According to the betting markets, the Dems are likely to pick up between 30 and 35 House seats. (Thirty or more is trading at 80 cents, 35 or more at 35 cents.)
The Senate, which was looking terrible an hour ago, now looks possible; GOP control is trading between 50 and 53 cents on Tradesports. But that overall number isn’t really consistent with the race-by-race numbers, or with the Iowa market.
According to Tradesports:
Tennessee Senate and Arizona Senate are gone, which needs we’d need to hold Maryland and take Virginia, Missouri, and Montana.
Maryland Senate looks 90+% likely.
Virginia Senate looks 80% likely.
Missouri Senate looks like about a 45% shot.
Montana Senate is about 70%.
.9 x . 8 x .45 x .7 is about 22%, which is consistent with the Iowa price of 76 cents on a Republican hold. At this point the probabilities in the various races should be just about uncorrelated, unless you figure that the military write-in vote will tred the same way in Missouri as it does in Virginia. But the Tradesports Senate control contract is more actively traded, with a narrower spread, than the Iowa Senate control contact or any of the component races.
So we’re where we’ve been for a month or so: it’s not impossible, but we need to run the damned table.
Update Va. now at 85%, and GOP Senate control is now at 44 cents on Tradesports; the spreads in the Iowa markets are so large as to be meaningless. Hard to believe, but this could actually happen.