What If RBIs Were An Accurate Stat?

I'll spare you the anti-RBI rant, as we've all heard it before.  But what if RBIs actually reflected a player's offensive production?  To answer that question, I took Justin's runs-above-replacement stat (RAR, which is BaseRuns-derived linear weights with park and league adjustments) and found the best-fit equation with RBI:

deserved RBI = 17 + 1.6*RAR

So, if RBIs were useful, here's how the leader board would look.  I've basically just taken RAR and translated it onto the RBI scale.

I don't know about you, but I love the idea that Brian Roberts could have more RBI than Raul Ibanez.

Here are the trailers, with a minimum of 400 plate appearances:

Uh, single-digit RBI totals?  Yes, those guys have been just that awful.  Remember when Luis Castillo couldn't pass the 40 RBI mark?  Yeah...  And did anyone predict that the Yankees would have two of the worst full time hitters this year?

How about the hitters who are most overrated by RBI?

That list is pretty much dominated by low-walk, middle of the order hitters.  I'd make fun of Ryan Howard, but it's really just sad how far he's fallen in two years.

And, finally, who are the hitters most underrated by RBI?

And here you've got a whole bunch of high-walk leadoff hitters, plus some huge bats having phenomenal years.  Matt Holliday underrated?  Crazy.

To be completely forthcoming, I used RAR data through September 5th and RBI totals through September 9th.  Considering the depth of research in this post, I didn't think that would be much of an issue.  Thanks to The Hardball Times for supplying the stats Justin uses for his calculations.