Using “plus” stats to evaluate Reds pitchers

One problem that can arise in interpreting baseball statistics is the context — the run environment — in which they occurred. Hitting 40 home runs in one era was a different accomplishment than hitting 40 homers in another era. For that matter, how does hitting 40 home runs compare to other batters today. It’s nice to know, for example, that a pitcher strikes out 9 batters per 9 innings, but is that good or great?

With the rapidly changing run environment the past few years and strikeouts at an all-time high, are the numbers we usually think of as great, really great? Maybe they are now merely average. Context is important.

To address this concern, statisticians have created “plus” and “minus” stats for a wide range of metrics. Examples of those include wRC+ or FIP- which are adjusted and weighted against a league average number. For a while, there were a few plus and minus stats, but not many.

Now, FanGraphs has released plus stats for a wide range of metrics. We can now compare players across time and season in stats like walks, strikeouts, ground balls, pull-rate, isolated power and more.

Before we look at the Reds and specific plus/minus stats, let’s look at how they work.

League average is 100. Every point above or below 100 is a percentage point above or below league average. For example, a 115 BB%+ means a player walked 15 percent more often than an average hitter would have. A player with a 90 BB%+ means he walked 10 percent less often. Minus statistics work the same way. If a pitcher has a 74 score in ERA-, that means his ERA is 26 percent below league average.

Home Runs vs Fly Balls

The inspiration for this post came from Steve Mancuso’s tweet about Tanner Roark’s home run rate. Roark’s home run rate is low considering how many fly balls he allows. Immediately after Steve called this out, Roark proceeded to allow three home runs in his most recent start against the Brewers.

To highlight how we can use plus stats to dive into ideas like this, lets look at Roark’s FB%+ (Fly Ball Rate Plus) and HR/9+ (Home Runs per 9 innings Plus). With an FB%+ of 102, we know that he has allowed 2% more fly balls than league average. While that is higher than his GB%+ of 83, it is certainly close enough to average to be sustainable, especially considering his career FB%+ is 100.

The issue arises when we look at his HR/9+. Currently in 2019 it is a 56, meaning he has allowed 44% less home runs per nine innings than the average pitcher. Obviously, that is a huge number and would certainly contribute to Roark’s strong overall start to the year. As we saw on Tuesday, however, it may only be a matter of time before more and more fly balls start getting over the fence. His career HR/9+ is 87, so he has always done a good job minimizing the long ball. But a HR/9+ of 56 is most likely unsustainable over an extended period.

David Hernandez and Robert Stephenson are two other pitchers who have a favorable gap between FB%+ and HR/9+. Both have higher fly ball rates than Roark, with Hernandez’s FB%+ at 124 and Stephenson’s at 117. However, their HR/9+ are down at 37 and 66, respectively. These gaps, especially for Hernandez, are massive and appear unsustainable considering each of their career HR/9+ are much higher.

One Reds pitcher who has been a bit unlucky with the home run ball relative to their fly balls allowed is Tyler Mahle. He has allowed a below average fly ball rate (96 FB%+) yet has given up significantly more home runs than league average (130 HR/9+. While he has only thrown 220 innings in his career, Mahle has struggled with this prior to 2019 as well and may be something to keep an eye on as his development continues.

Strikeout Rate vs LOB%

There are several ways for a pitcher to get out of a jam without allowing any runs, Help from the defense or just getting lucky are definitely part of it. The most surefire way, however, is to get strikeouts. While not a direct correlation, it is safe to assume that pitchers with high strikeout rates will have higher left-on-base rates.

Do any Reds have K%+ and LOB+ that are out of alignment?

Those who have been more fortunate include Jared Hughes and Matt Bowman. Hughes K%+ (79) is 21% below average, though his LOB%+ of 102 is slightly above average. This does fall in line with his career numbers (K%+ of 77 vs LOB%+ of 108), meaning his very high ground ball rate (career GB%+ of 137) makes this a sustainable trend. Ground balls can be a great way to get out of an inning especially if they lead to double plays. Bowman, though only with about a third of the innings of Hughes, has a similar trend. It will be interesting to see if these two can continue to strand runners without the strikeout as a major factor.

Hernandez and Stephenson show up again here, this time on the less fortunate side. Each have well above average K%+ (128 and 137, respectively) but are both below average at stranding runners. A patterns seems to be forming here,  with both pitchers sporting well below average ground ball rates, the exact opposite of Hughes and Bowman. For Hernandez and Stephenson, when they are not getting strikeouts, they are more often than not allowing line drives and fly balls. These outcomes are much more likely to drive runners home.

ERA- and FIP-

As I mentioned earlier, ERA- and FIP- are nothing new to anyone who has been on FanGraphs or Baseball Reference. But with new plus stats like BABIP+ and BB%+, we can more quickly identify why players are underperforming or over-performing their ERA.

By far the player with the worst variance between ERA- and FIP- is David Hernandez. His 5.40 ERA looks unsightly (120 ERA-) while his 2.55 FIP is nearly 50% better than league average (FIP- 55). His poor LOB% is certainly playing a part in that, but so is his 0.362 BABIP, which is good for a 122 BABIP+. Allowing 22% more hits on balls in play than league average does not account for the entire variance here but it certainly explains some of it. As BABIP tends to normalize over larger sample sizes (Hernandez career BABIP+ is 97), that should help close the gap a bit.

Two high profile players with a better ERA- than FIP- are Amir Garrett and Luis Castillo. Both have pitched phenomenally so far for the Reds, with Castillo being awarded an All-Star selection and Garrett among one of the first snubs. For as good as they have been, both have high walk rates pulling up their FIP. Garrett is at an even 5.00 walks per nine innings (155 BB%+) while Castillo is right behind at 4.76 (150 BB%+). If they cannot get those totals under control for the second half of the season, it is likely that we start to see their ERAs rise. They should still be well above average even with some regression, but if they want to maintain their current dominance, cutting back on free passes is the first thing to focus on.

Matt Habel

Matthew Habel was born and mostly raised in Cincinnati and was always a Reds fan growing up. Ironically, he did not become die-hard until moving to Pittsburgh after college and experiencing the 2013 Wild Card game behind enemy lines. While the "Cueto Game" is one of the worst sports moments of his life, he became enamored with the analytics side of the game after reading Big Data Baseball and watching the Pirates organization end their postseason drought. He started writing for Redleg Nation in 2017 and has enjoyed continuously learning more about the sport. He currently lives in Portland, Oregon where he loves exploring the great outdoors. Find him on Twitter @MattadorHeyBull