Originally Posted by
sportsfan33
It is difficult to postulate what CO is doing. We can only let past history be our judge.
I think the real driver here is that the lack of EB2-ROW demand. Outside of all other constraints, there is another constraint that a fixed number of visas should be given out each quarter. Since the PERM slowdown and the government shutdown has started, there should be a severe lack of EB2-ROW demand, which in short term has kept the EB2-I date to where it is. In fact I will go on a limb and say that the date may move forward even in December (end of quarter) and in the best case, match EB2-C (Nov 2008).
In this scenario, it doesn't matter even if EB2-I exceeds its annual demand. Sure, some spillover is used here, and sure, people in later months of 2009 will lose out. But in the grand scheme of things, it is immaterial. It is October already and we will start the holiday season soon. With the shutdown prospects still looming, I think the backlog will just increase in near term, which will be a boon to EB2-I. I saw that Q has posted a very aggressive FY-2014 estimate on his site. I think along the similar lines. EB2-I will cross mid 2009 this year for sure. The question is how the forward movement will be accomplished (early in the year or later).
P.S. I don't think CO runs any deep optimization algorithm to derive his strategy. His algorithm should be very simple meant to satisfy a bunch of constraints. When he considers "demand", he must literally take a snapshot of approvable cases ready on day X at time Y and that's it. Finally, there must really be a strong disconnect between the USCIS and the DOS and that disconnect is what propelled the dates forward in 2012. CO simply applied QSP because for whatever reason, the demand from other categories was invisible to him and in order to satisfy the quarterly limits, he ended up pushing the EB2I date. Unfortunately, it is difficult to reverse engineer this because whatever internal metrics the DOS uses are unknown to us. That's my view in a nutshell.