Did you hear about that book Moneyball Billy Beane wrote about how all he does is draft college players and acquire guys who walk alot, and how he doesn't care about defense, makeup, scouts, and how he has a potty mouth? I thought you might have.

No, this isn't going to be another "hey, scouts and stats can get along, beer and tacos, man!" article, or anything else like that. Rather, I want to take a roundabout look at the (apparent) shift in emphasis in the on-line sabermetric community from on-base percentage to defense, and along the way examine a few players to see what would be make them more valuable: having a league average walk rate or being an average defender.

It should be clear to everyone who has read Moneyball that the first paragraph above was written with tongue firmly planted in cheek (although I'm not sure what that means when I'm typing rather than talking). Moneyball wasn't really about how walking is better, or how taking college players the draft is better, or whatever. It's not worth rehashing everything in detail or going into what extent Michael Lewis (the person who actually wrote the book) may have misunderstood and thus distorted certain issues.

The real point of Moneyball, let it be said for the nth time, is/was that the Oakland As were winning on a tight budget because they had found and exploited undervalued commodities -- that is, players -- on the market. In the mid 1990s, for example, that meant acquiring and playing defensively-challenged guys with low batting averages like the redoubtable John Jaha, the immortal Geronimo Berroa, and, of course, one of the acknowledged all-time greats, Matt Stairs. While many teams would pass such players over completely because they didn't like the strikeouts or the lack or range, the As picked them up at bargain prices and got plenty of value because those players had certain skills that were relatively undervalued at that time by other teams: good walk rates and isolated power.

Whether accurately portraying the As or Moneyball, in popular parlance, the slow, low-average, high-strikeout, power-hitting player became the Moneyball stereotype. It persisted (and persists?) no matter how many times it was pointed out that it wasn't so much a particular kind of player that defined the "Moneyball philosophy," as the idea that simply finding market inefficiencies was the point. And so on. I'm not here to talk about the past, particularly as regards this book.

The point, rather, is that if one reads around on sabermetric blogs and sites the last several years, it becomes apparent that there has been a serious shift in what people think an undervalued position player looks like. Defense is all the vogue. Granted, both critics and fans of Moneyball seems to forget that Lewis mentions the As looking at and developing new defensive metrics in the book as they realized that on-base percentage was no longer undervalued, and started acquiring players with a different skillset but whom were also undervalued. As Billy Beane said himself in a later interview:

...we're in a finite market and we're always trying to take advantage of any inefficiencies. Right now, you take on-base percentage and it's en vogue. It wasn't 10 years ago. We could get guys like Matt Stairs and Geronimo Berroa.... guys like Scottie Hatteberg. Now people are recognizing the value of that and they're paying for it. And if we're in a bidding war, we're going to lose that. So we have evolved. If you look at some of our first playoff teams, the '99 team that won 87 games, it was a power, on-base team. Now we're tops in the league in defense and pitching. For us, it's all about filling in on the backend and figuring out what people are undervaluing. You know, one day we're going to have a team with guys who steal 50 bases because people aren't paying for it. But it's all about wins. That's all that matters.

The As aren't the only organization doing such things these days, and the 2008 Rays certainly come to mind, as is well-documented. Indeed, while there are always going to be some GMs who overpay for, say, Raul Ibanez, the slow market for players like Bobby Abreu, Manny Ramirez, and Adam Dunn shows that perhaps the Day of Defense is upon us. Either that, or collusion, I mean, "the impact of the economy on baseball."

But back to the subject at hand: the baseball-obsessed intertubez have definitely picked up on the shift in emphasis. If strikeout-and-walk-and-homer heavy Adam Dunn is a typical icon of the "old" Moneyball, noodle-batted defensive wizard Endy Chavez is the icon of the "new."

To repeat, the issue is not that defense is more valuable than walks. It's that on the free agent market (until this past offseason, at least), it was less expensive to obtain. So the issue isn't whether a player with Adam Dunn's skills is better than a player with Endy Chavez's skills, but rather which would be more expensive to sign, and whether that's relatively efficient. Because they're closer than you think. How close? Let's assume that our Endy-esque player is just going to play the OF corners to make the comparison easier. Using CHONE's offensive projections for park-neutral linear weights above average per 150 Games, Dunn is +16 runs/150 -- a very good hitter. Endy is -18/150 -- not so good. That's 34 runs of difference. But now look at Rally/CHONE's defensive projections for outfielders. Dunn comes in at -13/150, subtracted from his +16 offense leaves him +3 runs above average, and that's before the positional adjustment. Chavez is projected for +20 defense in the corners, which makes him a +2 player before the positional adjustment -- only 1 run worse than Dunn.

Now, maybe you don't buy these projections for each player. But assuming they are accurate, if both were on the FA market this offseason, do you think they'd get anywhere close to the same money? Dunn is an awesome hitter, but Endy is about three wins better on defense. 'That is not something to sneeze at,' indeed.

But what is the point this rambling column? While everyone knows that Dunn has massive power, everyone knows that home runs are good. But the stereotypical "old" Moneyball approach finds that walks are/were undervalued, so that's interesting... while the "new" approach says that defense is underrated. Like "scouts vs. stats," only blowhard columnists think that one has to choose. A player with massive power, plate patience, and great defense is obviously the ideal. But sometimes you can only get a player with one or the other.

With that in mind, let's take some players who rate as very good or bad defenders, or good or bad walkers, and then to "adjust" their bad or good defense to league average, and/or bad or good walk rate to league average, and see what is more valuable in each case. I hoped to do a larger study, but at the moment FanGraphs (which rules, of course), doesn't have the sort of export options that make that very easy.

