PredictWise Blog

Election Update - 10/8, 27 Days

Bookmark and Share

The senatorial forecasts have remained remarkably steady over the last week with two small exceptions. First, Kansas has come into stronger focus and the Independent Orman is pulling away from the incumbent Roberts. Second, Michigan, which was leaning heavily Democratic is now a solid Democratic (and off of our chart!).

I have also updated the forecast for balance of power. For those of you who watch the site closely, you may have seen a small, but reasonable discrete jump when I went live with my updates about 2 AM ET on 10/8/2014. The reason is that I was previously assuming that Orman was 100% likely to caucus with the Democrats. Now, the calculations assume that Orman is 100% to caucus with the Democrats if they get 50 or more seats, 100% to caucus with the Republicans if they get 51 or more seats, and 50% to the caucus with either party of the final tally (excluding him) is 50 Republicans and 49 Democrats. Thus, what I have done is derived the balance of power as if the Kansas race did not exist and then add in the possible effect of the Kansas election. I will update this choice as Orman’s choice comes into focus. Further, the prediction market forecast that is directly forecasting balance of power is a few percentage points different from the aggregated balance of power prediction generated from the state-by-state elections. In order to ensure consistency, the topline balance of power numbers now reflect the aggregated forecast of the states, but all data is noted.

The forecasts from the major news organizations are converging, which is not surprising. More unique fundamental forecasts that dominate early stage forecasting are now completely supplanted by heavy polling, which everyone sees.

Here is New York Times and FiveThirtyEight compared with PredictWise. Not too much difference:

Updating Predictions: senatorial, senatorial balance of power, and gubernatorial.

Election Update - 10/1, 34 Days

Bookmark and Share

The biggest update of the senatorial elections over the last few days is the good news for Republicans in both Colorado and Iowa. Both of these have moved to leaning Democratic to toss-ups. Again, nothing extremely newsworthy in either of the races, but a gradual movement in the polls.

Imagine for a second that there were 99 senate seats (eliminating Kansas’ seat that is up this year) then the Republicans are 52% likely to get 51 seats (and have the majority) and the Democrats are 28% to get 50 seats (and have the majority). Thus, the Republicans are 62% to win the senate outright and the Democrats are 28% to win the senate outright. The Kansas race is basically 50% and relatively uncorrelated with other races, at this point. If Orman chooses to caucuses with the Democrats if/only if he is pivotal to Democratic control, then he will caucus with the Democrats about 50% of the time, but 100% of the time it matters. So, the Democrats have about 38% likelihood of retaining in that scenario. If he goes Republican at 49 Democrats and 51 Republicans than he will go Republican at 100% of the time it matter. So, the Democrats would have 28% likelihood of retaining the senate. More to follow on this soon …

Here is New York Times and FiveThirtyEight compared with PredictWise. Not too much difference:

Updating Predictions: senatorial, senatorial balance of power, and gubernatorial.

Election Update - 9/25, 40 Days

Bookmark and Share

Over the last eight days there has been substantial movement in four of the key senatorial elections. The first is Kansas, where I feel my prediction is a little low. The reason is that the fundamental model, which still has some power, is not well identified on independent elections! But, you have to run with what the data says. Second, Colorado, Arkansas, and Louisiana have all trended a little bit towards the Republicans over the last few days. None of these are serious issues, as much as the release of new polling.

It is important to note two weird structural things about this election. First, the independent in Kansas has declared for the Democrats if they get 50 seats or more and the Republicans if they have 51 seats or more. There is a 20% likelihood of the senate being 49 Democrats and 50 Republicans, with Kansas not sure. Second, the Louisiana race goes to a December 6 runoff if neither candidate gets 50%. This is extremely likely, making the standard three-way polling very confusing, because the third candidate is a Republican. This gives the Republican an increased edge in any runoff election.

Here is New York Times and FiveThirtyEight compared with PredictWise. Not too much difference:Updating Predictions: senatorial, senatorial balance of power, and gubernatorial.

Possible error in Scottish predictions really large and likely favors NO

Bookmark and Share

Using betting market data from Betfair, I have the NO vote in the Scottish independence election at 84% to succeed. And, I can break that down:

6% for 0% to 40% of the vote for YES

26% for 40% to 45% of the vote for YES

51% for 45% to 50% of the vote for YES

12% for 50% to 55% of the vote for YES

3% for 55% to 60% of the vote for YES

2% for 60% to 100% of the vote for YES

There are two key points. First, this is a really wide range. The polls are all clustered at NO being a few points ahead, but the markets are giving 32% that NO wins by 10 percentage points or more! Second, if the polls are way off, the swing is going to favor the NO vote. Intelligent review of the polls are both surmising and determining this possibility.

To provide some perspective on this crazy error possibility, let us consider what an 84% election the in US looks like right now. Gary Peters is 83% to defeat Terri Lynn Land for the open Michigan senate election. My current fundamental and poll-based forecast has Peters getting about 54% of the vote. I would put the possibility at him getting 60% approaching 0%!

Election Update - 9/17, 48 Days

Bookmark and Share

I last addressed the senate 6 days ago and there has been meaningful movement in two states. First, North Carolina has gone from leaning Democratic to strong Democratic. This is on the strength of some new polls. Second, Kansas has gone from very difficult to possible for the Democrats. That is assuming that Orman is going to caucus with the Democratic Party. This on the strength of the first polls showing the Democratic candidate basically gone and looking increasingly likely the Democratic candidate will not be on the ballot, despite the Lt. 'Governor's best efforts. It amazing to think that Kansas’ top three races: Senator, Governor, and Lt. Governor may all go Democratic.

Here is New York Times and FiveThirtyEight compared with PredictWise. Not too much difference:

Updating Predictions: senatorial, senatorial balance of power, and gubernatorial.