I meant to get this update done yesterday, but it is now looking to me like Trump is in a good position for Tuesday based on deconstructed polls. The following is based on the Clinton 2016 bias reduced to 85% of that bias impact for Biden this cycle. I think there will be a couple more updates between now and Tuesday but the momentum in the polls seems to be on Trump's side over the last week, a good sign. The predictions below are based on some assumptions that are listed below wherein I am trying to be as objective as possible. Personally I think the president could over-perform these results by quite a bit. Conversely the pollsters could be more correct than they were in 2016, but I cannot find evidence that such is the case.
Taking a look at the RealClearPolitics average of polls in the battleground states, president Trump currently has 279 electoral college votes. Joe Biden currently has 245 electoral college votes and there are 14 electoral college votes that have not yet been decided.*
*Based on the following parameters; polls taken in the last 13 days, with a sample size of at least 500 among likely voters, with a margin of error of less than 4.3%, among pollsters rated C/D or better, using the Pollster Tracking by FiveThirtyEight.com, and removing 538's allocated pollster bias as well as my calculated temporal pro-Clinton bias vs. actual election results from 2016. States must have at least 2 valid polls to be considered.
First the good news.
In Texas, Trump leads Biden 51.8% to 48.2%. This represents a difference of 3.6%. In a regular turnout election, this represents a vote difference of 230,728. In a high turnout election, this equates to a vote lead of 251,703. This is measured across 3 valid poll(s). Trump wins the electoral college votes for Texas because while it has tightened, it is not a swing state.
In Florida, Trump leads Biden 51.9% to 48.1%. This represents a difference of 3.8%. In a regular turnout election, this represents a vote difference of 198,592. In a high turnout election, this equates to a vote lead of 216,646. This is measured across 8 valid polls. Trump I think wins the electoral college votes for Florida by enough that there will be no recourse for Democrats to challenge it.
There are a couple of other states that the Trump victory should be similarly too large to contest: In Ohio, Trump leads Biden 53.4% to 46.6%. In Iowa, Trump leads Biden 52.4% to 47.6% (In a regular turnout election, this represents a vote difference of 43,976, too much for a small population state to be thrown into chaos recounts). In Arizona, Trump leads Biden 51% to 49%. (in a regular turnout election, this represents a vote difference of 35,710, probably too much for Democrats to contest).
Similarly in Minnesota, Trump leads Biden 51.5% to 48.5%. enough of a difference to negate any sort of recount or missing ballots showing up in the trunk of a car. But there's a caveat to this one - it is measured across only 1 valid poll, which I am always hesitant to do. I would leave it as a toss-up for now, despite non-polling evidence momentum is in Trump's direction.
Then there are the states that are going to vote Trump but are worrisome due to the tightness of the race.
In Pennsylvania, there's a tight, tight race but Trump leads Biden 50.1% to 49.9%. This represents a difference of just 0.2%. In a regular turnout election, this represents a vote difference of 5,576. In a high turnout election, this equates to a vote lead of 6,083. This is measured across 6 valid polls. Trump should win the electoral college votes for Pennsylvania. But what is concerning is this is a tight enough race for Democrats to try to flip the result to Biden by voter fraud or through some type of court challenge followed by endless recounts a la their Florida 2000 efforts.
In North Carolina, Trump leads Biden 50.1% to 49.9%. This represents a difference of 0.1%. In a regular turnout election, this represents a vote difference of 2,998. In a high turnout election, this equates to a vote lead of 3,270. This is measured across 7 valid polls.
In Georgia, Trump leads Biden 50.1% to 49.9%. This is only a difference of 0.2%. In a regular turnout election, this represents a vote difference of 5,422. In a high turnout election, this equates to a vote lead of 5,914. This is measured across 3 valid polls. That's definitely a recount scenario effort for Democrats.
Now the challenging parts.
There are a few states polls after factoring in bias still show Biden leading: In Michigan, Trump trails Biden 50.4% to 49.6% across 7 valid polls. In Wisconsin, Trump trails Biden 51.2% to 48.8%, measured across 4 valid polls. In Nevada, Trump trails Biden 52.3% to 47.7% across 2 valid polls.
Virginia appears to be a lock for Biden but I see only one poll that I would factor in, and I don't think the Biden lead will be as massive as the polls predict. And in New Hampshire, Trump trails Biden 52.2% to 47.8% but only across one poll that is worth mentioning. Should this be accurate, Trump cannot demand a recount, the lead is just too big.
In Colorado, New Mexico and Maine there are no polls that are worth factoring in but I suspect Biden will win all three.
All of this said, (1) there may be a few new polls posted today that need to be considered and (2) this is with a bias towards the Democrat at 85% of what we saw in 2016 for Hillary Clinton in October. That polling error/bias could be 0% this time around (giving Biden a massive win). It could also be 150% of the previous bias (which means a Trump landslide. For example, if the bias is 100% of 2016, Michigan flips to Trump's column by 50,000 - 60,000 votes.
No comments:
Post a Comment
Disagreement is always welcome. Please remain civil. Vulgar or disrespectful comments towards anyone will be removed.