Valuing Versatility

by $cHu on Nov 29, 2011 10:30 PM PST

I've been thinking for a while about the value that versatile players provide beyond their direct contributions. It seems that players who are able to competently play multiple positions help their teams win more games because it allows teams to maximize efficiency. For instance, lets say Mike Carp gets the full-time starting job in left field next season. If Justin Smoak needs a day off (or gets injured), he could essentially be replaced by Casper Wells as Carp could move to first base. If Carp couldn't play first at a somewhat respectable level, then the Mariners would have to use a roster spot for, and give playing time to, someone like Matt Mangini / Alex Liddi / insert replacement level first baseman here. Wells is better than a replacement level player, but the only reason that the Mariners could replace Smoak with him is because of Carp's ability to play first.

Players like Ben Zobrist and Chone Figgins (pre-collapse) who have been able to effectively play numerous positions have probably contributed more wins to their teams than their WAR would indicate because those players' versatility gives managers more options. Because of versatile players, managers can give more playing time to the best players on their benches rather than being forced to replace starters with the player on the bench who plays a certain position. It seems that flexible players would allow managers to platoon more effectively, and even give General Managers the flexibility to pursue players who are the best value without having to worry as much about filling a specific hole.

Has anyone seen research done to measure this hidden value, or have thoughts on how to measure the value that players who can adequately play multiple positions provide?

Tweet Comment 7 comments | 0 recs |

Do you like this story?

Comments

I think eventually WAR's built in positional adjustment accounts for this already, in theory

First thought – if someone was providing more than 2 WAR (more than league average) at a particular position, why would you platoon them to begin with? Which means to me, if most of the platooning is being done with people providing 0 – 2 WAR at a position, then really the platooning is just trying to lock in a ‘floor’ for that position, guaranteeing it’s no worse than about average. So it’s more about maximizing competence at a position rather than allowing efficiency to sneak in more value elsewhere in the roster.

More generally, I couldn’t find a study on your topic on a quick googling. But since positional adjustments are applied even/especially if the player’s played multiple positions, I’d think a starting point would be:

- identify WAR of utility players and/or players that spend time at multiple positions
- determine a cutoff point for what “competently play multiple positions” looks like (through your fave defensive metric)
- compare them to similar players who play exclusively at one position

It should put you in a position to say how common these competent multi-position people are, and thus whether it’s viable to develop or pursue versatile players vs. individual expensive monsters at each position.

It should also point to how many of these guys play multiple positions really well. I’d be surprised if you find a lot of people who were able to provide 2 WAR at one position, 2 at another and 2 at a 3rd one. I’d suspect most of the time even the utility players are most valuable at one specific position over others.

Overall, I suspect the conclusion will be “at the end of the day, wins are wins, so it’s irrelevant whether you got it through 1 fulltime guy giving a consistent 3 WAR while another position was close to 0 WAR, or got it through 3 guys platooning through 2 positions to combine for 3 WAR from those 2 positions”.

Apologies if this was just random unsupported blather. Only one cup of coffee, and no tables of players from fangraphs yet, this morning.

by Chris_FB on Nov 30, 2011 9:15 AM PST reply actions

I'd disagree with the first paragraph

It wouldn’t be a platoon in the traditional sense, but say you have four players. Assume all are equal at defense

1B/LF Prince Fielder
1B Adam LaRoche (Lefty hitter, not so good against lefties)
LF Casper Wells
LF Carlos Peguero

Now, assume Casper Wells absolutely has to be platooned for whatever reason. The team gets more value out of Prince Fielder if they can shift him to left, allowing LaRoche to play first, because the other option is to leave Fielder at first and put Peguero in left.

Since LaRoche>Peguero, Fielder adds value to the team by being able to accomodate that switch.

Determined, Jonesing Commentor

by Corco on Nov 30, 2011 12:03 PM PST up reply actions

Logical approach, but dang that'd be hard to look up easily

Also, it still feels more like establishing a ‘floor’ for certain positions than getting some huge advantage. More Fielder more often in that scenario probably is what… an extra 1, 1.5 WAR over the other combinations that would feature him exclusively at 1b?

by Chris_FB on Nov 30, 2011 2:50 PM PST up reply actions

It would depend on what the other options are

Think of it this way- a team definitely wants to put the best 9 players it has on the field for any given matchup. Sometimes, you’ll have two quality players that share a position, so that value is maximized if one of them is capable of playing somewhere else.

Example- in 2004, after Bucky Jacobsen went down, there was pretty much no question that four of the best players on the team were Ichiro, Reed, Winn, and Ibanez. Edgar was going to DH no matter what. Essentially, on a roster without much major league talent, there were four good players for three positions. Since Ibanez could play first, we were able to put him there so that Reed could get in the lineup. If he couldn’t, we would have finished out the season with Jolbert Cabrera at first.

Now, the upgrade from Cabrera to Reed may not be a huge upgrade (it actually might be if you assume Ibanez was as good at 1st as he was at LF- which I’d bet in 2004 was the case, and Reed>>>>Cabrera at playing the field), but an upgrade is an upgrade.

An extra win or half win is an extra win or half win. A skill that adds a full win is a pretty valuable skill to have.

Determined, Jonesing Commentor

by Corco on Nov 30, 2011 3:32 PM PST up reply actions

I guess it is entirely dependent on the rest of the roster though- so I sort of see your point

A team full of Prince Fielders that field like Jack Wilson would get a lot less marginal gain from a versatile player.

Is it fair to say, then, that bad teams benefit more from versatility than good teams? Somebody with time on their hands should figure that out.

Determined, Jonesing Commentor

by Corco on Nov 30, 2011 3:38 PM PST up reply actions

I'd agree on bad teams getting more out of it, and on someone else doing the math

Also, I suppose another way to look at it is, how many PA does someone need to be able to do enough to increase their WAR? And conversely, how many PA can you take away from the average player until they’re unable to make their WAR any better?

Platooning could maximize the value because you’re optimizing matchups or presence in specific parks or health. But platooning could also limit value for all but streaky players on a great run or truly great players, as they just won’t have enough time to contribute more than their part of league average performance for that position.

by Chris_FB on Nov 30, 2011 4:25 PM PST up reply actions

I would assume that the value of versatility is directly related to the amount of WAR a team gets by freeing up a roster spot

and indirectly related to the value of resting a player. Mark Mclemore was definitely worth more to the Mariners in 2001, but I doubt the degree of difference is much more than the roster spot, and the value of resting a player unquantifiable. That’s just my guess though.

...and now I'm here

by CapSea on Nov 30, 2011 3:34 PM PST reply actions