Former Vice President Joe Biden is favored to win the US election regardless of whether you look at the betting market or poll. But the tough money in the election is hardly as inclined to Biden as the polls do.
Goldman Sachs analysts Blake Taylor and Alec Phillips say that predictive models maintained by data journalists and academics put Biden’s probability of winning up above 80%. But prediction markets show that election results are still highly uncertain, implying just over a 60% chance that Biden will win the Electoral College.
One reason, they say, is just the way the Electoral College operates, it values the larger states over total voting profits. In the so-called tipping point state, Biden led the year instead of nine. In 2016, while polls skewed one or two points nationwide, they missed an average of 4 points in Wisconsin, Michigan, Pennsylvania and Florida.
Markets are expecting President Donald Trump to outperform polls, by about 3 percentage points, analysts at Goldman Sachs said. One reason may be the so-called shy Trump voter, who may not be willing to show support for the Republican candidate. Goldman Sachs ran its own test using the gap between live interviews and automated or online polls, and noticed a point or two away, though scene analysts report that it may be statistical noise. Another problem, Goldman analysts say, is poll errors based on incorrect voter-turn assumptions.
Large differences in voting methods between parties create additional uncertainty, with Democrats more likely to be affected by legal issues or challenges, voting by mail, they say. Conversely, the recent revival of COVID-19 could reduce the number of Republican voters.
Interestingly, the gap was not the same in major Senate battles, where prediction markets followed the polls more closely, Goldman analysts found.
S&P 500 SPX,
has fallen 4% in the past five sessions, although traders have also had to consider new European closures, rising US coronavirus cases, and a deadlock in the fiscal stimulus talks. US key.