We live in an era of short benches. Recently, MLB Commissioner Rob Manfred made news when he admitted that the league office has had discussions about new restrictions for teams around how they can use relievers (and how many). There’s nothing imminent, but Manfred cited the dominance of relievers in recent years and hinted that the move would bring a little offensive spark back into the game. Plus, we all know the complaints about innings where a team uses four pitchers. Because they have eight relievers on the roster.

The thing about an octopus bullpen is that it means that there are only a couple roster spots left for substitute position players. Four and even three-man benches are common. And that has another important effect on the makeup of a roster. Because there’s little room to carry both a dedicated backup middle infielder and a dedicated backup corner infielder. The fourth outfielder had better be able to hack it in all three spots out there. Or maybe the utility infielder is also willing to grab an outfielder’s glove and do his best out there.

Or maybe the regular position players need to be willing to move around a little bit. At the end of last season, I took a look at how many of these multi-position regulars there were in baseball. In 2015, 18 players played at least 81 games and played in at least five games at four different positions, and eight of them added a fifth position to their resume. We’ve gone beyond the days of the utility infielder who was the shortstop who had enough bat to fit on a bench and more than enough glove to fake it at second and third. Teams are starting to recognize that having a player (or five) around who can shift around makes things more interesting for the manager. I’ve estimated that the “Zobrist Effect” could actually be worth a half a win or so in the right circumstances. But first, you need a Ben Zobrist.

Well, why not grow one? Do teams actively cultivate these types of players in their farm systems? There’s plenty of incentive for them to do so. For one, even if they are being groomed to be the “third baseman of the future,” the future that we speak of might be in five years. Even if he projects to stick at third with no problem, in five years, his team might accidentally find itself with a pretty good third baseman already on the major-league roster. Might as well hedge the bet and get him some experience at second or in left field or whatever makes sense right now, even though he’s just in Low-A.

There’s also the reality of the fact that even among the lucky few minor leaguers who make it to the bigs, they are most likely to be bench/role players, at least at first, and will have to fill a utility role. So, why not train players for the reality that they will eventually face? And we know from previous research that even if a player is a highly trained shortstop, the idea that a player can simply step into second or third with limited experience—or even pick it at first—is flawed.

But then there’s another side to moving players around. Some teams don’t like to do it because players only have so much bandwidth. They are trying to learn the art of hitting, how to field one position, and how to be adult humans all at the same time. Trying to add another position or two to the mix might be overload, and frankly you don’t need to be good at two positions at the MLB level. But you’d better be handy with the bat.

So, I vote we take a look at what’s actually happening on those minor-league diamonds. Are teams getting all that they can out of their minor leaguers?

Warning! Gory Mathematical Details Ahead!
I looked at data for 2015, for just a snapshot in time. What might surprise you is that in 2015, among all minor leaguers who appeared in at least 20 games and logged at least 100 plate appearances, three quarters of them played more than one position over the course of the year. It might have been only once, but it seems that everyone dabbles at least a bit.

Now, one token appearance at a position does not make for adequate training. When we raise the bar to 10 games or more, it’s still a respectable number—45 percent—who could boast that they had spent time at more than one position. One in eight had actually spent more than 10 games at three or more positions. Even moving the bar up to 25 games, we still see that about one in six minor leaguers had at least two positions in which they got regular action.

One thing that we do know is that the most specialized position of all is catcher. Catchers do get to spend some time wandering around without all that stuff on in the minors (and occasionally in the majors), but for the moment, we will politely excuse them from our sample.

So, here’s a question. When is it a good time to start teaching a player a second position? To get some idea of how teams approach the question, I looked at the age of a player and then how likely he was to be a multi-position guy with at least two spots with 25 games. An interesting table:


Played 1 Position (25+ Games)

Played 2 Positions (25+ Games)

Played 3+ Positions (25+ Games)









































Teenagers are generally not allowed to roam freely about the field, but between 21 and 23, the percentage that stick to one position drops by 18 points. Now, some of that might be the fact that some (though not all) of those 22-year-olds are recently graduated college seniors who may have already done some double duty for State. But the fact that at 23, we continue to see a drop suggests that this seems to be the point where teams really begin introducing the idea in earnest.

As the years progress, we see that the ranks of the multi-positional players begin to swell. Now, it’s not entirely clear what the forces are (broadly speaking) behind these numbers sliding. Early on, it might simply be that teams don’t want to overload the kids. But as the years go on, and minor leaguers start to get to that “Why are you still in the minors?” age, some of the shifting to a more utility role might be the mark of someone who is learning a few new tricks to try to stick around and make it as a utility infielder, or perhaps someone who is realizing that to even keep playing minor league ball, he’s going to need to fill a utility role on the Triple-A team. And maybe he’s happy with it and he’s the kind of guy that the team wants to keep around because he’s a good role model for the kids.

So what combinations of positions do we see? I looked to see what position a player played the most often, and then looked at what that group of players did when they were roaming around.

