Republican hold in congress?

The Economist indicated today that Democrats would need to have a lead on the generic congressional ballot of about 7% to take control of Congress.  That is because of factors like clustering and such, which are explained in the video link above.  Anything short of 7% and the Republicans probably maintain control of the House. Let's take that analysis at face value. What the upcoming midterm elections hold is not clear. However, what is certain to me, is that there is not going to be a big Blue Wave.

I looked at the polls taken in 2018, month by month (based on the final polling date of a given poll), of only polls where likely voters are questioned.  That's because polls of all adults or registered voters tend to be misleading. Also rather than applying a flat average, I use a weighted average as some polls are three times the size of others and therefore carry a different level of statistical significance, margin of error etc. 

Here's what I see:

In October, the lead for Democrats is only 6 points.  That's close to the 7% they'll need to win the House.  And what appears better for Democrats is that this is an increasing percentage over the last 3 months.  But as a conservative I'm not worried, because this analysis does not even look at voter crosstabs and weightings between Democrats/Independents/Republicans.  And the methodology of determining likely voters can vary from poll to poll too.

In aggregate, the RealClearPolitics averages are not unimportant however, because they can be directionally indicative.   

I dove a little further.  It looks like Democrat-favorable polls peaked in the immediately prior buildup and immediate wake of the Brett Kavanaugh confirmation on October 6th.

Furthermore there's an NBC/Marist poll putting Democrats at +9.  I discounted that source in 2016 in the presidential elections for good reason, and I'm inclined to do the same her.  On the Republican favoring side, Rasmussen has Democrats only at +3. If you exclude those 3 (and I'm almost ready to hop back on the Rasmussen bandwagon, but for now, I'll hold off), that leaves just one poll in late October mattering. Reuters/Ipsos has Democrats +7.

Right now it looks close, but there's still a week left and there are a few polls left to come in that will probably be much more relevant.  But if I were a betting man, right now I'd predict a stalemate, with Democrats making some gains but not quite enough to take the House.

That's not my final prediction just yet, but at this point there sure does not look like a 35 to 40 seat gain Blue Wave in the House to me.