What is the job of the major league position player when he comes up to bat? Simply put -- to create runs.

What is the job of the major league position player when he is fielding? Simply put -- to prevent runs.

But what about those characters who, since the American League's adoption of Rule 6.10 in 1973, are labeled "designated hitter?" The obvious presupposition would be that they have no job when his team is not hitting. However, I will argue, in fact, that his job is also to prevent runs. In fact, there are some players, not only in the American League, but in the National League as well, who, in fact, would have prevented more runs by being a designated hitter than by playing in the field. While the Gold Glove not officially awarded to Designated Hitters, I believe it should be -- thus, this award is a counterfactual award. So who are the (counterfactual) Gold Glove winners at DH for 2008?

Even among the statiest of statheads, it is acknowledged that there isn't yet one "perfect stat" for judging relative player value, or even simply for judging a player's offensive value.

Still, it would be nice to get a total picture of what a position player contributes, wouldn't it? Given the advances in publicly-available defensive stats, it seems more possible than ever to look at a player's hitting contribution in runs during a season (whether by EqR, bRAA, or some other system) and add/subtract your favorite defensive metric, whether it is Baseball Prospectus' Fielding Runs above average, Hardball Times' RZR/OOZ system, or systems like Dewan's or UZR.

[Naturally, I waited so long to finally publish this that something has come out for public consumption -- Fangraphs, due to the generosity of MGL,  has started publishing not only a version of UZR based on data from Baseball Info Solutions rather than the Stats, Inc. version that is no longer publicly available. In some circles, they are distinguished by calling them bUZR and sUZR, respectively. With the addition of year-by-year wOBA-based offensive linear weights above average, Fangraphs took the next logical step and added a total value stat in terms of Wins Above Replacement. I'll come back to this.]

So, it can be done, but not in a simple one-shot for the amateur baseball "fanalyst" such as myself. However, through a post on RBI for Fielders by Sky Kalkman at Beyond the Box Score (subsequently linked in my award-winning 'RBI for Fielders and the 2008 Royals'), I discovered Justin Inaz's utterly cool Total Value Spreadsheet (this was months before Fangraphs started publishing their own WAR values. Since then, I have discovered some other home-brewed Total Value stats, although Justin's is still my favorite).

You can check Justin's stuff out for yourself -- on his blog he has a series of entries that explains just how he came up with the pitching, defensive, and offensive run values for each. It's quite sophisticated. At some point, I want to get more into BaseRuns, but this isn't the time or place. On defense, Justin takes (pre bUZR-era) data from both BIS and STATS, INC converted it to +/- plays and then runs (he also did this for 2007 MLB fielding). He also prorated positional adjustments by innings per position played. Great stuff for the lazy man such as myself! This is quite impressive and has been used by the Hardball Times their discussion of the 2008 cyber-MVPs as well as the basis for Sky Kalkman's series of position-by-position player value review at Beyond the Box Score. Again, for something like this to be done by (as far as I know) a layperson who makes it openly accessible to all is something very valuable and cool. It probably isn't perfect* (he runs through all of his methodologies in this series of posts on his blog if you want to check his work), but it's free, I like it, and it's useful for my purposes.

What were those purposes? Well, remember, I was going to give Gold Gloves to the best DHs in both leagues (yeah, I know about the National League, just give me a minute). But first, I would like to offer up a somewhat random list of position player total run values (above replacement) from Justin's spreadsheet, without comment.

Albert Pujols 97.6

Chase Utley 76.2

Grady Sizemore 75.5

Alex Rodriguez 68.6

Mike Aviles 35.1

Derek Jeter 28.5

Justin Huber 4.1

Mark Teahen 0.4

Emil Brown -3.8

Jose Guillen -6.7

Ross Gload -12.3

Tony F. Pena -16.6

But what does this all have to do with the defensive value of DH?

It is often asked whether or not a player should be a DH, or if he hits well enough to justify being played at DH. This is understandable, given the name of the position. However, as pointed out above, run creation is only half of the job of any given position player. On defense the player is supposed to prevent runs.

