Baseball Prospectus is looking for a Public Data Services Director. Read the description here.

In my commentary on our Martin Prado mock arbitration case a few weeks ago, I wrote about why we might be undervaluing multi-position players:

It’s not as sexy a sabermetric frontier as quantifying catcher framing or clubhouse chemistry, but the value of multi-position players is another area in which cutting-edge stats are still struggling to capture the complete picture.

So how do you assess the value to a team of a manager’s peace of mind or a GM’s flexibility in filling holes? Right now it’s not something we can put a number on, but neither is it something we should ignore—especially in the era of eight-man bullpens, when multi-position players help dwindling benches pack more punch. “Utility player” has in some circles come to be a derogatory term. But Prado proves that there’s more truth to the term than the stats might sometimes suggest.

We’re a little closer to being able to put a number on it now, thanks to one of the Research Paper Finalists at the 2013 MIT Sloan Sports Analytics Conference held in Boston this past weekend, “The value of flexibility in baseball roster construction.” You can read the whole paper here (PDF). (Warning: intimidating equations.)

The authors, Timothy C.Y. Chan of the University of Toronto and Douglas S. Fearing of the Harvard Business School, drew inspiration from “the theory of production flexibility in manufacturing networks” in their attempt to quantify the contributions of multi-position players. What that means, in practice, is treating baseball players like factories that produce car parts:

Consider a car manufacturer with multiple factories and multiple product lines. Suppose each factory is capable of producing only one type of car and each car type can be produced in only one factory. If there is a sudden loss of production capacity in a factory (e.g., machine breakdown), the company may be unable to produce enough cars to meet demand. On the other hand, if the factories are flexible enough to produce multiple car types and each car type can be produced at multiple factories, then the company would be better able to respond to changes in supply or demand. Baseball players are like factories producing innings-played at each position.

It’s a clever approach, and it yielded some intriguing results. The authors studied how likely players are to get hurt, and for how long,* as well as how they perform in the field at their non-primary positions. Then they developed a model to calculate how much team performance suffers as a result of injury, and to assess the difference in the magnitude of the impact on teams with and without many multi-position players.

*In summary: older players are more likely to get injured, but not to miss more time when they do.

Their conclusion? Flexibility matters, not only because it aids teams in overcoming injuries, but because it helps them gain the platoon advantage more often. The most-flexible team in their simulations of the 2012 season (the Cubs) added 30 runs above replacement, or 15 percent of their overall offensive value, due to flexibility alone. The least-flexible team, the White Sox, added only six runs, or 3.4 percent of their overall output. Although they studied the effect of flexibility only on the team level, the authors note that the “same optimization model can be adapted to quantify the value of flexibility each player provides his team.”

If anything, the authors’ model is still underrating the value of multi-position players, since much of the benefit of a flexible roster comes in the planning stage. As I wrote in my Prado piece, “Not only do multi-position players enable teams to maximize the returns on their roster by helping them avoid replacement-level production when guys get hurt or face an unfavorable matchup, they also give general managers greater flexibility in putting their rosters together, warding off weak positions in the first place.”

That doesn’t mean just any old multi-position player will do: as the authors note, “Although it provides a team with options, flexibility alone may not be enough. A team needs flexibility and capability–that is, players with the ability to play multiple positions with skill.” But it certainly suggests that there’s real value in acquiring or developing players who can float all over the field, even if it doesn’t show up immediately in a team’s projected WARP total. Which reminds me of a few paragraphs from an article I read a few days ago:

Offensively, the Rays do things differently. For instance, this season, Maddon will have regulars at third base (Evan Longoria), shortstop (Yunel Escobar) and center field (Desmond Jennings.)

He'll platoon at every other position, playing matchups, defensive moves and other factors compiled from the reams of data supplied by the Rays' brilliant executive vice president of baseball operations, Andrew Friedman, and his staff.

"That's how we are built. We have to be built that way," Maddon said. "We can't afford the large contracts with 150 games per position. Andrew and the boys do a great job of presenting all these different scenarios. That's who we are. We kind of dig it."

There'll be more of the same in 2013. The Rays have 12 players in camp who've started at three or more positions and eight who've made starts in both the infield and outfield.

The Rays are already exploiting multi-position players to eke out a few extra runs from their roster, but any advantage they have in that area won’t last long. For many teams, the future might hold a bench full of Bloomquists.

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
Ben Zobrist is one of the most valuable players in baseball, I'm convinced. Versatility could well be the new undervalued asset, and that an infield-outfield guy like Jordany Valdespin or Grant Green (if their development continues) can make the eight-man bullpen much less onerous, or allow a team to keep a Jason Giambi type of specialist around.
We just had our auction draft and Zobrist went for too much money for me, but I targetted and got Prado, who is eligible at 2B, SS, 3B and OF in our league.

As for roster construction, he let me wait until the end of the draft to fill either 2B or 3B becuase I knew I could plug him in at either one. He saved my season last year when I could plug him in for Longoria while he was out.
That's sort of how I imagine it working for a real team, too. It's not just that a player like that can help you fill a hole in season, it's also that he allows you to be more flexible in putting your team together. With someone like Zobrist on the roster, a GM doesn't have to be locked into a need to acquire a second baseman, a right fielder, or a shortstop. He can wait and acquire whichever one comes cheapest or makes the most sense. More options, better leverage.
It also means a farewell to "Prospect who is blocked by _____." The Nationals are trying to make Anthony Rendon into a 3B/2B type for this reason.

This is part of why the Yankees look like they're in trouble this year. Not only are they thin on depth, they have a plurality of starters who are old enough to require frequent rest. They may be paying ~$200M and still getting a healthy dose of Eduardo Nunez, Matt Diaz, Juan Rivera, etc.
To be fair, this was one of the insights/case stidues that I normally enjoy in the BP annual team capsules that I missed this year. I expected to read more about how the Rays and Oakland A's play match ups with platoons to get more from their rosters (hello Jonny Gomes and Brandon Moss). Though this concept was mentioned it wasn't explored in any detail. Nice to see it being addressed here.
Duly noted, and I agree that it is a factor that deserves an in-depth look.

As the author of the A's chapter, I mentioned the platooning in the team essay but I made a point to address the specific platoon patterns of Moss and Seth Smith in their player comments (and added the lefty-mashing bit to Gomes). I also mentioned the team-wide tendency toward platooning under manager Bob Melvin's comment. I certainly could have gone more in-depth in one space, but instead I chose to build that theme within the chapter.

I expanded on the positional flexibility and the strategic platooning in the A's episode of Effectively Wild, and I made sure to cite the Rays as a point of comparison. I encourage you to check it out for a bit more detail and nuance, if interested.