Sep 17, 2013; Detroit, MI, USA; Seattle Mariners relief pitcher Carter Capps (58) pitches in the eighth inning against the Detroit Tigers at Comerica Park. Detroit won 6-2. Mandatory Credit: Rick Osentoski-USA TODAY Sports
Earlier this off-season the Marlins acquired Carter Capps from the Seattle Mariners in exchange for Logan Morrison. The move was seen as counter-intuitive by Marlins fans and the media surrounding the team. Many asked why would the Marlins trade one of their only power bats for a middle reliever that had a 3-3 record, 5.49 ERA and .289 BA against.
The standard stats only tell part of this story. Sabermetricians talk of “luck” as an important factor in arriving at how a season’s player panned out, the luck stats are the ones that the pitcher can’t control and in turn adversely his ultimate results, especially ERA.
These stats are often called “peripheral stats” because they plug into ERA and give it inputs. In the case of relief pitchers the most relevant of these are left on base rate also called strand rate, home run fly ball rate as well as strikeout and walk percentages. BABIP, batting average on balls in play, is also a major influence on a pitcher’s final ERA. FIP, fielding independent pitching, is composed of home run flyball rate, strikeout rate and walk rate
The purpose of FIP is to figure out a pitcher’s actual contribution when his isolated from his team’s defense, park effects, weather, so forth. But unsurprisingly, a pitcher with a high BABIP against corresponds equally bad to a pitcher with unfavorable FIP metrics. BABIP and FIP may measure different ways of getting on base and getting outs they still ultimately measure many of the same results.
According to linear weights a walk and a single are worth about the same and a home run is worth much more than a double or a triple. In short both FIP and BABIP measure a pitcher’s run prevention but in different ways.
The previous disquisition about deep sabermetrics is important because it informs why Carter Capps’ 2013 stats are so misleading. In 2013 Capp’s was extremely unlucky giving up a ridiculously high 18.8% HR/FB rate. That placed him in the same company as Brandon League, Heath Bell, Chris Perez and John Axford, who all sported FIPs over 4. Capps’ HR/FB rate, one of the three components of FIP was very high and thus skewed his FIP up. Although he let the ball fly out of the yard at worrying rate his strikeout and walk per nine innings was quite good at 10.07/3.5.
As well as allowing a lot of home runs, Capps was unlucky in the BABIP department allowing a .365 BABIP this suggests that hitters were hitting good balls and a 23.7% line drive and 36.2% fly ball rate supports this.
Capps in 2013 was unlucky that is undeniable, both in fielding independent metrics as well as in balls in play stats. Again a high FIP suggests bad luck and as does a high BABIP in 2013 Capps had the seventh highest FIP among qualified relievers and the second highest BABIP in MLB. These two factors fed to create a very ugly 5.49 ERA despite good K/BB rates and per 9 IP stats.
The best illustration of the importance of the fielding independent stats is Carter Capps in 2012. In 2012 Capps had nearly identical strikeout and walk per 9 IP stats, 10.08 and 3.96 respectively but he didn’t allow a single home run in his 25 innings pitched. Along with very similar strikeout and walk rates Capps sported an almost identical BABIP in 2012 at .357. The lesson is simple that with almost identical strikeout, walk and BABIP numbers but an extremely lower HR/FB rate an ERA can decrease massively from an indefensible high 5.49 in 2013 to a reasonable 3.96 in 2012.
Sabermetrics isn’t complicated and it values the inputs above as keys to building a reliable pitcher, a K/9 of above 10 suggests that he has good stuff but two consecutive years of BABIPs in the mid 300s bodes for a regression to the mean in the following years. It may not be as much as we wanted for LoMo but at least Capps has the potential to be integral in the Marlins seventh and eighth inning plans for years to come.