All evidence points towards a comfortable win for Biden
The American election will be on November 3, 2020. Information about the results will begin to go public around 10:00am (Bangladesh time), November 4, 2020. There has been a lot of excitement and anxiety about this election. This article sets forth my predictions and the reasons for my conclusions.
How to predict an election
There are three methods: One is an analytical model using historical data and polling data. The polling results are the most important, as this attempts to put information together that is recent, and the underlying probability model allows one to estimate the change of being right.
The second way is for a knowledgeable student of elections in the area to formulate an opinion based usually on extensive discussion with voters and local leaders. There are many people who claim this knowledge or claim to have been visited by a vision that revealed the truth. Generally, this approach is useless, although it draws a large number of participants.
The third method is to establish a voting market and let the public vote on the candidates with their money. The betting market attracts knowledgeable persons. But how well this works is uncertain. Persons participating in the betting markets often use the polling data to analyze prospective outcomes. Buying and selling in financial markets draws a lot of persons who have specialized knowledge; but in a large election, there is limited knowledge outside the polls.
The polls show a consistent story both at the national level and for the states. Biden has a lead at the national level, of 7-9% depending on what set of polls you average. If this difference at the national level holds, then the lead in popular vote will be as follows
Popular vote [million]
Trump (2016) | Clinton (2016) | Total |
63.0 | 65.9 | 128.9 |
Trump (2020) | Biden (2020) | Total |
64.1 | 75.3 | 139.4 |
The total popular vote for the two candidates is projected to increase from 128.9 million to 139.4 million. This increase is partly from the expected voter turnout and larger number of potential voters, but also due to the lack of a significant third-party candidate. This projection says that Trump will get about one million more votes than in 2016, while Biden will receive 9.4 million more than Clinton. The difference between Biden and Trump will rise to almost 10 million votes.
If this is the outcome of the popular vote and the Electoral College and Court mediated disputes award the Presidency to Trump, then most Americans will not accept such a government as legitimate.
However, the outcome of the Electoral College will almost certainly go to Biden. The Economist estimates the Electoral Outcome as 350-188. The Financial Times 273 for Biden, 125 for Trump, and 140 uncertain; as 270 votes are needed to win, this implies Biden will win.
We have done a statistical calculation based on the distribution of votes, determined from the polling for each state, using the uncertainties [sampling errors] to divide the votes for each state between the two candidates. This gives 346 votes to Biden and 194 to Trump. [This calculation is consistent with the winner take all system for awarding the Electoral votes.] Our estimated margin of error is between 5 and 6%, so at the extremes Biden is still the winner.
Real Clear Politics reports the voting odds on October 30 at 63-37 in favour of Biden.
All evidence points towards a comfortable win for Biden. There is a great deal of discussion about the possibility of disruptions in the voting revolving around the counting of the mailed in ballots. In my opinion, this is much less of a problem than was first thought. The voters in the United States are voting early and getting their ballots to the authorities. The magnitude of the problem is much less as a result of all of this early voting.
There will be a lot of claims of wrongdoing and much litigation, but ultimately, everyone will see that these are small effects and interest will be lost. The business, judicial, congressional, and civil service have no interest in widespread conflict over the election of a man like Trump. This is not 1860 when a deep moral issue, slavery, divided the United States. Trump may do all he can to cause trouble, but he will not be able to steal the election from Biden.
In the final days of his presidency, with the Covid-19 pandemic becoming worse and worse in the US, Trump is saying that the problem is solved. He appears to be a fool even to his supporters. In addition, he is dragging people into mass rallies which will lead to many more persons getting sick.
The fact that Trump decided in the last month of the election to invent an imaginary world where the pandemic was fading away, presenting this myth to the American public when everyone knew the pandemic was growing in intensity, reveals a mind that is either living in fantasy or operates at a level of cruelty and disdain for other human beings that is appalling. The people of America have finally grasped what this monster’s real nature is and are going to throw him out of the White House.
Forrest Cookson is an economist who has served as the first president of AmCham and has been a consultant for the Bangladesh Bureau of Statistics.
Leave a Comment