From an offensive standpoint, the first half of the Washington Nationals’ 2014 has been fair to middling. Ranking sixth, seventh, and tenth in weighted on base average, weighted runs created plus, and wins above replacement, respectively, in the National League, the team thus far as produced runs at a slightly disappointing level, given the level and depth of hitting and run producing talent the lineup carries. Despite this mildly disappointing aspect of the Nationals’ 2014 season, the team has remained within shouting distance of first place in the NL East, making the expected unfulfilled, at least, as of yet.
A statistic that can be used to gauge the variation between expected and observed tendencies in hitting and help discern whether a spike or a slump in production is a product of skill or some other variable is batting average on balls in play, otherwise known as BABIP. Simply put, it measures how often a ball put in play by a hitter ends up a hit by taking their batted ball profile into account. As a rule of thumb, BABIP sits around .300, but can vary greatly between players and even between individual player seasons. From BABIP, additional calculations can be performed to derive a hitter’s expected BABIP (xBABIP), which can further refine the ramifications of a batted ball profile. While there are a number a methods to calculate xBABIP, the following is felt to be the most accurate:
xBABIP = 0.392 + (LD% x 0.287709436) + ((GB% – (GB% * IFH%)) x -0.152 ) + ((FB% – (FB% x HR/FB%) – (FB% x IFFB%)) x -0.188) + ((IFFB% * FB%) x -0.835) + ((IFH% * GB%) x 0.500)
…where LD% is line drive rate, GB% is ground ball rate, IFH% is infield hit rate, FB% is fly ball rate, HR/FB% is home runs per fly ball rate, and IFFB% is infield fly ball rate.
With the combination of BABIP and xBABIP, some of the more finicky aspects of a player’s season can be parsed out and determined as something that is indicative of a player’s skill, or something outside of his control and is one way to take stock of player performance at the halfway point and determine whether a streak or a slump will carry on into the summer months. Below, I have provided the career (cBABIP), 2013 (BABIP 2013), and 2014 (2014 BABIP) BABIPs as well as the projected 2014 BABIP based on 2013 numbers and the expected BABIP for the rest of the season (xBABIP 2014) based on this year’s performance thus far for the eleven Nats hitters who have had at last 100 plate appearances this year. With these values, we can identify Nats hitters who might be due for an uptick or drop in production based on their batted ball rates thus far; this can also be compared to last year’s numbers as well as career values to find help determine whether the waxing or waning of their 2014 BABIP is something that could be indicative of skill, or perhaps other variables, such as an injury, a change in hitting approach, a change in pitcher approach, or how a defense plays a hitter in terms of alignment or shifting:
With the help of the color coding, we see that Ryan Zimmerman’s BABIP is pretty resistant to change, with the respective BABIP values over his career, 2013, and throughout this year staying within a couple of points of one another. On the other hand, Jayson Werth’s fantastic start to this year hasn’t fulfilled expectations that were in place using his final 2013 batted ball values, but is still in line with his career BABIP, which is encouraging. However, using up-to-date values and calculating his 2014 xBABIP, it appears he will possibly suffer a light drop in productivity. Adam LaRoche’s season has been a positive across the board in comparison to both last year and his career averages and appears to have the potential to get even better. We can also hope to see a over-correction in Denard Span’s BABIP later this season, eclipsing both his current and career BABIP.
The calculations for BABIP/xBABIP are based on batted ball data and as such, the swings in these values across and within a season can be caused by changed in one or many of these stats. Research has found that while BABIP itself does not correlate strongly year to year, metrics like GB% and HR/FB% can, thus providing additional layers of complexity when looking at the above table. With that in mind, provided below are each player’s change in the batted ball rates inherent to xBABIP, to help identify what is truly at the root of any egregious disparities in BABIP or xBABIP. First, differences between 2014 and 2013 data:
…and here, differences in 2014 data compared to career averages:
With both of these tables, positive numbers indicate 2014 data being an improvement over either 2013 or career averages. Overall, we see the volatility in year-to-year BABIP values reflected in the batted ball data, consistent with the effects of injury and game-to-game changes in hitting approach and defensive alignments being played out over a small period of time. Looking at the 2014 compared to career averages, we do see some significant changes in Denard Span’s ground ball rates, as well as with Bryce Harper’s HR/FB%; however, given the comparative lack of games played by Harper due to both MLB service time and injury, these values can be expected to swing a wildly as his year-to-year values for the moment. Other changes of interest include the career decline reflected in Nate McLouth’s numbers and the change in line drive and homer run rates for Wilson Ramos, possibly a reflection of an injury-marred career more so than a change in hitting philosophy.
Converting expectations into actual results is a precarious endeavor and can take unexpected turns during the course of a season; slumps, injuries, even the fashion in which opposing defenses line up for a given hitter can all make the most obvious and conservative of projections worthless, or at the least, frivolous. However, with xBABIP, we are provided a more refined and data-driven approach to prognosticating what’s in store for Nats hitter come the second half of the season.
Statistics courtesy of FanGraphs; current as of July 7th.