Kevin Correia’s GB/FB

Earlier today, Marc Normandin and I were discussing Fausto Carmona and somehow the topic of groundball-to-flyball ratio came out alongside the amount of line drives Carmona is (or rather: isn’t) giving up. The agreement was that skews his figure when compared to other pitchers with differentiating line drive rates. If that point sounds familiar, it’s because Dave Cameron wrote a similar article in early April; as you can see here.

After sifting through the leaderboards a bit I decided to write about Kevin Correia. The 29 year old Padres’ starter has made 14 starts and currently features a groundball-to-flyball ratio of 1.71; a career best. Despite that, Correia is actually giving up one of the highest ratios of liners in his career. Correia’s 48.3% groundball rate is a career high figure too, though, which could make him a victim of scorer bias. Take a look at his fliner (FB+LD) rates:

2007: 54.9%
2008: 61.8%
2009: 55.2%
2010: 51.7%

And once you create a groundball-to-fliner ratio, here’s how Correia looks:

2007: 0.82
2008: 0.62
2009: 0.81
2010: 0.93

From Dave’s thread, the wise Tom Tango suggested that he prefers GB-FB per contacted ball since the majority of pitchers will fall between a close range of line drives given up. Now how does Correia look?

2007: 4.8%
2008: 1.3%
2009: 8.8%
2010: 20.1%

That suggests that Correia has gotten better since joining San Diego in getting groundballs instead of fly balls. Using GB-FB assumes that half of the line drives are really groundballs and half are really flyballs, thus canceling out their inclusion. What if we assume that most of Correia’s liners are really fly balls; say 75%? Here’s how the GB-FB breakdown looks:

2007: -2.60%
2008: -11.28%
2009: -0.83%
2010: 8.55%

The only major change between even this and last season is an increased usage in his change-up (nearly double the amount); otherwise Correia’s increased groundball tendencies are a product of small sample size, location, or sequencing. Without becoming overly loquacious, the intent here wasn’t to find the reasoning as to why Correia is getting more grounders (although I suspect location, sequencing, and sample size all play into the equation at various amounts) but rather to see if the improvement is real or just a mirage. It looks legit through these lenses.