Sabermetrics: wOBA

For my second sabermetrics post, I thought I would talk about a complex sabermetric that combines many simpler, more widely understood statistics: wOBA, or weighted On-Base Average. An improvement from the tradition triple slash-line, the goal of wOBA is to assess a player’s “offensive value” in measuring how capable a hitter is of moving himself and his teammates around the bases to generate runs.

Before I attempt to explain wOBA itself, however, it is important to understand the simpler stats that factor into wOBA. Batting average measures how often a player gets a hit, but does not factor in walks. On-base percentage measures how often a player reaches base, regardless of how. Slugging percentage weighs how many bases a hitter covers per hit, but does not include walks. wOBA attempts to combine all of these stats into a number that takes into account not only how many bases are covered per hit, but the odds of a hit also removing another play from a base, so it takes walks into account.

The formula itself uses average weights of how valuable a type of hit is to the overall possibility of said type of hit generating a score in order to correctly factor in all ways of reaching base (walks, hits, homers, getting hit by pitch) to calculate what a player contributes offensively.

With that in mind, I’ll let FanGraphs present the wOBA formula:

wOBA = (0.690×uBB + 0.722×HBP + 0.888×1B + 1.271×2B + 1.616×3B +
2.101×HR) / (AB + BB – IBB + SF + HBP)

**FanGraphs specifies that the weights in this formula are for the 2013 season, and change slightly each year.**

It is important to remember that wOBA does not take into account disparities in ballparks (some ballparks are considered “hitters ballparks,” meaning it is easier to hit a home run, for example, based on the outfield configuration/ length of the outfield than other ballparks) or in-game context (this includes runners on base or the score of the game during the player’s at-bat).

In general, an average wOBA is about .320, with anything above that being about average to excellent; excellent is .400. There are a number of MLB players currently with season wOBA averages significantly above even the excellent mark. They are all in close contention,  none of their names are surprising, and some have absolutely fabulous haircuts. Joey Votto (.429), Bryce Harper (.427) and Aaron Judge/Paul Goldschmidt (.426) are the current 2017 wOBA leaders. These players, then, are exemplary in scoring production solely from the plate–stolen bases and other on-base events do not factor into wOBA, so these guys are some true productive sluggers.

Cue Bryce Harper hair flip.

Sabermetrics: BABIP

I thought I would start with a stat that is slightly more obscure than, say, WAR, but is both useful and extremely relevant: BABIP. BABIP stands for Batting Average on Balls In Play, which is a measurement of how many non-home run “balls in play” end up as hits. In other words, the stat provides a numerical measurement of how many balls put in play by a specific hitter end up as, for example, seeing-eye singles versus ropes straight to the center fielder. The numbers factored into the BABIP equation include strikeouts, hits, home runs and fly outs (I cannot BELIEVE I’m doing this, but here is the equation below, courtesy of FanGraphs).

BABIP = (H – HR)/(AB – K – HR + SF)

“The numerator is the number of hits minus the number of home runs and the denominator is at bats minus strikeouts and home runs with sacrifice flies added back in.” – FanGraphs

There are three major on-field factors that influence BABIP: defense, luck and talent. If the batter is facing a more skilled defender, perhaps one withs faster reflexes or a stronger arm, then the ball in play might become an out rather than a single. The luck of whether a batter hits  into a shift, for example, or hits a ball slightly out of a fielder’s reach, is also a factor. And, of course, talent-based factors such as exit velocity and the ability to hit both sides of the plate also play a role in BABIP.

BABIP is a difficult stat to use on its own because of these uncontrollable factors that go into whether or not a ball in play ends up as a hit. Therefore, BABIP is best used in context with a given hitter’s average BABIP over many at-bats. Because the league BABIP average is .300 for a hitter, it is safe to say that a player whose career average BABIP is significantly above .300 is especially skilled at turning contact into hits. Whereas an altered BA over the course of a few weeks most likely signifies either a slump or a streak, the fluctuations of BABIP over a few weeks or a month could be the result of bad luck or good defense rather than talent.

The BABIP stat also exists for pitchers, but requires about a season more of stats to calculate than would an accurate career hitting BABIP. Pitching BABIP is also much more reliant on the pitchers’ team, as his own defense plays a significantly higher role in his BABIP.

Interestingly enough, both Jose Altuve and Aaron Judge, two vastly different players in terms of build, position and overall strengths, boast the third and fourth highest BABIPs in the Majors right now, respectively (behind Chris Taylor and Ben Gamel). Judge is a dominant outfielder and loping power-hitter; Altuve is a small and quick, but still boasts 15 home runs and a .367 season BA. It is likely that Altuve’s speed and Judge’s power (exit velocity) produce similar BABIP results.

In other words,  there is more than one way to put balls in play.

Baseball math

People love math. They love the objectivity, the specificity and the incredibly convoluted yet patterned and predictable answers that numbers provide. People feel comforted by numbers, wowed by the information they can provide, and interested in manipulating them in as many was as possible.

I am not one of those people.

So you can imagine how difficult it is, even considering my rabid passion for baseball, to delve into sabermetrics–because in the heart of my favorite subject lies an ever growing and ever important trove of insightful information based in my least favorite thing…math.

Up until now, I have mostly focused on the few numbers I understand: BA, ERA, OPS, etc. In other words, the old stats. I have explored, though hesitantly, fielding shifts, WAR and various StatCast home run speed/distance statistics–but I am hardly a regular visitor of FanGraphs.com.

And then I wrote a column last week about fatigue in the Cubs pitching staff following their World Series Championship year, and I just didn’t have the evidence that many baseball fans expect today. Though I included what I believed to be important stats, I was lacking in saber metrics, and my writing suffered.

It isn’t that people like me do not understand baseball, or the statistical direction in which baseball is moving. It’s just that we don’t like math.

So, in an effort to expand and share my knowledge of sabermetrics, I have decided to write one post a week explaining a different sabermetric. Look out for my first post next week!

I can’t believe I’m saying this, but let’s go do some math.