Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

I have a fascination with super-utility players, the guys who can play anywhere on the diamond. Players like Tony Phillips, Ben Zobrist, or even Denny Hocking. They're so handy to have around because a manager can fill out a lineup with a little more flexibility and know that he has someone to fill whatever hole is left. He's a wild card that gives a general manager more choices when putting together a roster. He's the type of player who adds a little extra value that the box score— and WARP—don't really capture.

In 2000, BP alumnus (and full disclosure, a man with whom I have worked) Keith Woolner asked a series of questions that he titled "Baseball's Hilbert Problems." At the time, it was a compendium of questions that he felt would drive the field in the century to come. At no. 6 on his list, he asked the question of how we might actually value this sort of positional flexibility. Reading through the rest of Woolner's questions, we've actually made quite a bit of progress in the first eighth of this century on several of them. But not this one.

I'd propose that one reason that sabermetrics hasn't progressed in this particular area is that while we've gotten good at describing various pieces of a player's performance that provide value, we've only ever looked at the player himself. When a player hits a home run, we credit his account. But what do we do when some other factor about the player allows his team to do something positive more often? Maybe a player can play both first base and left field decently, and his team happens to have a right-handed-hitting first baseman who mashes lefties and a lefty left fielder who is unkind to northpaws. The team can construct a nice platoon, with much better results than playing either player full time, but only because of the versatility of the third wheel.

Getting the platoon advantage more often is a big deal. There's a reason that teams carry LOOGYs and go to all the trouble of making five pitching changes in an inning. The platoon advantage is worth about 20 points of OBP. Because our super-utility guy is so super, his team can take advantage of that more often, and it will perform better as a result. It's not a stretch to say that some portion of the credit for the team's increased production should go to the versatile guy for enabling that to take place.

I'd like to begin an arc of articles here at BP on the matter of what I'm calling "latent talent." It's talent that derives not from what a player does, but because of the effect that what he does has on the other players on his team. Can a team extract extra value from finding parts that fit nicely together and make something that is more than the sum of its parts? Is there an effect of Ben Zobrist for just being Ben Zobrist? I'd suggest that the answer is yes, and maybe the rewards can be even bigger than you might think.

The platoon-enabling issue is an obvious case. But I that think before we encourage each team to find its own Ben Zobrist, there's an issue to tackle. While a team could use the platoon to its advantage, even mid-game, it would require that our Zobrist actually ping back and forth between positions on a regular basis. We know that there is a "penalty" that hitters pay when they pinch hit. This will sap some of the advantage of the platoon. But we need to ask a related question. After the pinch hitting appearance, we will have one player just entering the field, while another has to switch between positions. Must we give back some of the advantage that we think we're getting to a "pinch fielding" penalty? Do substitute fielders perform worse than starters?

Warning! Gory Mathematical Details Ahead!
I looked at two types of events: infielders fielding ground balls and outfielders catching fly balls. I used the 1993-1999 Retrosheet database, since it has data on hit locations. For infielders, I looked at all balls hit in their immediate zone (i.e., the “6” zone for the shortstop) as well as to the zone immediately to their right and left. I calculated the percentage of balls that the infielder reached and turned into outs over the course of the season. For the outfielders, I used fly balls to the outfielder's primary zone, as well as to the zones immediately behind, in front of, to the left, and to the right of that zone. I turned that percentage into an odds ratio and took the natural log.

I then created a binary logit regression and put in the control variable of the fielder's general prowess at that position. I then coded for whether the fielder was the original starter at that position, or a substitute (who started the game on the bench), or a starter at another position who had shifted to this new spot. I ran a series of tests for each position and various comparisons. (Did it matter whether the fielder was a starter at this position vs. a starter who was from another position? Was there a difference between a transplanted fielder and a direct substitute?)

For infielders, there was a general pattern that for third basemen, there was some benefit to being the starter (rather than a substitute from the bench), in the sense that while numbers for first basemen, second basemen, and shortstops generally did not go near significance, the p-values for third basemen were around .20. For outfielders, there were no significant effects observed.

What it all means
For the most part, substitute fielders do not suffer the same problems in fielding their positions that a substitute batter does in hitting the ball. This should be interpreted as, "Players field their position with the usual efficiency that we see from them, no matter whether they started at the position or moved there." Our versatile guy might be decent at one position, and bad at another, and that has to be taken into account, but he won't suffer any ill effects of switching positions mid-game.

I should also point out that this is not license for managers to simply shift any player anywhere on the field. The players who made these moves did so because their teams felt that they were capable of doing so (or the team was desperate). But then again, those are the same versatile utility guys of whom we are already speaking. It means that managers can pursue these split platoons with reckless abandon because of this Zobrist Effect.

And now that we have an idea of what the Zobrist Effect can do for a team, we can get about the business of conceptualizing how much it is worth.

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
kenarneson
6/18
The A's use of Jed Lowrie, Adam Rosales, and Eric Sogard this year seems to indicate that the A's have reached a different conclusion from you.

Rosales and Sogard are in a platoon, and Lowrie plays every day. Sogard only plays 2B. Rosales and Lowrie play both SS and 2B.

When Rosales starts, he is at SS, and Lowrie plays 2B. But when Rosales comes into the game as a substitute, they leave Lowrie at SS, and put Rosales at 2B. I can't figure out why the A's would do that unless they feel there is some sort of penalty for mid-game position switches.
walrus0909
6/18
That is a really excellent example. Offhand, all I can think of is the effect on turning the double play: maybe they think that moving Lowrie midgame messes with his footwork. That seems like such a small percentage of plays as to be negligible; however, it is something Russell left out of his analysis.

Another possible overlooked aspect is throws from the outfield: even if moving between LF and RF doesn't affect which balls you get to, it might affect opposing baserunners. If, say, you moved Manny Ramirez from LF to RF midgame you'd expect to see a lot more runners going from first to third on a single.
lboros
6/18
The Cardinals' roster decisions in the middle infield this year seems like a pertinent "latent talent" case study. They only have one true middle-infielder on the roster, Pete Kozma (the everyday SS). But instead of keeping the deadweight backup MIF they signed this winter (Ronny Cedeno), they played the versatility card w Descalso (a natural 2B whom they stretched into a serviceable backup SS) and Matt Carpenter (a natural 3B whom they molded into an everyday 2B). The position-shifting ability of both Descalso and Carpenter enabled the Cards to jettison Cedeno and give his roster spot to Matt Adams, likely netting at least a full win in the standings over 162 games.
lichtman
6/19
Russell in your calculations, are you controlling for the pitchers, batters, weather, lighting, etc.? I would think you would have to. Entering the game after it has started implicates all kinds of contextual differences. Without controlling for them, I am not sure how reliable any results would be. What do you think?

Also, the pinch hitting penalty is easily explainable by two things: One, and probably the biggest factor, is that a pinch hitter faces a pitcher one time. Two, they tend to be injured or tired players.

I don't think there is any particular reason why a sub fielder or one who switches positions would do any worse than one who enters the game, although I suppose you could create an argument similar to the batter's situation where the fielder gets used to the park, lighting, batters, as the game goes on. I guess there IS a reason to think that a sub fielder might have some kind of penalty. Again, I am not sure that your test is valid one way or another without controlling for the changing context as the game goes on, unless somehow you did do that with your logit method...