Friday, December 30, 2016

The Election In Retrospect;State Polls Showed Obvious Trump Win

Having reviewed the election night broadcasts from the major networks it is clear the entire punditry on show were utterly befuddled by the result. "How did we not see this coming?" Nobody predicted this."

to state "nobody predicted this" is really just an extension of the blindness inherent in 'we did not see this coming' as a number of pundits, including in all modesty myself, and an academic who has always been right, saw a Trump win.

The stunned reaction included 'the polling industry is broken, they got it wrong." As the night started the consensus was "Trump has a very narrow path, he has to thread the needle whereas Clinton has multiple paths."

 It is this "consensus" based on "data and received wisdom" which, along with what must be willful blindness that was so wrong.

Once Trump had won Florida, North Carolina and Ohio the scales (and the countenances) started dropping from their collective eyes the tune changed to "this is a complete reversal of what everyone thought, it is Trump who now has multiple paths and Hillary whose path is narrowing."

Lets look at the reality in the polling and the actual results in the "battleground states".

The aggregate of polls in Florida gave Trump an 0.2 point lead with the final poll from Trafalgar group on 11/6 giving Trump a +4 lead. The final result was Trump +1.2. Clearly a Trump win was more than probable.

North Carolina; Final aggregation Trump +1.0 actual result Trump +3.7

Ohio; Final aggregation Trump +3.5  actual result +8.1

Iowa; Final aggregation Trump +3.0 actual result Trump +9.5

It is perfectly clear that if the punditry had dwelt on the final poll aggregates at the very least the comment could have been"if the polls are correct in these battleground states the Trump only needs to break through in one or two of Clinton's "rust belt blue wall to have a strong chance." 

Most certainly the polls were massively out in Wisconsin giving Clinton a 6.5 point lead and he ended up winning by 0.7 but in Pennsylvania Clinton's lead was only 1.7 points and the final Trafalgar poll had Trump at +1 and he won it by +0.7. In Michigan Clinton's final aggregate lead was 3.4 with Trafalgar  giving Trump a +1 lead and Trump won the state by 0.3

Again, anyone not blinded by either bias, group think and probably both in most cases could see that Trump was in the margin of error in Pennsylvania and Michigan and if he won, where he led in the final polls, Florida/North Carolina/Ohio and especially if he won the latter by a large margin, which he was obviously going to do as the early returns came in, he had every chance of winning the presidency.

The battleground state polls, in the main and where it counted were not wrong, neither were the national polls which gave Clinton a popular vote win. It was the punditry that was willfully blind that was wrong and they reaped their reward on election night

No comments: