates while having positive indicators. Here they are.
| Pitcher | PA/HR | FB% | FB Velo | K% |
| Andrew Heaney | 19.2 | 33% | 90.2 | 26.9% |
| Yu Darvish | 24.3 | 30% | 92.3 | 29.3% |
| Adbert Alzolay | 20.7 | 26% | 95.0 | 24.8% |
| Zac Gallen | 27.5 | 26% | 92.4 | 26.6% |
| Sonny Gray | 30.2 | 23% | 90.2 | 27% |
| Aaron Nola | 28.8 | 27% | 91.3 | 29.8% |
Some notes:
Yu Darvish - Gave up a fair amount of fly balls, but with the 29% K% and pitcher-friendly Petco Park, you definitely would have expected fewer homers surrendered. I am on the Darvish train again this year.
Aaron Nola - His 27% FB% was higher than we usually see from him, but the strikeout rate was elite and he didn't get hit particularly hard either. Some of this may have to do with the ballpark (Citizens Bank Park is one of the more hitter-friendly parks), but I would be pretty surprised if Nola doesn't significantly improve in this category this season.
Adbert Alzolay - The 25%-6% K-BB ratio was really encouraging for Alzolay, but unfortunately, his end-of-year stats were pretty bad. He had a middling flyball rate as well, which makes this bad home rate really seem unlucky. He did get hit quite hard, which maybe suggests some pitch-tipping or lapses in command, but I want to be on the buy-side of Alzolay this year.
Sonny Gray - He doesn't stand out much here since the PA/HR he posted was right at league average, and he pitches in the very homer-friendly Great American Ballpark. However, the 27% K% and low fly-ball rate show some signs that he could improve here in 2022.
Take a look at the interactive plot linked above and see who else may stand out. Reach out to me on Twitter for any requests or questions about what I've been doing here. Thanks for reading, RotoBallers!
Download Our Free News & Alerts Mobile App
Like what you see? Download our updated fantasy baseball app for iPhone and Android with 24x7 player news, injury alerts, sleepers, prospects & more. All free!
More Fantasy Baseball Advice
