NL East | NL Central | NL West | AL East | AL Central | AL West

Update 2: Well, you live and learn. I couldn't help myself, and after some thought and email discussion with Sky Kalkman of Beyond the Box Score, I decided to use Pythagenpat, a more precise win% estimator. I also tweaked a couple of other things with respect to how FIP is scaled and the run environment, if you want the details, just ask. The main effect of Pythagenpat, so far, is that gives a bigger "spread" between pitchers; as opposed to traditional Pythagorean win%, in this run environment, Pythagenpat adjusts things such that the further above/below average a pitcher is, the more/less valuable he is relative to the win% generated with traditional Pythagorean win%.

Update: I've added row in each teams table to include the sum of the the "top 5" projected SP WARs, since some might want to see what the "real" rotation is worth. See my newly added note in the "details" below.

After I wrote my post about the 2009 A's rotation, I realized that I had spent a lot of time setting up a spreadsheet, collating the data, and improving my calculations, and that it would be shame to let it go to waste. So, struck with inspiration (or was it desparation?), I decided to refine the methods that I'd learned from various places around the web, take the the three best publicly-available projection systems, Baseball Prospectus's playing time estimates for each team, and go through each MLB division and rank the projected rotations (including spot starters, since those definitely matter!) by projected Wins Above Replacement (WAR).

Some details:

• The projected 2009  FIP for each pitcher is generated by averaging the projections of the three best publicly available projection systems: CHONE, ZiPS, and PECOTA. The projected FIP for CHONE and ZiPS is that listed on the player pages at FanGraphs. I generated the PECOTA FIP myself from the raw stat predictions given in the spreadsheet. Playing time/depth chart projections are based on Baseball Prospectus's.
• The primary "total" and rankings are based on the value of all pitchers projected to start at some point during the season, since no team only starts 5 players all year long. However, if you want to compare the "real" rotation, I've included a "top 5" figure for each team, based on the top 5 projected starter WARs for each team (not the top five projected IPs).
• Value is based on a Win% generated using PythagenPat in relation to a average FIP-ERA of 4.40 and an RA of 4.76.
  
• Run values are generated by are scaling FIP to RA rather than ERA for better accuracy and to properly value pitchers with respect to hitters.
• I used an appropriate replacement level win percentage for starting pitchers while also adjusting for relative talent level of the leagues: .390 is set as replacement level for starters in the NL, and .370 for the AL.
• Adjustments for ballparks use the five-year regressed park factors used at FanGraphs.
• Adjustments are made for each pitcher's influence on his run environment through a dynamic run-to-win conversion for each individual pitcher.
•  Note that neither the rates nor playing time projections are "mine." I have simply taken the performance projections and averaged them, and BP's playing time projections are carefully done. I haven't included the complete projected "lines" for each pitcher -- you can find them at the home sites for each projection system (and FanGraphs, for CHONE and ZiPS). Also, don't blame me when these turn out to be way off...
• Obviously, I'm not smart enough to have figured this all out myself. I've picked up stuff from all over and from various people, and while I use methods very similar to those employed at FanGraphs (much of which comes from Tom Tango), there much, much of information out there for free. For a good list on introductory links, see this post at Royals Review. In partcular, check out the FanGraphs series on pitcher value, which covers much of the material I'll be using here, if you want the details... if not, just read on.

The series begins with the NL East. Hey, did you know that Beyond the Box Score is having an NL East Chat at 2 pm EST today? What a coincidence!

The individual player/team numbers, rankings, and brief commentaries are after the jump.

Comment: A picture graph is worth a thousand words:

Hey, at least they've got the next Sandy Koufax on board...

Comment: Say what you want about the collective legacy of Schuerholz's former lackeys, but Frank Wren has put together a heck of a rotation. Of course, you really need to buy into FIP to a certain extent to buy into the greatness of Javier Vazquez (and if you haven't already, check out Josh Kalk's Pitch f/x analysis of Vazquez at THT), but also remember that neither the park nor the outfield defense in Chicago did him any favors. Kenshin Kawakami isn't projected by any of the systems, but the way, so I did what had to be done and projected his FIP as exactly average. That adds a greater degree of uncertainty to the Atlanta projection.

Comment: We have to do without Jamie Moyer's PECOTA projection because they "deadly accurate" system only projects easy players, I guess (just kidding Nate, you're still awesome... .hey, call me sometime when you get the chance...). Decent top half of a rotation... I'm guessing that Chan Ho Park's numbers reflect his time as a reliever.

Comment: Do you ever wonder if Marlins' GM Larry Beinfest would really be awesome if he didn't work for Jeffrey "Welfare Queen" Loria, or if he'd be spoiled with, you know, a budget appropriate for a major-league baseball team? Then again, this group is about as good as Omar Minaya's collection, and we haven't even mentioned payroll. Hey, the Tigers should could use a guy like Andrew Miller, I wonder what it would take to get him?

Comment: What's more sad, the fact that the Nationals project worse that the Marlins, or the fact that Bowden/Karstens were actually trying to be good (as opposed to Loria)? Can you think of a more appropriate place for periennial "breakout" candidate Daniel Cabrera to end up than a franchise enduring the last gasps of the Jim Bowden regime?

The Braves look to have a pretty clear edge in starting pitching in the division, although the Phillies are pretty close (and who knows how good/bad Kawakami  will be? I don't). Much of the Braves advantage comes from their two big offseason acquisitions: Derek Lowe (free agent signing) and the trade for Javier Vazquez. Maybe you don't buy Vazquez as a better pitcher than Johan Santana, and I'm not sure I do, but that's how the systems project his FIP. The Mets, on the other hand, are struggling to stay ahead of Florida, and one injury to Santana... As for the Nationals, hey, at least they get two picks in the upcoming draft. And Aaron Crow will be available! (okay, cheap shot).

We'll continue the series on Monday or Tuesday of next week. Unless Kyle fires me for skimming projections off of Dominican prospects...