Baseball Analysts has a tool that generates the best and worst lineups for a team, when given 9 players, their OBPs, and their slugging percentages.
I ran this for the Rangers, using BP's PECOTA projections for the likely starters for the 2008 team, and the system cranked out this as the best lineup:
Frank Catalanotto
Milton Bradley
Ian Kinsler
Josh Hamilton
Jarrod Saltalamacchia
Ben Broussard
Hank Blalock
Marlon Byrd
Michael Young
What jumps out at me is TFOTF hitting 9th...but then, PECOTA projects Young to have the lowest OPS of the 9 starters.
I was prompted to do this little experiment by David Pinto's blog post about Trey Hillman considering using Mark Teahen or Ross Gload in the #3 slot, with Pinto noting that some models show "the number three slot isn't as important as it seems to be in the traditional lineup."
Which made me think about the sound and fury surrounding the talk that Josh Hamilton will likely be hitting 2nd and Michael Young 3rd, rather than vice versa.
I do think that people tend to get a little too worried about, say, who hits 2nd vs. 3rd, that sort of thing. I think you want your best hitters (particularly your best OBP guys) at the top of the lineup, and the crappy hitters at the bottom, but at the end of the day, whether someone hits 2nd or 3rd doesn't make a whole ton of difference, I don't think.
I do think, though, that there is a certain psychological aspect that plays into certain slots. The #3 hitter, for example, is supposed to be The Man, the Professional Hitter, the George Brett type who is your alpha offensive player. The #4 hitter is supposed to hit a lot of home runs (which is why I think Hank Blalock, who would be best served not getting pull-happy and homer-conscious, shouldn't be in the cleanup slot).
And I think that's why Washington wants to put Young 3rd and Hamilton 2nd, even though Hamilton seems like a perfect middle of the order hitter right now.