An early look at the 2013 rotation

Assuming health, the Cubs rotation appears to be set with Matt Garza, Jeff Samardzija, Travis Wood, Scott Baker and Scott Feldman. I was curious what we could reasonably expect from this group next year and decided to use the CAIRO projections to calculate percentile forecasts. Before that, there's been some recent talk about how this rotation is better than last year's rotation.

My first thought was that, no, this is not better and it may not even be as good. My second thougt was that it might have more upside though that's not certain. So I decided to look and see which rotation is better.

I used last year's CAIRO projections and compared the rotation that I thought would give the Cubs their best rotation entering the season. I competely dismissed the idea of Jeff Samardzija starting until midway through spring training when it appeared a certainty he would be starting. I was wrong in thinking that was the dumbest idea I'd ever heard of. Anyway, I used the following 2012 rotation: Ryan Dempter, Matt Garza, Carlos Zambrano, Travis Wood and Chris Volstad. (UPDATE: obviously Volstad and Z can't be in the same rotation since they were traded for one another so switch out Z for Samardzija).

That rotation was projected to produce 8.5 WAR. If we remove the worst starter in that group, Volstad, and replace him with what was projected for Samardzija, it would lower it to 8.0. This year's rotation is projected by CAIRO to produce 6.8 WAR.

CAIRO projects the rehabbing Scott Baker to throw only 30 innings so if we're generous and adjust that to 180 innings, the group would then be worth 8.3 WAR, which is equal to last year's rotation entering the season.

Zambrano was later traded so if we swap him out with what we can now reasonably expect from Samardzija, that rotation would have been worth about 10 WAR, so it's still better thant his year's rotation.

The percentile forecasts for this year's rotation are below. The WAR uses FIP.

The baseline WAR is a bit different for this purpose because I calculated it myself.

Because Samardzija had been a reliever prior to last season, he's still projected for some relief appearances. As a starter only, those rates would be just a bit higher (WAR a bit lower), but it's not that important to be precise at this point.

The Cubs top two are relatively good, but afte that there's a considerable drop in expectations.

I wouldn't put a lot of faith into the percentiles for Baker since they're based on only a 30 inning baseline. What I mean by that is that I wouldn't go taking the 2.65 FIP and 1.1 WAR in 36 innings and prorating that to 200 innings. Don't even consider that.

Here's where using FIP for WAR becomes problematic. Feldman's FIP projection is half a run lower than his ERA. A RA-based WAR would probably have him at replacement level at best.