Are GB Spikes Sustainable?

At the top of the ground ball leaderboards for starting pitchers, there are three pretty familiar names – Derek Lowe, Trevor Cahill, and Jake Westbrook. All three rely heavily on their sinker and are primarily pitch-to-contact starters who rely on their infielders to make outs behind them. They have long track records of extreme ground ball rates, and their performance this year is nothing new.

However, coming in right behind Westbrook is James Shields, who is getting ground balls on 59 percent of his balls in play this year. Shields is nothing like Lowe, Cahill, or Westbrook – he’s a strikeout pitcher who throws a lot of change-ups and has never relied much on the ground ball in previous years. In fact, from 2009 to 2011, Shields GB% was 43.3%, slightly below the league average. Shields extreme spike in getting ground balls is fascinating enough on its own, but is compounded by the fact that he’s not the only guy who has transformed himself into a ground ball pitcher in the first two months of the season.

Over in Philadelphia, Cliff Lee is also seeing a significant uptick in ground balls, posting a 54.5% GB% this year after coming in at just 43.1% over the prior three seasons. Like Shields, Lee is also a high strikeout pitcher with one of the game’s better change-ups, but he has been incorporating a two seam fastball more over the last several years, and the pitch seems to be gaining in effectiveness. In fact, this isn’t the first time Lee has dramatically altered his batted ball profile – Lee went from being an extreme fly ball pitcher in 2007 (35.3% GB%) to a slightly above average ground ball pitcher (45.9% GB%) in 2008, the year he turned into one of the game’s elite starting pitchers.

Finally, Jason Hammel has also seen a drastic rise in his ground ball rate as well, and his early success has been one of the keys to the Orioles surprising start to the season. Hammel has jumped from a 45.6% GB% from 2009 to 2011 to a 53.8% GB% this season, while also seeing a dramatic increase in strikeout rate. Often times, pitchers have to trade strikeouts (which are often generated through pitching up in the strike zone) for ground balls (which come from pitching down in the zone), but Hammel has managed to get both more ground balls and more strikeouts at the same time. That’s a pretty remarkable change.

With three guys all seeing big shifts in their batted ball profiles this season, you might think this is a fairly normal occurrence, but in reality, these guys are defying a significant amount of recent history. GB% for starting pitchers has a year to year correlation of 0.85, the highest mark of any results-oriented pitching metric out there. The only numbers for starting pitchers with a higher correlation relate to pitch selection, as guys who throw a lot of change-ups in one year tend to throw a lot of change-ups the next year, and we don’t see a lot of massive changes in types of pitches thrown. Ground ball rate actually holds as strongly from one year to the next as rate of fastballs thrown, which shows just how few changes we actually see in a pitcher’s batted ball profile.

In fact, if Shields were able to maintain his current ground ball rate, he’d post the largest single year spike in GB% of any pitcher since 2002, the first year batted ball data was collected. However, we can look at how other recent pitchers who have experienced large increases in ground ball rates have fared, and whether they were able to keep performing at a new level for a significant period of time. The most notable ground ball spikers of the last 10 years:

Joel Pineiro, STL, 2009.

Pineiro had always been a slightly above average ground ball guy, but once he bought into Dave Duncan’s affection for the two-seam fastball, he really took off. His GB rate jumped from 48.5% in 2008 to 60.5% in 2009, and he posted the best season of his career as a result. He cashed in on his success by signing a two year contract with the Angels after the season ended, and he essentially split the difference between the ground ball rates he’d posted in the prior two seasons, coming in at 54.9% in his first year in Anaheim. However, in 2010, he went right back to posting a similar mark to what he’d done in his pre-Duncan career, and performed poorly enough that he hasn’t pitched in the Major Leagues since.

Charlie Morton, PIT, 2011.

Morton famously revamped his mechanics after the 2010 season, deciding to emulate Roy Halladay’s delivery in hopes to achieving success in the big leagues. The new delivery and pound-the-bottom-of-the-zone approach resulted in a GB% jump from 46.8% to 58.5%, giving Morton his most successful season in the big leagues to date. Because this transformation occurred just last year, we don’t exactly how this change will play out over the long term, but it’s worth noting that Morton’s ground ball rate (56.5%) this year is almost as high as it was a year ago, and at least for now, it appears that the change has stuck.

Johan Santana, MIN, 2004.

After spending four years bouncing between the rotation and the bullpen, Santana was finally given a permanent starting job to begin the 2004 season. He’d been the most extreme fly ball pitcher in baseball the year before, posting a 28.2% GB%, but in his first full-time gig as a starter, he cranked the ground ball rate up to 40.6%, just a bit below league average. Santana would go on to be a unanimous Cy Young Award winner that year, and he kept his ground ball rate around 40 percent until 2008, when arm problems began to erode his velocity. The slower-throwing Santana of the last four years has gone back to being more of a fly ball pitcher, but given the spaciousness of Citi Field, that could be also be a conscious decision.

Of the three most significant spikes in ground ball rate, Santana maintained his GB% gains for several years, Pineiro held on to part of his for a year before regressing and getting injured, and Morton has held onto most of his gains during the first several months of this season. Given what we know about the strong year to year correlation of GB rates, this is about what we’d expect – changes in batted ball rates tend to be much more real than changes in other types of results, as pitchers have a lot of influence over the trajectory of the balls they allow to be put in play.

So, what does this mean for Shields, Lee, and Hammel? If they can sustain significantly higher ground ball rates while still getting as many strikeouts as they are now, this could be the start of legitimate improvement for all three. Since we’re dealing with a smaller sample of just two months, we should still expect some regression back towards previous career norms, but they don’t have to keep getting ground balls at their current rates in order for these changes to represent legitimate steps forward. For Lee, this shift could vault him into the elite tier of starting pitchers (if he wasn’t already), while the extra ground balls for Shields and Hammel should help them continue to pitch well in a the toughest division in baseball.