Cristian Guzman and Position Changes

As Bill Ladson notes, the Nationals are giving strong consideration to the idea of shifting Cristian Guzman from shortstop to second base next year, due to what they consider to be diminishing range as he ages. For now, let’s put aside the fact that UZR doesn’t exactly agree with that sentiment and address what shifting across the bag will do to his value:

Nothing.

Over the last 10 years or so, the “advanced” statistics that became popular evaluated players against a position specific offensive baseline – VORP, for example. If a shortstop and a second baseman had the exact same batting line, the shortstop would rate higher by that kind of metric, due to the fact that second baseman hit better as a group than shortstops. As such, it’s become exceedingly common to see people write things like “he’s got enough offense to be valuable as a shortstop, but he doesn’t hit enough to play second or third”.

In fact, I guarantee you that someone will write that very thing about Guzman. The perception will be that moving from shortstop to second base will decrease his value. In reality, if the Nationals are right about his reduced range, it very well could increase his value.

Positions are essentially just a way to arrange players in a manner that produces the most efficient defense possible. You can literally play anyone anywhere – there’s no rule preventing the Nationals from sticking Adam Dunn at shortstop, for instance. They realize, however, that they will field a better team by minimizing the amount of times that Dunn has to move laterally in order to make a play, so they hide him at first base.

No one would think that Dunn would be more valuable if the Nationals lined him up at shortstop. The loss of defensive value would more than offset any gain the team got from getting an extra first baseman into the line-up. So, what I’m saying isn’t even controversial, though it may seem like it on the surface – everyone agrees that there is a point of defensive ability where a player’s value increases as he moves down the defensive spectrum.

If Guzman has lost significant range (again, we’re ignoring UZR and just assuming they’re right about this for the sake of discussion), then it is quite possible that the Nationals will get a larger benefit from reducing the amount of balls hit in his direction than they would by squeezing a marginally better bat into the line-up at second base.

This is where offensive position adjustment statistics, such as VORP, fail. If the Nationals would save themselves 5 to 10 runs a year by having a better defender at shortstop than Guzman, then they’re likely to get a net gain by moving him to second base (assuming that this better defender hits like a shortstop and not a pitcher, of course). And if the move produces a net gain for the team, then it’s impossible to accept that Guzman is getting less valuable in the process.

Now, the Nationals could be wrong about Guzman’s defensive abilities, and they could be marginalizing the value of a guy who UZR thinks is still a pretty decent defender at the position. But don’t just let someone tell you that Guzman is automatically losing value because his offense doesn’t compare as well to the average second baseman. That’s only part of the picture.




Print This Post



Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.


26 Responses to “Cristian Guzman and Position Changes”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Gaupo says:

    Excellent article. Puts the whole VORP/UZR discussion into perfect context.

    Vote -1 Vote +1

  2. Joe R says:

    I think (most) people get this, though, that average defense @ 2nd = sub-average defense @ SS. His VORP goes down, his defensive stats relative to his position go up, but surprisingly this can be novel to some.

    Vote -1 Vote +1

  3. snapper says:

    Dave, I’m a little confused by your analysis. I think you are conflating two different concepts: the change in Guzman’s value from the move and the net effect of the move on the Nats.

    If we lay it out mathematically,

    Guzman’s value change = Defensive Value (at SS) – Defensive Value (at 2B) – 5 (positional adjustment).

    The net value change for the Nats = [DV(new SS)-DV(Guzman at SS)] + [Offensive Value(new SS)-OV(alternative 2B)] + [DV(Guzman at 2B)-DV(alternative 2B)]

    Guzman’s value (in isolation) may go up or down, depending on how he adjusts to 2B. However, even if his value it goes down, it still might be a good idea to move him, if the net value change is >0. It all depends on the quality of the alternative 2B and SS.

    Is that what you mean?

