David M Rothschild on Posted on

Clinton is 88 percent to be the next president, with Trump at 12 percent. Democrats are now 72 percent to take the Senate and 11 percent to take the House. Below is the joint probability of the different parties controlling the president, senate, and house:

PredictWise20161028x2

1) Markets moved fast today as the FBI announced they were looking at more emails related to Hillary Clinton. The markets were probably right to freak-out, because they thought this could be something really serious: did the Russians hack more accounts and hand evidence of wrongdoing to the FBI? It turns out that the FBI was investigating Anthony Weiner and found some of his wife’s email on his computer. It is probable that they were already investigated and if they were not, they are not even from or to Hillary Clinton. They are things her aide forwarded to her home account to print. Highly unlikely to either be classified or have anything to do with Hillary Clinton.

I have no idea why James Comey sent a deliberately vague letter about Clinton’s emails 11 days before the election. The fact that he was forced to leaked additional information downgrading the impact of the letter within hours, is an indication that he did not think this through very well. If he was trying to help Trump, he probably succeeded in giving him a one or two good news cycles and something to push for the rest of the campaign, but that could have been accomplished by releasing the details about the Weiner investigation at the same time. But, by not doing so, he also greatly diminished his own reputation by needlessly sending the financial and prediction markets on roller coaster.

2) The markets did pretty good at closely up as soon as they realized it was not an actual situation, but something the FBI had created. But, not all markets have been efficient as others. In short, the state-by-state markets (i.e., the contracts on Clinton or Trump to win any state) quickly rebounded to near pre-FBI values (with a few exceptions in the toss-up category), but the topline contracts (i.e., the winner of the election) have not recovered. There are few reasons for this.  (1) State-by-state traders, really into the weeds of the state markets, are probably more knowledgeable on average leading to more efficient results. (2) Topline is hard! State-by-state is directly translating polling and GOTV and possible events into a single outcome. National topline is a product of 2^51 different possible outcomes. If it hard for Nate Silver and I, then why would we assume market do it better? (3) If you wanted to manipulate anything you would crush the topline and ignore the states.

3) Polling will not show the impact of this incident for a few days, if ever. But, I find it hard to determine the mechanisms that it really derails Clinton. Most voters are going to hear Clinton and email and go to their default positions of ignoring it or blood boiling.

1) Strong Clinton: Clinton is 94 percent or more to win the states that add up to 268 Electoral Votes (she needs 270 to win). New Hampshire’s 4 electoral votes are at 92%.

2) “Toss-Up”: There are six states that Trump needs to run, which are all leaning Clinton: Nevada 77%, North Carolina 75%, Florida 69%, Ohio 42%, Iowa 39%, Arizona 45% (if Clinton wins any of these states, she will win). Maine-2 and Nebraska-2 are both worth one electoral vote and both toss-ups.

3) Lean Trump: She has a serious shot in one further state: Georgia 13% (if she wins Georgia, it is a landslide).

PredictWise20161028x1

Clinton continues to look strong in the Huffington Post’s Pollster (Pollster) and RealClearPolitics (RCP) rolling averages for the strong Clinton states. The national polls are continuing to show a solid lead for Clinton. She is up 5.2 percent in RCP and and 7.32percent in Pollster.

PredictWise20161028x3

In the senate I am following seven races very closely. The Democrats are down four seats and will almost certainly win in Wisconsin and Illinois. They are only defending one tight seat in Nevada, holding steady 62%. The remaining six are possible pick-ups for the Democrats.

1) Lean Democratic: Democrats are looking good in Indiana 68%, New Hampshire 51%, and Pennsylvania 62%. If all of these go Democratic (and they hold Nevada) they will have 51 seats. The most likely, as of right now, they win 2 of 3 and we are 50-50 in the senate!

2) “Toss-Up”: Missouri is 43% and North Carolina 38%. If these also go Democratic they would be up to 53 seats. The same polls that showed the tight presidential race also swung well for the Republican incumbent in the senate race.

3) Lean Republican: Florida, Marco Rubio’s seat is at 15% for the Democratic challenger.