Sam Wang, who writes the Princeton Election Consortium’s blog entries, has an entry entitled “Congressional predictions I: Senate 59-41, House 258-177” that is well worth reading. He routinely does a fine job of aggregating data from polls and examining them scientifically.
The voting makeup of next year’s Congress is a simple sum of votes, and therefore poses an easier estimation problem than the Presidential race. Many other sites give just the average estimate. Here I will give a probabilistcally-based analysis that (a) estimates the likelihood of Democrats reaching a supermajority in the Senate, and (b) puts confidence intervals on the estimates. This projection will be updated as the election nears.