So here's what I did. FanGraphs' uses MGL's bUZR (Ultimate Zone Rating using data from Baseball Info Solutions), which  is expressed in runs. They also use Tango's year-by-year calculations for wOBA, which anyone with a SQL database can do. I took the custom linear weight for unintentional walks for 2008 (0.3142 runs). Then I selected different players from different pools. I calculated their walk rate versus MLB walk rates (initially I separated AL from NL, but for various reasons I decided that would be a bit of false precision in this case, since the custom linear weights are for all MLB, I think).

Since bUZR is expressed in runs, it is easy enough to see how each player would have been more or less valuble if their defense had been average. For walks, I took the players unintentional walks, and multiplied that by .3142, then multiplied their plate appearances by the MLB average unintentional walk rate (I got 8.7%) and subtracted the two to see how much their value in runs would have increased or decreased if they had an average walk rate. Who would have been helped the most? Keep in mind that because of the unsystematic nature of my selection of players, this shouldn't be taken to prove anything about what "kind" of player is more valuable. It's just (to me) an interesting and fun little toy.

Here is the spreadsheet. You can see that I didn't do it terribly systematically. I took the 2008's worst defenders, then some of the good defenders, and then some guys who don't walk too much. Yes, some of these categories overlap... Anyway, most of the columns are pretty self explanatory. "Defense" is Fangraphs' UZR in runs above/below average for the season. "BBDif" is how many runs from walks the player would be more or less valuable with an average walk rate over the same number of PAs. To compare the two for each player, simply reduce/increase the "BBDif" to zero and compare it to the "Def" number, then reduce/increase the "Def" number to zero and compare it to "BBDif." That's what I'm going to do, anyway.

Since I discussed terrible defenders recently, let's begin by pulling examples from that pool. Let's begin with our Gold Glove DH winners   for the NL and AL this past season. Bobby Abreu is a guy who is clearly on the decline in every aspect of his game, but still had a good season. While his walk rate was down, he still would have been 4.2 runs less if that rate was just league average. But, wow, that defense is bad -- -25.2 runs according to bUZR. So, yeah, the patience for which Abreu is/was famous is great, but maybe playing an average center field and merely having an average walk rate would have made him about two wins better. I feel bad picking on Brad Hawpe again, but, wow. Decent hitter, and had 8.2 runs worth of walks this year above average. However, if he'd just had average patience, and played average defense,  he would have been almost three wins better. Pat Burrell is an interesting case. While his defense was bad (-10.8), he's a walk machine, and with an average walk rate and average defense, he would have been less valuable, 4.4 runs less valuable. That said, if you look down this list, average defense would have helped a lot of good "old Moneyball" players with good walk rates and bad defense (we'll come back to this).

Looking at the list of good defenders, we see some guys with bad walk rates -- Cesar Izturis, Carlos Gomez. But in neither case does that problem outweight their outstanding defense. In fact, from our (admittedly selective) list, the only player who would lose value this way is Albert Pujols -- a great defender at first who almost beats the heavy  positional penalty for 1B with +10.2 defense this season, but who would lose 15 runs of value with a merely average walk rate. But (as in everything baseball), our selective sample has Phat Albert as the exception, not the rule.

How about some infrequent walkers who aren't (generally) exceptional defenders? here, in every case but one to which we'll come back, an average walk rate will help the player more than average defender would help/hurt him. But it does seem that the difference is much lower than for the bad defenders.

But those extreme cases aren't so interesting, since their defense is already in the middle (for the most part) anyway. Let's go back to the bad defenders list and find guys who were bad defenders in 2008 and had a below walk rate. In both cases, they could stand to improve. What would have helped them more? Take Matt Garza-bait Delmon Young. If the former uber-prospect could have walked at an average rate, he would have been 6 runs better. But check out that defense... yikes, can this guy do anything? Having average defense and the same lack of patience would have made him 18.4 runs better, or 12 runs better than an average walk rate. Jermaine Dye doesn't walk that much, either, but average defense would  have helped hm be 12 runs better than having an average walk rate. Robinson Cano is an exception in his all-around horrible year, as an average walk rate would have made him 9 runs better, while he was "only" 7 runs below average defensively. But for the most part, these low-walk, bad-defense players, all of whom had bad years in general, would have been better off being average defenders than average walkers. Yes, I'm looking at you, Yuniesky Betancourt.

Where does this leave us? Nowhere, in particular. Smart teams will keep looking for  undervalued skills to exploit, the league will start catching up with them, and they'll have to move on to the next thing. And then the internet will catch on. Yes, a broad base of skills is the best thing for a player. So far, it seems that bad defenders are generally hurting their teams more with their bad defense than they are helping or hurting with their good or bad walk rates, though. It's a hypothesis that needs further testing.

Let's conclude by looking at our two poster boys. Endy Chavez, as mentioned above, isn't exaclty a run creation machine. He would have been 2.4 runs better with an average walk rate. But no one would give that up for the 14.2 runs (almost 1.5 wins) he saved on defense in limited playing time.

Adam Dunn, the Big Donkey, is a fun player. In addition to his awesome power, he walks a lot. In fact, he'd be more than 20 runs less valuable if he just had a league average walk rate. However, I missed something in the bUZR spreadsheet I put together for the Gold Glove DH awards. You see, I exported the leaderboards, as one might expect. Except those leaderboards don't deal well with players who played multiple positions, or were traded during the season. So Adam Dunn didn't look atl that bad (if not very good) on the leaderboards. But I wish I'd looked at his player page, because now he's going to want a recount. For all of Dunn's  base-on-balls prowess, he would have been more valuable with merely average walks (-20.5) and average defense (+22.8).