The sabermetric movement in Major League Baseball has fundamentally changed the way the game is played and analyzed. Billy Beane and the Oakland A’s of the early 2000s brought advanced analytics to the attention of the public masses via Moneyball. Beane and his right-hand man Paul DePodesta (aka Jonah Hill) found that teams undervalued players that simply got on base, whether that be from a hit, walk, or hit-by-pitch — and thus, were able to acquire them on the cheap. Meanwhile, the rest of the league drove up the price of players that hit a lot of homeruns and had high RBI totals. Being a small-market team, the A’s could not afford such players and had to find creative ways to find successful hitters that scored runs. This led the A’s to acquire players that had high on-base percentages (OBP) regardless of their homerun prowess, defensive ability, or physical appearance.
As the rest of the league began to catch on, the A’s and other small-market teams had to continue to innovate. In past five years or so, the Pittsburgh Pirates and GM Neal Huntington shifted their focus to pitching and defense. They found that teams undervalued qualities such as defensive range (ability to get to balls that others may not reach), pitch framing (ability of catchers to turn borderline pitches into strikes), pitchers that allowed lots of ground balls (you can’t hit a homerun on a ground ball), and defensive shifts (most commonly known as aligning three infielders to the right side of second base). While all are interesting and purposeful, this article will focus on the defensive shifts and their impact thus far in 2016.
For most of baseball’s existence, defensive alignments have remained unchanged. Players were position equidistant from each other to minimize open space. But, as time progressed and more and more data entered the game, people much smarter than myself began to realize that this traditional way of position players may not be the “best” way. This was found to be especially true in the infield. Many batters tended to hit a large majority of their ground balls to one side of the infield — mostly to their pull-side. So, in an attempt to turn more ground balls into outs, teams began positioning three infielders to one side of second base. Like this example with a left-handed batter:
Coined as “shifts,” these new defensive alignments have become increasingly popular over the last five years. Per Baseball Info Solutions (a baseball data gathering firm), the number of shifts increased from 2,357 in 2011 to 13,298 in 2014. Why? Well, in short, analysts believe they work. Here is the breakdown of the % time each has shifted so far in 2016 (as of 6/14).
The defensive team’s goal of employing a shift is to minimize hits, and in turn, runs. You’ll see from the above table that this hasn’t been entirely true so far in 2016. The team that has shifted the most (Astros at 48.3%) have allowed more runs than the team that has shifted the least (Marlins at 14.9%). Obviously there are other factors that affect runs allowed, like the skill of the pitcher and defense, but the results are interesting nonetheless. For you visual learners (like myself) here is the relationship between shift% and runs allowed in graphic form:
Again, not much of a correlation at all. We would expect to see a significant downward sloping trend if shifting was working to prevent runs. Is this a case of simply only looking at less than half a season’s worth of data? Well, lets look at shifting from 2013-2015.
Still nothing. While others have proven that shifting decreases runs, I cannot seem to prove the point. If you for some reason didn’t click on that linked article, the takeaway is that employing shifts does work. In the first-half of 2015, shifting saved 190 runs across MLB according to Baseball Info Solutions.
While this article seems to go in circles and not come to any concrete conclusion, I hope to at least show you my thought process behind determining the effectiveness of shifts — which are becoming the norm in today’s game. Do they work? Well, I suppose that’s the question I didn’t really answer. Most of baseball seems to agree they do, and there’s data to back it up. However, my seemingly obvious analysis did not agree. Teams that shifted more did not necessarily allow fewer runs.
Now, I think the next step here to see is to look at the relationship between batting average (more specifically, batting average on ground balls) and shifts. One would expect that infield shifts would decrease the likelihood that a ground ball becomes a hit. That’s the goal of the shift anyway.
***P.S. I don’t mind Lebron.