    Vote -1 Vote +1

  4. John says:

    you might want to throw in something around this being applicable given that the other middle infielder can play 2nd or SS equivalently well and will hit the same playing either. Would bring home the point of maximizing the group versus the individual a bit. This scenario obviously wouldn’t make a ton of sense if you had fewer/lesser SS options for replacing Guzman than 2nd base options.

    Vote -1 Vote +1

  5. Randy says:

    I think the article is really discussing an accounting mechanism. Fangraphs uses a positional adjustment to calculate WAR; this serves to explicitly separate a player’s positional value from his batting value. VORP essentially adds those two components together into a single number on their way to calculating WARP. By my way of thinking, it’s a matter of personal taste which method a person prefers; do you prefer to count 100 pennies one at a time, or lump them into groups of 10 and count the groups?

    Vote -1 Vote +1

  6. Big Oil says:

    Ian Desmond looked pretty good last night (SSS alert, clearly), other than airmailing that throw in the 9th.

    Vote -1 Vote +1

  7. Phillies Fan says:

    Usually when someone says ““he’s got enough offense to be valuable as a shortstop, but he doesn’t hit enough to play second or third”, they are referring to a situation where an acceptable defensive player is moved to an easier position to accommodate a teammate. Consider Milwaukee, for example. If they wanted to keep both Hardy and Escobar, and moved Escobar to 2b (or Hardy to 3b) then it’s certainly the case that the value of the two players would be less than the sum of the values each would create at ss.

    The case where a player is moved down the defensive spectrum as he gets older is different. In this case, the move is an acknowledgement, though not necessarily a cause, of lower value. If Guzman’s ability to play ss has in fact declined then it may well be the case that he is now more valuable at 2b then he would be at ss, but he is certainly less valuable than he was when he was a more capable ss.

    Vote -1 Vote +1

  8. Doug Melvin says:

    “This is where offensive position adjustment statistics, such as VORP, fail.”

    That’s… harsh. Getting on VORP’s case for something it doesn’t pretend to be?

    Vote -1 Vote +1

    • Joe R says:

      I’m gonna give Cameron the benefit of the doubt here. I don’t think he’s getting on VORP as a stat, he’s getting on people who immediately think moving a player to a position where his VORP decreases hurts him and/or the team. Even if you do take defensive metrics with a grain of salt like I do, it’s common sensical that if a player had a +40 VORP at SS but butchered it at the position year after year, gets moved to 3B where he becomes +32 or something, but makes it back up by playing better defense relative to his new contemporaries, that it’s a lateral move.

      Just like when people get on Ichiro for playing RF. Well, his defense in RF and CF give him pretty much the same value. If he likes the corner, he can have it.

      Vote -1 Vote +1

      • Teej says:

        Yup.

        I don’t think Dave intended to hurt VORP’s feelings, but it’s a fact that VORP doesn’t incorporate the differences in defensive skill among players at the same position. VORP would tell you that Brad Hawpe has been a more valuable right fielder this year than Nick Swisher, J.D. Drew and Hunter Pence. I think that’s a flaw that’s worth pointing out.

        Vote -1 Vote +1

      • Joe R says:

        I remember once when I told a Rockie fan that J.D. Drew > Brad Hawpe, they came back by telling me that Hawpe could be a great CF and Drew couldn’t.

        I suspected prolonged drug abuse.

        Vote -1 Vote +1

      • Doug Melvin says:

        “I’m gonna give Cameron the benefit of the doubt here. I don’t think he’s getting on VORP as a stat, he’s getting on people who immediately think moving a player to a position where his VORP decreases hurts him and/or the team.”

        Fair enough. Danke.

        Vote -1 Vote +1

  9. Matt B. says:

    Let’s be honest, he isn’t a gamebreaker by any metric.

    Vote -1 Vote +1

    • Joe R says:

      I don’t think anyone’s calling Guzman a gamebreaker. He’s a very average player who’s probably fine for a platoon or one-two year fix.

