Hi, Yves. Rajiv Sethi makes an important point: the events affecting the presidential election are so unexpected that many of the correlations in statistical models are no longer useful. Sethi has more faith in prediction markets. I am less positive. Not as bad does not necessarily mean that it is as good. There have been many failures of prediction markets, some of which are very glaring, such as predicting that the Brexit referendum would fail by 4 points despite being won by 2 points. Barry Ritholtz has been a skeptic for many years. Take a peek at his many posts and articles on the topic. here.
By Rajiv Sethi, Professor of Economics at Barnard College. His Website
There are still three months to go before the finish line of what has become one of the most surprising election cycles in living memory. The history of this season could fill many books, but many of the key events that shook the presidential election can be read into prediction market price trends, which pack a lot of information into a very compact space.
The following two charts together pretty much tell the tale and allow us to consider counterfactual scenarios in an interesting way: The top chart shows the prices of three contracts in the presidential winner market. PredictMeanwhile, the chart below shows the price of contracts referencing the winning party.
Many key events – the debates, the pressure for Biden to step down and Biden’s resistance to this pressure, the assassination attempt on Donald Trump, the rallying of party support for Harris – are all seen in the above fluctuations.
In early May, markets were betting that Biden had a slight lead in what was expected to be a close race. By mid-May, Trump had surpassed Biden in odds of victory, but the party contract still gave the Democrat a lead. Traders were pricing in some chances that Harris (or someone else) would replace Biden at the top of the shortlist.
The parties’ contract prices crossed in early June, but the gap remained relatively narrow leading up to the June 27 debate.
During the debate, Biden’s contract prices plummeted and those of the Democrats also fell, while Harris’ contract prices rose. As pressure mounted for Biden to step down, the trend continued, eventually leading Harris to overtake Biden in the presidential winner market in early July.
During this time the party’s prospects continued to deteriorate. Uncertain timesBiden and his supporters solved the problem.
The assassination attempt occurred on July 13, causing a spike in Trump’s chances of victory. These chances peaked a few days later and then began to fall as the price of Harris’ contract rose. By the time Biden formally left office on July 21, the gap between the two parties had already narrowed significantly.
As soon as the decision to step down was announced, the price of the Biden contract completely collapsed and the price of the Harris contract skyrocketed. The gap between the contracts of the two parties narrowed and continued to narrow over the following days.
Yesterday, for the first time since early June, the market rated the likelihood of a Democratic victory higher than the likelihood of a Republican victory.
You can also use these charts to consider counterfactual scenarios. At the time of the debates, Biden was already six weeks behind Trump. Without the debates, this gap probably would have remained, but it would have remained small enough to thwart a serious challenge. What the debates did was expose weaknesses early enough to spur action, rather than resignation or paralysis in the face of panic. When people look back on this period, the debate debacle may be seen as a blessing for the party; it opened up its only chance of a competitive election in November.
This election season has exposed the limitations of statistical models. Important departures This is a departure from historical norms. Once it became clear that Biden was not running in this election, the largely poll-based model Paused It had to be revised and re-adjusted to accommodate changes in candidates. One of them is now Back onlineBut others remain silent.
Markets, on the other hand, adapt instantly to changing conditions. They tell an uninterrupted story, absorbing all kinds of twists and turns, and establish an accurate historical record that can supplement other sources of information once these events start to fade into the distant past.
