Commissioner of Baseball Rob Manfred has made comments in the past about banning defensive shifts in order to increase offensive production. But as we delve deeper into the logistics, we're not sure how he'd do it -- or why. Let's take a closer look.
How we categorize shifts
There are many variations of defensive shifts. At Baseball Info Solutions, we have two principle designations for shifts:
1. Full Ted Williams (or TW) shift: TW shift is one in which there are three infielders playing on one side of the infield.
2. Partial Ted Williams shift: Two players are significantly out of their normal position but on their normal side of the infield.
If Manfred were to implement some kind of ban on shifts, the most obvious choice would be for him to eliminate only full shifts. It's easier to see whether or not three infielders are playing on one side of the infield as opposed to trying to determine whether or not players are "significantly" out of position on their regular side of the infield. The latter would require lines to be drawn to regulate whether or not a player crosses them.
So for the purpose of this exercise, let's say Manfred banned only full shifts. Let's examine what percentage of overall shifts would be banned.
Over the past five years, 60 percent of all shifts on balls in play have been full shifts, where all three infielders are playing on one side of the infield. As the table shows, the percentage of shifts that are full shifts has gone down this season compared with the previous two seasons. However, the overall number of full shifts is far greater than in prior seasons, and there's still a month left in the season. Teams are shifting against more and more players that are less extreme shift candidates (i.e. they pull the ball less than others), and these are the players that teams have a higher likelihood of using a partial shift against rather than a full shift, which leads to a lower percent of shifts that are full shifts.
Issues with trying to regulate the shift
Here's the problem: We're not even sure how Manfred would go about setting the rules. Consider these relevant questions:
What exactly qualifies as a shift? In order to ban shifts, Manfred would need to develop a clear-cut definition of what constitutes as a shift. He would need to draw lines that infielders would not be able to cross, and determining those lines would be extremely difficult. There are many plays when fielders are out of position due to a specific situation in a game, such as infielders playing in on the grass to prevent a run from scoring on a ground ball.
Would a ban apply to bunts as well? If Manfred banned movement of infielders, this could also have an effect on a corner infielder's ability to move in to prevent sacrifice bunts or bunting for a hit.
How is an infielder defined? While it's theoretically easy to ban three infielders playing on the same side of the field, could a team still bring in its left fielder to play in short right field when a left-handed batter is up and move the third baseman into left-field? This would give the same location of players as a shift, but technically there would be only two infielders on each side of the field.
The best question of all: Why is banning the shift even necessary?
The whole idea behind banning shifts would be to make more things happen when the ball is in play and ultimately increase offensive production. Guess what? It wouldn't.
Shifting actually affects a small percentage of balls put in play. Let's examine batters' performance with and without shifts. Infield shifts mostly affect balls hit to the infield, that is, ground balls and short line drives. The following table gives the overall batting average on grounders and short line drives, which we'll call BAGSL, over the past five seasons with a shift on, with no shift and overall.
So we can see that the batting average on ground balls and short line drives is higher with no shift than it is against the shift. Interestingly, the overall BAGSL with the shift this season has actually increased by 20 points since last season. Anyway, as the table shows, over the past five seasons, the batting average on ground balls and short line drives is 25 points lower when a shift is on than with no shift. However, the shift has been implemented on only 10.7 percent of the total plays that resulted in ground balls and short line drives. If we look at the overall batting average on grounders and short liners over the past five years, it's .260, just two points less than the .262 average on plays with no shift. This means the shift has had only a slight effect on the performance of batters on these types of plays.
The infield shift does not have an effect on the plays in which the ball gets hit to the outfield. Ground balls and short line drives have made up 50.6 percent of all balls put in play during the past five seasons. So if we put it all together, the shift has been a factor on only 5.4 percent of all balls put in play.
Let's look at this from a broader perspective by checking the performance on all balls put into play. Let's compare the batting average on balls in play (BABIP) when the shift is used, the BABIP when no shift is used and the overall BABIP.
We can see that the BABIP for the past five years with the shift is only six points lower than the BABIP when no shift is on. The overall BABIP is only one point lower than on plays with no shift. While both the BAGSL and BABIP are lower with a shift on than without, the shift's overall effect on the offensive production of the game is minimal.