      Vote -1 Vote +1

  10. Todd Boss says:

    I know everyone is in love with UZR, and that UZR doesn’t necessarily show that Guzman’s range is bad. However, i’ve watched most every game this year and I can tell you; he cannot go to his left. I’ve watched numerous balls hit almost directly to the SS position that he wasn’t within 5 feet of.

    2nd basement are not asked to make plays to their left that often; either a lefty pulls the ball into the hole or a right hander hits a weak grounder that way. Thus, moving him to 2nd makes perfect sense. The positions are interchangable skill wise and moving him there will make the Nats a better team.

    Vote -1 Vote +1

    • Joe R says:

      I wonder if anyone has a play by play index to verify that.

      If he does suck at going left, it’s a no brainer to put him at 2nd.

      Vote -1 Vote +1

      • nny says:

        Going by plus/minus, Guzman is +10 to his right, -6 strait on, -3 to his left. That’s plays. not runs.

        What’s interesting is that it was basically the opposite last year. He was +13 to his left and -6 to his right in 2008.

        Vote -1 Vote +1

      • Joe R says:

        So if we call those all singles, maybe throw a double in there, we can use .6 runs per play (+0.1 runs for the out, +0.5 for preventing the hit).

        So, +7.5 to +8 runs going right vs. left.

        One year sample size, halfassed analysis, seems pretty lateral to me, if the Nats have a solid SS option, could definitely be a move worth looking into. And if Guzman can still field his position (UZR and +/- says he can), he could very well play at a Pedroia/Hill level at 2B. Still can’t believe after the beginning of his career, that we’d be calling Cristian Guzman an above average player.

        Vote -1 Vote +1

    • brendan says:

      with dunn showing poor range at first, the nats 2B might make more plays to his left than the avg 2B.

      Vote -1 Vote +1

  11. John says:

    poor range going to the left is not good for a 2nd baseman. anything to their left that they can knock down is an out, going to their right is less of an issue because the throw back to first is difficult enough that even if they reach the ball its not guaranteed or even likely to be an out. On both of these I’m talking about the left and right extreme part of a 2nd baseman’s range. Poor going right or having a weak arm are the best conditions/reasons for the switch from SS to 2nd.

    Vote -1 Vote +1

  12. LeeTro says:

    While his UZR shows his range being average, his errors are atrocious (1 every 50 innings). I don’t see the Nats much, so I’m not sure what the problem is. If he has a majority of fielding or catching errors, the move probably won’t help. If they are throwing errors, then the move should help. I wonder if it’s possible to split up the errors category into “fielding errors” and “throwing errors” for UZR calculation in the future.

    Vote -1 Vote +1

  13. pm says:

    This is a big flaw in UZR. Guzman’s UZR looks great because Ryan Zimmerman is easily the best 3B in baseball. Zimmerman steals a lot of balls to his left (Guzman’s right) with CF-like range. Just to put it into perspective. Zimmerman leads all league 3B with 89 plays out of his zone. 2nd place has 49 plays out of his (Reynolds). That is almost double.

    Vote -1 Vote +1

    • BlackOps says:

      Wouldn’t that hurt Guzman’s UZR?

      Vote -1 Vote +1

    • Teej says:

      Yuniesky Betancourt put up historically atrocious UZR numbers while playing next to Adrian Beltre for years.

      From my readings of MGL’s explanations, UZR does not dock players for playing next to good teammates. And if it does, it’s to a very small degree.

      Vote -1 Vote +1

  14. Neil says:

    Guzman doesn’t even get close to many grounders that look like sure outs off the bat. Couple that with Ian Desmond’s long awaited breakout season, and I think the Nats have to at least try Guz at 2nd. They’re already paying him $8mil in 2010 so that would be preferable to, say, making a run at Orlando Hudson. But if the Nats have any hope of contention, they need to figure out second base which has been an absolute black hole in their lineup.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current ye@r *