Short-Stop Analytics #2: A Potential Evolution for the Shift

by Matthew de Marte – March 13th, 2018

Shifting has changed the game. Defensively, many clubs employ shifts in an attempt to prevent hits through the savvy positioning of fielders geared towards players’ spray charts. In recent years, the shift has become a massive part of the game, and every team employs it. Even the old-school fans of the game recognize that shifting is here to stay and is a fixture in the modern day game. Has shifting hit a creative wall? By this, I mean shifting has not really evolved.

Let me further explain. Organizations have information on players batted ball profiles that are available to the public, and I assume this is how they create the shifts they employ. Is it time though for advancement in the defensive shift? Players are trying to hit the ball in the air. This is no secret. So, against a player like Kris Bryant, Joey Gallo, or Trevor Story with no runners on base, why play four infielders when these players will rarely hit ground balls, especially to the opposite field? They rarely even hit ground balls up the middle. A first baseman and two infielders on either side of the infield should be sufficient. Look at the following spray charts for various Major Leaguers.

These hitters have an extreme tendency to pull the ball. They also rarely hit ground balls the other way. So why have defenses not adapted accordingly?

To me, there is only one purpose of the shift, and that is to place your defense in the most optimal position possible to limit the opposition’s chance of getting a hit. The current model of the shift is employed to take away hits and is focused mainly on preventing singles. Considering there are hitters with such extreme batted ball tendencies as the ones above, I think the shift can evolve. Instead what if we looked at shifting as a means to limit opposing hitters total production, not just singles? With the hitters above, we can be reasonably confident they will not hit ground balls to the opposite field. This kind of batted ball tendency leaves an excess infielder who plays up the middle. What I think the shift can evolve to is making this fielder a potential fourth outfielder or rover. It has happened before. Look at the following shift employed against Jim Rice in 1978.

Teams are looking for an advantage any way possible. This potential advantage would enable organizations to potentially take away extra-base hits. If teams were to employ a shift with four outfielders, they would have to make sure they are going to allow more singles to the point where the shift provides negative value.

While I do not have the tools currently to create a shift that could be an optimal four-man outfield shift, I imagine the image above would work pretty well for right-handed batters, and just moving the two infielders to the other side of second base would be sufficient for left-handed hitters.

I understand this would be a radical advancement and met with some criticism, but if a shift like this provides any sort of added advantage to a team, it should be employed. The success of this shift can be valued by how many hits and extra-base hits this four-man outfield would take away versus how many hits it creates. Now with the addition of a four-man outfield, there will be doubles that are taken away. So when evaluating the shift instead of looking merely at hits taken away versus hits given up, this model with RE24 should be used. RE24 measures the change in run expectancy from the beginning of a player’s plate appearance to the end of it. This would be more useful because it gives an exact measure when accounting for doubles compared to singles. Doubles are obviously more valuable than singles so it would have to be accounted for when evaluating a four-man outfield.

I think the four-man outfield will happen sooner than many who read this piece will anticipate. The moment when one organization employs it and its success is quantified, other teams will hop on immediately. Of course, there is the risk of the four-man outfield being detrimental to a team. Either way, with baseball savant data, organizations are becoming smarter and are understanding how to position players to take away hits better than ever before. Soon, this could lead to a four-man outfield designed to take away extra-base hits.

In the future, I hope to advance my argument from this piece to actual four-man outfield shifts computed through coding. For the purpose of sticking to the goals of the Short-Stop analytics feature, I believe it is more beneficial to introduce this concept before going further into detail about it.

Leave a Reply