Astros' Team Construction and Flyball Hitters

Last off-season, Baseball Prospectus theorized that the A's had discovered another market inefficiency: fly ball hitters.  The article, "More Moneyball: Oakland's Other Platoon Advantage." posited that Oakland's team construction strategy favored flyball hitters; fly ball batters tend to be more productive hitters, and are more likely to have a platoon advantage over groundball pitchers.  The fly ball hitters' advantage over groundball pitchers becomes particularly significant as opposing teams' show a marked preference for acquiring and developing groundball hitters.  The 2013 A's AL West champion gave the highest percentage of plate appearances to fly ball hitters of any team in the past nine years---by a wide margin. In 2014, Oakland was No. 1 in fly ball batting percentage for the third consecutive year.

Have the Astros undertaken a similar strategy?  This question came to mind when I wrote the sabermetric profile of Colby Rasmus last month.  Rasmus has a consistent history of hitting fly balls, and is known to be a good low ball hitter (which is often true of left handed batters).  As I reviewed video of his hitting highlights, I noticed that he frequently golfed low pitches for home runs or extra base hits.

Dan Farnsworth, a private hitting instructor who writes for Hardball Times, discussed the characteristics of fly ball and ground ball hitters in "Groundballs: A Hitter's Best Friend?" He points out that major league hitters, unlike many amateur batters, face very good defenses.  His view is that major league batters should be aiming to hit the ball in the air, because wOBA of flyballs is significantly higher than groundballs.  As he concludes:

The bottom line: leave swinging down to bad hitters at the amateur level, who have no chance at playing at the highest levels of the game. Otherwise, hit the damn ball in the air.

Batters who hit bunches of HRs are usually fly ball hitters.  In 2014, batter fly ball rate exhibited a 0.45 correlation with HR/Fly rate.  In other words, hitters with higher fly ball rates tends to hit more home runs per flyball.

So, let's look at the fly ball ratios for the Astros' hitter acquisitions this off-season. The league average fly ball rate is approximately 35%.  (Career rates shown below, except for Valbuena, whose ratio is for 2014.)

The Astros' player acquisitions appear to match the Oakland strategy.   Sure, this doesn't prove that a fly ball strategy eixsts.  The high fly ball ratios could be a huge coincidence.  Each of the acquired hitters has a groundball to flyball ratio significantly less than 1, even though the MLB average GB/Fly ratio is 1.51.  During the off-season, the Astros added the same type of hitters that the A's have been acquiring since 2010.

Platoon Splits Vs. Extreme GB or FB Pitchers

The Book: Playing the Percentages in Baseball (authors: Tango, Lichtman, and Dolphin) pointed out that most teams use hitter platoons based on left/right handedness, but that platoons based on batted ball tendencies are less common.  In theory, according to The Book, fly ball hitters could be platooned against groundball pitchers, because fly ball hitters tend to produce better results against ground ball hitters.  The Book, published in 2007, suggested that teams have not fully exploited platoon possibilities because other factors affecting team construction over shadow the platoon advantages of batted ball chacteristics.  The Baseball Prospectus article speculated that this platoon strategy is more viable as teams focus on developing ground ball pitchers.

Farnsworth's Hardball Times article explains why fly ball hitters hit groundball pitchers better than groundball hitters:

Flyball hitters tend to swing on a sharper uphill plane, matching the steeper downhill plane of groundball pitchers. This results in more true contact and driving through the ball on the same level, producing more hard line drives and well-hit fly balls.

I have used Baseball-Reference.com's splits page to compile Astros batters' 2014 performance against extreme ground ball and extreme fly ball pitchers. On-base percentage plus Slugging (OPS) is the performance measure.  B-Ref defines GB pitchers as pitchers in the top one-third of groundball rate, and FB pitchers as pitchers in the top one-third of fly ball rate.  The remaining pitchers who fall between these extremes are considered neutral. "t OPS+" is the batter's OPS for that split relative to his overall performance.  OPS+ is the batter's OPS for that split relative to the league's overall average OPS.

As with all annual platoon splits, sample size is an important limitation.  So take the splits, above, with a grain of salt.  This is particularly true of hitters like Springer, Villar, and Marisnick, who had less than a full season of batting.  Also, remember that the table omits performance against neutral pitchers (average GB rate).  For example, Evan Gattis has a low tOPS+ against both GB and FB pitchers; logically, this implies that he has a very high tOPS+ versus neutral pitchers.  Despite the low tOPS+, his absolute OPS+ against GB and FB pitchers is good.  But his OPS+ is even better against neutral pitchers.

Rasmus, Lowrie, Gattis, and Valbuena--all acquired this off-season--had an average OPS+ of 134 against extreme groundball pitchers, compared to 98 for the average American League hitter.  (I excluded Conger, since he was likely acquired for his defensive skill.)

Based on the splits from 2014, the Astros' lineup will be tough on extreme fly ball pitchers.  However, extreme ground ball pitchers will be more challenging.  The newly acquired hitters should help though.

Taking the splits at face value, some platoon possibilities: