There's a pretty cool tool available at baseball musings (http://www.baseballmusings.com/cgi-b...upAnalysis.py?) that allows you to figure out the optimal batting order for 9 guys based on obp and slg, and projects the runs per game that the lineup will score. it's based on some really rigorous research that i've read and they cite, so don't dismiss what it says too easily.
here's what i found. i mostly used last years stats, but i guessed at some stats for EdE, used hatteberg's career numbers, chose larue over valentin because he's more likely to repeat what he did, and used paul wilson because he's about the average of pitchers we have at the plate.
The best lineup scored 5.362 r/g (50 runs more than last years offense):
hatteberg, dunn, larue, grif, felo, EdE, kearns, wilson, freel
weird huh? every good lineup had wilson batting 8, because it sets up the good hitters more to have a high OBP guy batting 9. The key to thinking about these new-fangled lineups is to realize that the leadoff hitter only "leads off" the game, and that the 9 guy is important to setting up the 2-4 hitters too.
the best lineup with womack pretty much the same, and scored 5.214 r/g:
hatteberg, dunn, larue, grif, felo, EdE, kearns, wilson, womack
what i imagine will be Jerry Narron's lineup scored 4.923 r/g:
womack, felo, grif, kearns, dunn, EdE, larue, hatteberg, wilson
and the worst lineup scored 4.553 r/g:
wilson, womack, EdE, hatteberg, kearns, larue, felo, dunn, grif
someone should send the link to Narron. The difference between his lineup and the "optimal" lineup would be about 71 runs over the season! Not that he would ever try something like that, it's definitely not "the right way" to make a lineup.
also, by those projections, just switching womack for freel costs the reds 24 runs over the season, about the same amount the reds improved by adding arroyo to the rotation. So Narron can effectively kill that trade by starting womack.