But how would a designated hitter prevent runs? How can we judge which designated hitter prevents the most runs?

Originally, I was going to go into a big digression on positional adjustments here, but given the extensive posts along the same lines on Fangraphs based on Tom Tango's adjustments (Rob Neyer even mentioned it on ESPN.com), I can keep it relatively short (always a warning, coming from me, you'll learn).

*Side Note added a bit later: I realize some of this will be old hat for many readers. However, I recently was reading some of my comments elsewhere and realized that one year ago, I wouldn't have understood half the stuff I was talking about. So I want to make sure and not be pretentious for those readers who might not be checking Fangraphs every three hours.

Pretty much every semi-informed fan realizes, on some level, that all offense from position players is not created equal. In our era, a first baseman with an OPS of .800 is considered below average, while a catcher with an .800 OPS would be considered far above average. Baseball Prospectus's uses this basic fact as the basis for its positionally-adjusted offensive statistic VORP (Value Over Replacement Player).

However, this thinking also works for defense. For example, take two players generally considered to be poor defenders at their position: Derek Jeter as shortstop and Adam Dunn and left field. Just as they are both good hitters for their position, they are, um, "below average," shall we say, defensively. But still, the bad (or good) defense at SS, all other thing being equal, still makes for a more "more valuble" overall player than the equivalent numbers of runs above or below average of defense at LF.

That's a bit of a confusing way to put it. A run given up on a botched easy grounder is as bad a run lost because a shallow fly drops due to the outfielders lack of range. What I mean is that the player with, say, -5 defensive talent at SS is more valuble than the player with -5 defensive talent at LF (assuming equal offense -- we aren't talking about Dunn and Jeter in particular any more). But the issue is that the pool of players who can play -5 SS is much smaller than the pool of players who can play -5 LF. Moreover, research and common sense tells us that a player who can play at a certain level at SS is usually able to play at a higher level at positions further across the defensive spectrum -- 2B, 3B, LF, 1B, and so on. I won't go further into how this is established. Dave Cameron did an excellent series on this not long ago at Fangraphs building on the excellent work of Tom Tango.

This is the well-known defensive spectrum. How does this work itself out in run values? Justin works this out in detail in his discussion of positional adjustments. I won't get into it here. Justin uses Tango's adjustments (which Justin prorates by innings played per position) for runs over the full season:

+12.5 C
+7.5 SS
+2.5 2B/3B/CF
-7.5 LF/RF
-12.5 1B
-17.5 DH

Seriously, though, tell me what this has to do with this dumb award you invented

What we have above are the values of a player over a full season (or whatever benchmark you want to use -- for this article, I will use 150 games, as does Fangraphs). So how does this effect how we look at "who would save more runs as a DH than in the field?"

Usually, of course, the usual question is "does this player's bat play at DH?" And that's a good point. Over the 150 games we will take as our benchmark, can he produce 17.5* runs above average?

* One thing worth noting is that the DH adjustment is "really" -22.5, but 5 runs are given back to account for the difficulty of hitting of off the bench for final value of -17.5. This will come into play in a bit.

Leaving offense aside, let's go back to our hypothetical SS and LF who are both league average hitters. Of course, we'll say that the SS is a better player because SS's have a lesser offensive expectation. And, ideally, we'd take both offense and defense into account... but let's just look at it from a fielding perspective. If the SS is -15 runs/150, that's bad, but still, his fielding stat + his position adjustment is -7.5 -- or the equivalent of an average fielding left fielder. But if the LF is -15 runs, well, his field+position adjustment is -22.5, or the same as he might be expected to hit if he was a guy who couldn't adjust to hitting off the bench (see the note above)...

And so that leads us, finally, to the point of this project I started and left aside, months ago:

Which players would have saved the teams the most runs, overall, including what position they played and their defensive "skills," if they would have (or, in the case of the NL, if they had been allowed to, hence the counterfactual nature of this award) just DH?

