Here are the NEIFI team projections for 2011.  I wanted to get these out before the PECOTA depth charts storm the blogosphere (incidentally, PECOTA this year looks absolutely awful... I can't go into too much detail since it's subscriber-only, but suffice it to say that it appears to be broken in every way imaginable.  The MLEs are wrong, the aging patterns don't make sense, up is down, day is night, Ruben Gotay is projected to outhit Colby Rasmus).

I have the projected rosters broken down by division, with additional tabs showing all the players sorted by WAR.

Projected rosters are intentionally oversimplified, because it's infinitely easier to deal with a 25-man roster than to worry about everyone who might see action.  As a result, you'll see strange things like Johan Santana and Brandon Webb listed in their respective teams' bullpens... I did this because I couldn't give either of them a rotation spot (Santana because of his injury, Webb because NEIFI only projects him for 51 innings), but I didn't want to completely ignore them either.  The projected rosters are generally generous toward the teams, i.e. the best players are projected to get the most playing time... the only exceptions to this are cases where it's clear that the team is determined to give playing time to guys who don't deserve it.  The Diamondbacks' bench is an example of this; I could have dug up guys from their minor league system who are better than Blum, Mora, and Bloomquist, but since those three are making seven figures each, their roster spots aren't in jeopardy.  Regardless, the depth charts themselves aren't that important... any reasonable tweaks you can think of won't move the needle for a given team by more than a win, two wins at the absolute maximum.  I can get almost identical results just by totaling up the projected WAR of everyone in each organization, without worrying about allocating playing time at all.

Eagle-eyed viewers will notice that the WAR numbers for Rockies hitters on the NL West page and on the "Players" page don't match... this isn't an error.  It's because of the Coors hangover effect, which depresses our hitters' value.  Tulo's value as a Rockie is 5.33 WAR... with another team, with no hangover effect to worry about, he'd be at the 5.64 WAR shown on the "Players" sheet.  The latter number is his true ability.

Similarly, the WAR values for relievers on the "Pitchers" sheet reflect their true ability, whereas the values on the division sheets are adjusted for leverage.  This is why Street gets significantly more WAR than Belisle or Betancourt, even though their ERC projections are really close.

I should also add that the projected bullpens include quite a few potential starting pitchers.  If you see a guy listed in the bullpen and you want to know how he'd do as a starter, just add 0.85 to his ERC projection.

Enjoy!