Jeremy Hellickson on March 20, 2016

Photo credit: By Arturo Pardavila III on Flickr [CC BY 2.0 (], via Wikimedia Commons.

In doing some research for a piece about the Mariners’ staff on Today’s Knuckleball, I was trying to figure out why Taijuan Walker has allowed so many homers. In 103 innings this season, Walker has allowed a whopping 20 home runs. While he leans toward being a flyball pitcher, his 41 percent ground ball rate is not on the extreme low end. Safeco Field has been neutral for homers this season, but it does appear that Walker would be an even bigger threat to give up dingers if he pitched in a worse home environment. On the road, he has yielded 11 home runs in 37 2/3 innings.

In poring over Walker’s stats, I noticed something I didn’t expect. Because many flyball pitchers, from Max Scherzer to Justin Verlander to Marco Estrada to Chris Young, have high spin rates on their four-seam fastballs, I just assumed that Walker would have at least a slightly above average rate. According to Statcast, an average four-seamer this season has spun 2241 rpm, but Walker’s rate is a mere 2119 rpm. He rarely gets above 2300 rpm, meeting or crossing that threshold on only 1.2 percent of his four-seamers.

That led me to wonder if there is an inverse relationship for flyball pitchers between how much spin they put on a fastball and how often they allow home runs. In general, we expect pitchers with high spin rates to allow more homers than those with low spin rates, because their pitches tend to be higher up. Among flyball pitchers, however, those with higher spin rates may actually allow fewer homers, since they may be better able to create deception while working up in the zone.

Towards this end, I started with the 50 starting pitchers with the lowest ground ball rates among those having thrown at least 50 innings. Then I queried the Statcast data on Baseball Savant for the percentage of all four-seamers thrown with at least 2300 rpm of spin for pitchers who had at least 1000 pitches this season. Seven out of my initial 50 did not meet the requirements of that query, so I was left with 43 pitchers.

From this small sample, there wasn’t a clear linear relationship between spin rate and home run-to-flyball ratio (from FanGraphs). For example, among pitchers who got at least 2300 rpm on 40 percent or more of their four-seamers, some (i.e., Blake Snell, Julio Urias and Adam Conley) had extremely low HR/FB ratios below 9.0 percent, while Chris Young, Dylan Bundy and Matt Cain were at the other end of the HR/FB spectrum.

Among those who rarely threw their four-seamers at 23o0 rpm or higher, there was a little more of an intriguing pattern. There were 10 starters who reached 2300 rpm less than 4.0 percent of the time. Half of them had HR/FB ratios above 15.0 percent, and a total of seven were above 14.0 percent, which is decidedly higher than the major league average of 13.3 percent. Because his ground ball rate was a little too high, Walker didn’t qualify for this list, but in pitchers like Eduardo Rodriguez, Jeremy Hellickson and Chase Anderson, he seems to have company.

Screen Shot 2016-09-03 at 3.34.29 PM

Of the three outliers with lower-than-average HR/FB ratios, two would seem to be explained by park factors. Despite a lack of spin, Matt Shoemaker and Stephen Strasburg have been relatively stingy with the long ball, but both pitch home games in pitcher-friendly environments. However, both have higher HR/FB ratios at home than on the road, so park effects don’t help us to understand their relative aversion to homers. The third outlier, Anthony DeSclafani, does have home/road splits that make sense, as he has a higher HR/FB at hitter-friendly Great American Ball Park. Still his 13.0 percent HR/FB at home seems to be on the low side.

This exploration requires all sorts of caveats. It’s a very small sample, and many factors, including the rate of home runs on pitches other than four-seam fastballs, were not taken into account. It’s a first step towards looking at the impact that spin rate might have on a pitcher’s ability to avoid home runs. It may also help us to understand the likes of Walker, Rodriguez, Hellickson and many others who give up homers at an even higher rate than their batted ball profile would suggest.