The formula is actually pretty simple -- if the players defensive runs above or (in the case of these players) below + their defensive adjustment is greater than or equal to the DH penalty, then they would have been better off DHing -- they would have saved the most runs. And the player who is the "most" over the DH margin in each league should win the Gold Glove for DH.

When I first started this, I just used -17.5 as the DH penalty, as the 5 run/150 bonus for hitting off the bench comes in. However, given that most of these players aren't DHing much now, I thought that would be unfair. So the margin is -22.5.

Using Justin's spreadsheet, which already prorates positional adjustments (there will be some discrepancy from how I prorate, but it isn't that significant, and doesn't effect the order of the the rankings) and gives the defense run above/below average, it was easy. Well, it would have been easier when I started if I had better data-handling skills at that point. Simply put, I manually went through (yeah, it was a while back) Justin's spreadsheet and put the players with bad defense into my own spreadsheet, then added in the positional adjustments, and prorated for 150 games. From there I added in the 22.5 DH adjustment. If the score is positive, then, well, despite sometimes terrible defense, one can say they still "helped" their team more (leaving aside issues of who else on the team could hit or field better in some combination) by playing the field rather than DHing. However, if they still score negative -- in other words, if their fielding + position is lower than -22.5, well, then they would have saved their team more runs by simply leaving their glove at home.

Finally, the Rankings

By the time I got my stuff together and then got this gig, Fangraphs started posting bUZR.  So I decided that I would stick with my original plan, to use Justin's stats (since I love good home-brewed stats, especially cool ones), and then also generate the positional adjustments, and use bUZR, add it together, and prorate it myself (a big pain in the butt for players with multiple positions). Yeah, it took longer than I thought. But by picking through some stuff, I figured I could then "average" the GGDH scores (Gold Glove Designated Hitter Field+Position Adjustment + 22.5) to get our winners. Sort of like the BCS computer and the polls -- two systems combined for one ranking. And everyone loves the BCS system, right?

It was more difficult than I thought it would be. Without going into the details, when I did the original sheet with just Justin's stuff. I did most of it by manual entry from Justin's Google Spreadsheet into Excel... so there was some picking and choosing. some of the rankings I came up with are pretty diverse -- most of that is because different defensive systems come up with pretty different results. I tried to take out players with relatively limited playing time, since the sample sizes would bias things too far one way or another.

Keep in mind that defensive stats should really only be taken to indicate a player's "true talent" once there's a decent sample size over three years. Thus, this little exercise isn't much of a "true talent" evaluator, but a one-year award thing. Just because some player isn't on the list, don't take that to mean that he's a good defender. He may have just had a lucky year. And, indeed, you have to be really bad to be a -15/150 corner outfielder, or a -10/150 first baseman.

Moreover, don't take this to indicate definitively what a team should or shouldn't do -- maybe they don't have any better fielders, or the ones they do have can't hit. This is (mostly) in fun. And remember that while defense is still generally underrated, that my rankings are for the Gold Glove -- they  have nothing to do with a player's offensive abilities. I'm sure Tony Pena has a better field+pos. than Pat Burrell. I'm still going to go out on a limb and say that Burrell is a wee bit better as an overall player.

One final note: there are different methodologies for Justin's  numbers and how I handled the Fangraphs' stuff -- Justin's positional adjustments include time already spent at DH, whereas the Fangraph's stuff didn't adjust for that. I did a version of the stuff from Justin's numbers such that it "weeded out" the DH positional adjustments, but I ended up leaving it in because the numbers got even goofier, and also to reflect a team's recognition that a guy like Jason Kubel isn't exactly Willie Mays in the first place.

I have three worksheets on the spreadsheet. Things got "messy," so I hid most of the columns. I trust I've given enough information above so that my work can be reconstructed. The first tab is just an info sheet. The second sheet sheet is "JustGGDH" (Justin Gold Glove DH). The third sheet is the supplementary "bUZRGGDH" (bUZR Gold Glove DH). The fourth sheet is the big Kahuna -- the final, all-MLB rankings. Don't mind some of the players at the ends of the lists -- they are on there for various reasons that don't matter at the moment.

