To improve in 2016, the Yankee middle infielders should be more selective

The Yankees acquired both Didi Gregorius and Starlin Castro in shrewd trades with the goal of becoming a younger (and cheaper) team, one not so reliant on aging veterans and their propensity for injury and ineffectiveness. The Yankees project to have a young, dynamic duo up the middle this season, but Gregorius and Castro are similar players in more ways than merely their youth and agility. They also happen to have had almost identical offensive approaches last season.

Neither Gregorius nor Castro walk very often; in 2015, Gregorius drew a walk in 5.7% of his plate appearances, and Castro walked even less at a 3.6% clip. This is not necessarily a bad thing, as a lot of successful high-contact hitters also reject taking many pitches in favor of swinging early and often in the count. For example, Lorenzo Cain, who accrued 6.6 fWAR in 2015, walked 6.1% of the time.

However, unlike Cain, problems arise for Gregorius and Castro in their contact-driven approach because they don't make great contact most of the time. There are three types of contact: soft, medium, and hard. Hard contact leads to hits most frequently; to quote Fangraphs, "the harder a ball is hit, the more likely it is to fall in for a hit." Comparing Gregorius and Castro's contact percentages to both 2015 league average and contact-master Lorenzo Cain yields the following results.

As shown above, Gregorius and Castro hit an above-average amount of soft contact and a below-average amount of hard contact. For Cain, the opposite was true in 2015, which helped him post an excellent .347 BABIP.

Part of the difference between these three players' soft, medium, and hard contact rates can be attributed to the types of hits they are having when they make contact. Here are their percentages of line drives, ground balls, and fly balls in 2015 (not compared to league average, because league average statistics include those of various types of hitters, e.g. power hitters).

Cain had the highest line drive percentage, and line drives are the most conducive contact type to classify as "hard contact" and to fall for hits. It is intriguing that Gregorius' and Castro's hitting styles actually diverge a bit in this domain, as Gregorius hit more fly balls and Castro hit more grounders. Fly balls are not likely to fall for hits, especially when hit by someone of Didi's small stature. In Castro's case, his ground ball percentage coupled with his soft contact percentage show that while grounders are helpful in creating hits in general, this is not the case if they are hit softly.

Given all of this information about Castro's and Gregorius' affinity for soft/medium contact and lesser tendency for line drives, perhaps it would be in their favor to try to be more selective at the plate. Castro swings at pitches outside the strike zone 33.9% of the time, and Gregorius follows suit, swinging outside the zone at a 33.8% clip. This isn't necessarily an issue, but for Castro and Gregorius it may well be. This is because they each make contact on these outside pitches at high rates, Castro 67.9% of the time, Gregorius 62.0% of the time.

Contact on pitches outside the zone is notoriously less effective than that made on pitches inside the zone. Thus, in order to increase their rates of harder contact, it would be in Castro's and Gregorius' favor to limit their hacking on pitches outside the strike zone. This will allow them to make better contact when they do swing (on pitches inside the zone) and to increase their on-base percentages by walking more.

Walks, especially coupled with good contact, are important. In the wise words of Moneyball's Billy Beane, "You get on base, we win. You don't, we lose." It's as simple as that.