Wednesday Rockpile: More of the Francis fret

One of the several reasons that pitcher wins aren't a great stat to use is because they can be so unreliable from season to season, even over multiple seasons at times, because it's impossible to predict where a team's runs will be distributed. Two pitchers on the same team can be equally effective but have vastly different win totals just because of an imbalance in run distribution. Yovani Gallardo had 13 wins for the Brewers in 2009 with a 3.73 ERA, Manny Parra had almost as many, 11, but had a 6.36 ERA.

So at any rate, I've got an issue with a lot of fans interpretation that the Jason Marquis to Jeff Francis exchange can only be successful in 2010 if Francis equals Marquis' win (15) total from 2009. Troy Renck alludes to this too in today's article about the left handed starter's pending return at the Denver Post. There can be no doubt that Francis' healthy return will be an important key to the Rockies success in 2010, but I don't thnk we've really taken a close look at just how much we will need from him. Maybe today's a good day to do that.

Let's start by establishing a wide range:

1. We know that a season like Marquis' 2009 should be enough to keep the Rockies competitive.

2. We know that a season like Francis' 2008 likely won't be enough. There's an issue with making this the bottom of the range, however, in that Jeff isn't going to be given the same leash in 2010 if he's struggling as much as he was in 2008. He's too much of a question mark right now and the team competitive enough that the Rockies aren't going to sink  another 24 starts his direction if he's performing at that level. Thankfully all reports are that he's in a much better place, but my point is that Francis 2008 is somewhat lower than what the actual bottom in 2010 will be, even if he himself is just as bad a pitcher.

More after the fold:

Second, let's get a general feel of what we'll need from the starting rotation on the whole. Here, when it's cumulative for the entire rotation in a full season, I think pitcher wins can have some value as a basic measure, as you lose the problems of the imbalanced run support distribution that you run into looking at them individually. So let's assume that a goal for the Rockies in 2010 should be 95 wins. They probably aren't at that talent level yet, but it's a number within their range and should be enough to win the NL West and challenge for home field advantage in the NL playoffs.

Assuming that 70% of those wins go to the rotation (league average last year was 69%), it means our starters should be targeting 66 or 67 wins, or just over 13 each. Last year, the Rockies rotation took on a larger share than the NL average and had 69 (among the playoff teams, the Rockies and Cardinals each had 69, Philly had 66 and Los Angeles had 58) so if you're taking a pessimistic approach and feel that the bullpen can't pick up that load and want to push the rotation's share up closer to 70 W's, that averages to 14 each and I would say 14 would probably be about as high as the Rockies are probably expecting from Jeff after missing a year.

Next question, though, would be the bottom of the acceptable range. Let me just throw this theory out there for everybody in the Francis must equal 2009 Marquis camp: I think you're looking at it wrong. Try this instead: 2010 Francis must equal 2009 Aaron Cook, 2010 Cook must equal 2009 Marquis. This makes sense to me and seems very doable. Basically, Cook has to rebound to a 14 or 15 win pitcher we all know he can be, and then the expectations for Francis can go down as low as the 11 win, 158 inning season Cook brought to the Rockies last year.

You saber-haters out there, keep that 11-14 win range in mind, and you might want to skip over these next couple of paragraphs, because I'm going to try and address all the people that just rolled their eyes over what I just did.

The Cook/Marquis replacement still seems to stand from a sabermetric standpoint, we just know better not to expect a specific amount of wins. The Rockies in 2009 competed with three pitchers posting sub 4.00 FIP's and two clocking into the 4.00 to 4.50 range. Last season, Cook had an FIP of 4.60 and an xFIP of 4.20, while Marquis had an FIP of 4.10 and an xFIP of 4.41. Projections for 2010 have Cook, Jason Hammel and Jorge De La Rosa regressing into a 4.00 to 4.32 FIP range, with Jimenez coming in a bit better than that.

The fact that this projects as a lower overall starting performance in 2009 isn't much of a concern since everybody who had decent starting pitching in 2009 projects to a lower overall performance in 2010, basically the team is still 90 wins good, Jeff Francis just has to be in that 4.10 to 4.50 FIP range. His career FIP is 4.54, but his 2004/2005 rookie seasons and the injury handicapped 2008 (which make up about half of his overall line) probably make that figure deceptively high, and will likely skew projections as well. If he is at full strength and has his usual location and movement with his pitches, posting a 4.35 FIP shouldn't be much of an issue. 

What about taking it to the next level, though? The easiest path to a 95 win season for the Rockies would be adding five wins from luck to the 90 win talent that the team has. Okay, that's hardly predictable, I'm sort of kidding. Here's this, though: the easiest path by using actual player skills would probably be to have some combination of JDLR and/or Hammel repeating their 2009 performances and Francis and/or Cook coming close to their 2007 level of play. Basically getting two of these four pitchers to put in near peak seasons along with the projected overall improvements from our hitter/fielders would get us there.

How realistic is that? Given that three of the four starters are between 27-29 years old in 2010, Cook is only 31 and Jimenez, 26, it's easier to envision peak or near peak performances from this rotation than it would be for teams like say the Braves or Dodgers, who have starters whose ages mostly outside a player's projected peak range.  We don't know who or if anybody is going to peak for the Rockies in 2010, but the age range of the rotation suggests that somebody will, and probably more than one starter at that. Wasn't 2009 our peak season for starters though? Not necessarily, that Hammel and De La Rosa both had peak type seasons in 2009 doesn't preclude them from having them again in 2010. In fact, it actually helps the likelihood of a repeat occurrence.

This last point is a large reason why I'm not at all concerned about the state of the Rockies rotation heading into 2010, even without a Francis hedge in place. Actually, I'll just list the main reasons since I'm already so late.

1. Four starters in their primes, one more in an area where he should still be near his peak value.
2. Positive health reports on Francis from everybody who's in a position to know anything. The closest we have to a negative is Jim Tracy's admission that "there will be some ups and downs" early in the season, but that has nothing to do with health as it implies there's going to be some wobbliness getting back on the MLB bike. That the Rockies front office has downgraded their insurance targets to the Tim Redding level is a huge sign of their optimism in Francis that we shouldn't overlook.
3.  The base level of our worst starter will probably be around a 4.50 FIP. This is a significant goal and usually a good delineation for a contending worthy team (if you have an offense like the Phillies, you can be an exception). I realize Cook's was slightly higher in 2010, but the peripherals suggested that it should have been better.
4. Franklin Morales and Greg Smith are probably better options for that hedge than we give them credit for.

At any rate, again my apologies for being so late with this. I've obviously come to that point in the offseason where I feel I've got a lot to say, but missing morning post deadlines isn't the best way to build a blog. I'm going to scour the Internet right now and start writing tomorrow's Rockpile tonight to try and avoid that happening again.