On each sheet, I've ranked starting  "best" DHs (worst def.+pos. prorated for 150 games based on 2008 stats) of 2008. The GGDH/150 score (at this point, it's not much of a value stat or metric, but a "score") numbers in red indicate that the player is below the -22.5 Dh threshold. If the number is in yellow, this indicates they are between -22.5 and -17.5 -- in other words, the player probably should be DHing if they can handle it halfway decently. The numbers in white spaces in the GGDH collun mean the player is "safe." Keep in mind that being "2" or "3"  here isn't that big of a compliment -- these players had a pretty horrible year on defense, and almost always at the "easier" positions. We're talking the worst of the worst here -- or, if they were to DH, the best of the best.

JustGGDH (using Justin Inaz's numbers): Brad Hawpe and Bobby Abreu just had "great" years this year. Truly amazing. Abreu's been on the slide for a while, but this year was a special effort. Don't know if he'll be able to repeat it. Hawpe has been outstanding for a while. Jason Bay may not get the credit he deserves due to switching teams, but a -9.73 JustGGDH doesn't lie, and  heck, Manny Ramirez (with all the numbers added up) is only in the yellow zone with a 4.26! And you can definitely see why the Royals had to get Mike Jacobs -- look at the runs he'll save (if he DHs), although they already have two good "yellow zone" DH candidates in  Jose Guillen and Ross Gload (no, that isn't a typo -- he's the man with the Gloaden Glove indeed). Oh, and nice work on the Delmon Young trade, Bill Smith. Shout out to Raul Ibanez  for not making the "red zone." And hey, Mr.  "only worth half as much as Rual on the free agent market" Pat Burrell didn't make either! Kudos to you, Ruben Amaro, on an outstanding off-season so far. A repeat is in the bag.

bUZRGGDH (using Fangraph's bUZR): not too many surprises here. Wow, as bad as Jason Kubel might have looked in Justin's rankings (yellow zone), prorate the bUZR for the innings he just played in the field and... WOW... stand aside, Bobby Abreu and Jonny Gomes (although I should probably  have left Gomes out due to relatively sparse playing time). Everyone agrees -- Mike Jacobs again looks like the Keith Hernandez of DHs. Pat Burrell somehow avoids the red zone.

and now, the moment you've all been waiting for...

Final Tally:

You are a witness to greatness. Your GGDH Red Zoners of 2008.

Everyone agrees -- Brad Hawpe is (or would be) the MAN when it comes to preventing runs as a designated hitter . Great job, Brad. Hand down Gold Glove winner for the 2008 NL. Honorable mention to Mike Jacobs -- his defensive talents will surely shine now that he's in the AL. Another shrewd move by Dayton Moore. He must have know that Jose Guillen would just miss the "yellow zone" cut by 0.98 points. Adam Dunn just misses the red zone -- better luck next time.

But what about the AL? Ryan Garko and Jonny Gomes made late charges based on bUZR, but still, Jason Kubel's monster bUZR season is simply stunning. Yet, I can't help feel this is unfair to Bobby Abreu. He played way more games, and hardly even had the chance to DH. I'm going to give the veteran credit, and, after all, he may never have the chance to do this again, even with Raul Ibanez out of the AL...

Congratulations to Brad and Bobby, and to all the rest, good luck next year in your pursuit of this prestigious award!

Many thanks to Justin Inaz for not only letting me use his information for this nefarious purpose, but for answering my embarrassingly long, rambling emails about this. Sorry I took so much of your time, Justin! Thanks, though. Justin can't be held responsible for my silliness, of course. The GGDH is an award of my own imagination. devil_fingers and Driveline Mechanics cannot be held responsible for GMs making stupid decisions based on this award, although in many cases that might still be a better idea than what some GMs actually plan on doing.