Early Voting & The Forgotten Lessons From 2016 Polls

Early Voting & The Forgotten Lessons From 2016 Polls

via ZeroHedge

Voters across the country have been crowding into polling places and mailing in ballots in numbers rarely seen in an off-year election.

In some states, as The Courier Tribune notes, more people are on track to cast ballots in early voting than in the entire election in 2014.

Democrats hope that surge indicates they may be succeeding in mobilizing a crucial demographic (because Democrats rely heavily on the votes of younger people and minorities, who are less consistent in their voting than are older whites, their candidates usually benefit from a higher turnout).

But Republicans are also energized, turning out in larger numbers than Democrats so far in Florida, for example, where a cliffhanger race for governor features a Donald Trump acolyte competing against a progressive who would be the state’s first African-American governor.

Nationwide, as early voting was coming to a close in many states Friday, more than 30 million ballots had already been cast. Turnout is hitting a pace closer to what’s typically seen in presidential elections. It has the potential to be the highest in an off-year election since 1966.

“When you look at some of these states, the numbers are eye-popping,” said Michael McDonald, a political science professor at the University of Florida and one of the country’s leading experts on voting patterns.

Overall, the most either side can tell is what has consistently been true of Florida since 2000: The statewide contests are likely to be very close.

Of course, none of this early-voting trend actually means anything statistically relevant – despite the best efforts of MSNBC et al. to spin it to the left’s advantage.

As Salil Mehta points out in his Statistical Ideas blog, everyone recognizes that the 2016 election polls were chaotic.

You can’t put a positive spin on fake 90% “probability”, with a straight face.  But to save face, the forecasters behind those polls have certainly tried!  Meanwhile in the months leading up to the 2016 election, we had correctly reasoned that Hillary Clinton had closer to only a 50% probability.  The gulf between what the mainstream news pushed out, and our reality, was indeed that wide.

Interestingly, all pollsters from back then are still in force.  Same work this go around.  Not much different.  Polls currently indicate that the Democrats have an 80%-85% “probability” of overtaking Congress.  To some, this seems nearly as (too) high as the fake “probability” they were being told in 2016.  Is it still a sure thing, this time?  We will briefly discuss below that we instead see the Democrats with a 55%-60% probability.  Indeed a gulf still exists, but obviously it is not as exuberant as was the case in 2016.

Now for ground rules, let’s reiterate that our website has never been politically biased.  We care about people with differing views, and our sole intention if to focus on the probability theory concepts.  With that, let’s discuss three major math themes that are worth reinforcing this go around, for the 2018 midterm forecasts.

The first theme is that the major pollsters are using a highly limited small sample size of longitudinal polls in order to formulate estimates of their errors.  They also assume a nice clean, normal distribution even though their limited data is noisy.  The result of these issues is that pollsters (such as competing statistician Nate Silver) continuously give absurdly high “probabilities” for outcomes that then fail to materialize. Here is a reminder of some of his high-profile failed forecasts, each with very high “probabilities” stated for occurring:

• 2015, 75% probability on United Kingdom election
• 2016, Donald Trump at 98% to lose GOP primaries
• 2016, Donald Trump at >90% in Alaska primary
• 2016, Hillary Clinton at >90% in Michigan primary
• 2016, Hillary Clinton at >90% in Indiana primary
• 2016, Hillary Clinton at 90% in Wisconsin primary
• 2016, Hillary Clinton at >70% in general election
• 2017, 75% probability on Alabama senate election

Don’t worry; for every high “probability” call that you can pick up that he has gotten right. We can provide at least one high “probability” call that he instead got wrong.  So much for high probabilities!  And this brings us to our second theme, which is that pollsters are still stating too low a self-assessed “margin of error”.  See this chart below…

Nassim Taleb in 2016 pointed this out as well:

What is alarming isn’t that they were wrong but that these idiots underestimated their error rate.
Similar to p-valhttps://papers.ssrn.com/sol3/papers.cfm?abstract_id=2834266 …

Statistical Ideas @salilstatistics

Each claimed ~5% error & all bad. Those in 90s% EVER (all were) were 100% wrong in demographic & probability calc. http://statisticalideas.blogspot.com/2016/11/abstruse-pollsters.html …

7:23 AM – Nov 10, 2016