Primary Position

Percent Who Played a 2nd Position (25+ games)

Most Common 2nd Positions (in order)






3B, LF, RF



SS, 3B



1B, 2B, 3B



2B, 3B



RF, CF, 1B






LF, CF, 1B

I doubt anyone’s surprised by this table, but I think that there are a few interesting things worth noting. Aside from catchers, the players who did the least two-timing were first basemen, possibly because first base is the last stop on the defensive spectrum, and it might just be that some guys are parked there in the hopes that enough bat will grow that it’s worth putting them on the cold corner. But the second-least-likely to be a multi-position player was the shortstop. Given that the surest path to finding a good utility infielder is to find a guy who can at least respectably handle shortstop, that’s a little surprising. Maybe teams are just hoping that their shortstops will stick there.

What is clear is that in the minors, teams seem to view their players in three buckets (four, if you count catchers). There are outfielders, infielders (2B, SS, and 3B), and first basemen. We commonly speak of a defensive “spectrum.” Indeed, our conceptualizations of value assume that defensive positions ride along a single sliding scale. This suggests otherwise. Instead it looks like there’s a defensive flowchart.

It seems that shortstops and center fielders are at the top. Shortstops flunk into becoming second basemen or third basemen, and third basemen eventually flunk into the last stop on the bus at first base. Center fielders tend to flunk into being corner outfielders, or if they can’t hack it out there, first basemen. This isn’t the sort of movement that is done frequently at the major-league level. But in the minors where the rules are made up and the score doesn’t matter and the whole point is to sort out who fits (and where) for the major-league roster, we see a different conceptualization of how things are when it comes to positioning. Our WAR metrics commonly make the implicit assumption that a player at a “lower” position on the defensive spectrum (such as right field) can be filled by a player at a “higher” position, like second base. Yet, we don’t see a lot of 2B-RF combos.

But that also means that the new breed of utility infielders who are willing to give it a go in the outfield are likely being sent to the majors with little to no experience in playing out there.

Now are there teams that place a greater emphasis on this sort of multi-positional skill? It’s hard to tell from the cross-sectional numbers, but again looking at 2015 (and this time, excluding catchers), we can see what percentage of their players were doing double duty (again 25+ games at a position).

Top 5

Percentage of Players

Bottom 5

Percentage of Players





Red Sox


White Sox














I’m sure that there are people right now who are trying to read the tea leaves on those lists and trying to figure out if multi-positional training is the domain of the Saber-savvy teams or not, but I don’t think we have enough data to really explore that. I wonder how much of any of this reflects a coherent philosophy (we’re going to get our kids reps in different spots!) versus random happenstance (we just happen to have some guys who were gifted athletes and circumstances called for them to spend some time at third). But there is quite the spread from top to bottom.

Finally I wanted to know how these multi-positional trends compared to the past, so I pulled data from 2010.


Played More than 10 Games at 2 Positions

Played More than 25 Games at 2 Positions








Preparing Them for the Real World
So this is one of those areas where there’s very little public work. We know that multi-positionality has plenty of value, and perhaps it would pay off if more teams were to encourage it in their regulars as well as the guys who profile as bench players. What’s not clear is whether teams have really explored this or thought about it more than just beyond happenstance. Maybe they have. It’s hard to tell from these data. For my next trick, I think I may also look at the other question of whether teams should encourage multi-positionality. Does it interfere with development at the bat? If it does, is it still worth it?

But in a world where there are more and more spots on the roster reserved for relievers and fewer for bench players, the need for those bench players to be jacks of all trades will be stronger. Maybe the team that figures out how to leverage that sort of training will be two steps ahead. Or maybe they’ll just have more interesting looking box scores.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
At the upper levels the Pirates have made a real effort to make infielders into multi-position players. At the Major league level Adam Frazier is the 1st homegrown player soon to be followed by Max Moroff & Alen Hanson. Right now Sean Rodriguez is the go-to guy, having played 1b, 2b, ss, & rf.
Much food for thought here. In the early years I believe that players were jacks of multi positions. We have specialized but maybe as you noted in this article we need to rethink that.

I would think the ability to play multiple positions well would be a capability that would vary across players. Some people need to have a single predictable role, or would get confused in constant switches, such as might happen with a platoon situation at two different positions.

Mind you, Manny Machado learned 3B is essentially no time to a very high level. I bet he could play anything. But I think that the ones that most benefit from multiple positions are not the superstars but an everyday player that creates cross-position platoon possibilities for his manager.
Looks like a typo in the table- 3bs are most likely to roam to 1b, 2b, and what? I'm going to guess RF with SS a close second (or actually fourth)...
Should be SS
Wonder if a bigger issue than number of multiposition players on a team is whether there is a player who can play multiple positions well. Last year the Indians had Mike Aviles play 300+atbats all over the field and this year it is Jose Ramirez. Ramirez is hitting well and seems comfortable at at least 4 positions, so there isn't as much need to have other multiposition players. With a small bench 1 really versatile player can really ease the manager's job as long as he can stay healthy. However I agree these players should be cultivated and appreciated. I
In some ways we are going back to the dead ball era, when rosters were limited to 18 players, and you had a bench of three after 7 pitchers: a catcher, an infielder, and an outfielder.

It seems to me MLB has two options: to increase the roster to 26 or 27, or limit the number of pitchers to 10 or 11. I favor the latter, because it would reduce the number of four-pitcher innings and still allow one-batter substitution when it really matters. For example, in the 1954 World Series, where Leo Durocher brought in Don Liddell, a lefty, to pitch to Vic Wertz, the one hitter on the Indians, who was really raking.

The result was Willy Mays's most famous catch. Liddell, when got back to the bench, said, "Well, I got my man."