We didn't weight a single poll for party id during the general election, and I think that was the key decision that made us one of the most accurate pollsters in the country- party id is a moving target and where it is in April might not be where it is in October, hence the choice to let it fall where it may.
It's been a different story since the election though. When we conducted polls in Colorado and North Carolina last month the unweighted results came back way too Democratic, with a spread that was 6-7 points higher for the Democrats than what party id showed on the exit polls in November. So we weighted both for party to bring them a little more in line with what our pre election polls were finding.
I'm writing about now this now because the same thing appears to be happening on our Missouri poll. Either there's been a major shift in party id toward the Democrats since the election or Republicans are too depressed to answer polls right now. While it is probably a mixture of both, I think it's likely more of the latter.