Depth, Platoons, and Versatility: (yet another) New "Moneyball"

Billy's up to his old tricks... - USA TODAY Sports

Much has been made of the A's depth and versatility in 2012 and even moreso in their offseason moves leading to the 2013 season. Melvin had the best mad scientist routine of his career: 3-4 platoons at any given time, a starting rotation in constant flux, and a bench full of players nearly as good as their starting counterparts. The fact that the A's attempted to maximize platoon advantages and build up the depth that allowed them to withstand injuries and ineffectiveness was seen almost as a new "moneyball"; a way to acquire a competitive advantage on the cheap. Beane has become king of collecting "assets" from the scrap heap, stashing them in AAA, then using them successfully at the MLB level.

This offseason, the A's signaled that they are doubling down on depth with the trade of the one-dimensional Chris Carter for the more versatile Jed Lowrie, and trading for Chris Young (who can play all 3 OF positions) rather than re-sign Jonny Gomes (primarily a DH).

But how much of an advantage do the platoons create? Is there any significant advantage to having players that can play multiple positions? How important are your 24th and 25th guys?

Tim Chan and Doug Fearing, two guys from University of Toronto and Harvard University, respectively, set out to put a value on versatility. Big ups to them. The result was a pretty interesting paper that caught the eye of Farhan Zaidi, A's Director of Baseball Operations.

From Athletics Farm:

"You know, I was at the Sloan Sports Conference this weekend in Boston, and one of the papers was about the value of roster flexibility. And just as an aside, just for your own edification, it’s an academic paper but it’s on their website, I think it’s worth checking out. And they talk about the value of having a roster built on players who are fairly interchangeable and can play multiple positions, because they’re able to be platoon players, maybe even across different positions, and because it insulates you against injury. And that’s the direction that our roster has been going in the last couple of years." - FZ

I really couldn't remember my last math class from 14 years ago (!) so much of this article was Greek to me. However, we don't need to know how the sausage is made. The results are what's interesting.

Tim and Doug studied position players only. They used things like WAR, UZR, and ZiPS to measure/forecast performance. While all of these have inherent biases, I think that's somewhat blunted because they use the same measurements across all players. We're comparing apples to apples at least.

[Note: Take any statistical commentary by me with a grain of salt. I think I got a C- in stats, because I couldn't spend my Tuesday and Thursday nights watching some guy drone on slide after slide behind an overhead projector. Please correct me if I read this paper wrong, and add any observations you have in the comments...]

There are a few hypotheses/premises: (1) platoons create more runs by taking advantage of matchups, (2) versatile players allow a team to replace an injured or resting player with a better player than they otherwise may have had available, and (3) having more versatile players means more options for the manager, which means more opportunities to optimize lineups/matchups/pinch hitting, etc. if you have a manager that's not Mike Scoscia.

The values were measured by runs above replacement (RAR) comprised runs saved in the field and runs added with the bat.

First, they measured the number of runs added for every team if there were no injuries. They noticed a solid positive effect in having more versatile players (ostensibly due to being able to maximize platoon matchups). The Dodgers were the biggest potential beneficiaries, being able to platoon 3 positions and gaining almost 6% more runs in the process. The A's were 8th by this measure, increasing their output by 3% (but again, this is assuming no injuries, so depth never really comes into play).

They also studied how often injuries happen, and how long they last, and then compared a "no flex" vs. a flexible roster, taking into account injuries. The A's graded out very highly, increasing their RAR by a whopping 12%. Increasing runs (or saving them) by 12% is pretty awesome.

Then, the kicker. The A's were apparently the best-positioned team to withstand injuries. The reason being that their bench and even minor league call-ups were nearly as good as their starters. So in other words, the A's would have an easier time overcoming bad injuries than other teams. It's not even close, the A's are so much better than all the other teams on this. Part of this is that the A's didn't have too many stars; they just had a number of very good players in 2012, and some atrocious players that were easily replaced.

Think Billy got tired of having injuries destroy his best-laid plans year after year?

Overall, "flexibility" (which was basically defined as having multiple players being able to play multiple positions) generated between 3% and 15% more runs than having a non-flexible roster (depending on the team). The conclusion was that the additional runs were generated (or saved) primarily through platoon advantage opportunities, and having more players be able to fill in at more positions in case of injury [click the table for a larger version].


Obviously, we all knew depth and flexibility was part of the A's strategy; many offseason articles and AN comments talked about it. However, I think this is the first time someone's really tried to get a measure on how important those qualities are in a roster. And the results are encouraging, at least for A's fans.

The A's 12% is nothing to sneeze at. By my rudimentary calculations, if you have a team with all average position players (i.e. 2 WAR players), that means your depth added 3 wins per year. That's the difference between AL West Champs and WC2.

Generating 12% more runs above replacement out of cheap players (who come cheaper because of platoon disadvantages, and the fact that they are backups or part-time players to begin with) is definitely a "moneyball" type strategy, and I'm glad the A's are once again showing that they are trying to find every possible advantage and crushing it on a low budget.