Yves here. Rajiv Sethi makes an important point: that the extreme events of the presidential election have been so far out that they make most of the correlations in the statistical models invalid. Sethi has more faith in the prediction markets. I couldn’t be so optimistic. A little bad doesn’t mean all that good. One can point to the failure of many prediction markets, some as prominent as the prediction that the Brexit referendum would fail by four points when it won by two. Barry Ritholtz has long been skeptical. You can take a gander through his many posts and articles on this topic here.
By Rajiv Sethi, professor of economics at Barnard College. Originally published on his website
This is already the most dramatic election cycle in living memory, and we’re three months away from the finish line. The history of the season could fill volumes. But many of the big events that have shaken the presidential race can be seen in the movement of the market prices of predictions, which compress and reveal a lot of information in a very compact area.
The next two charts, viewed together, tell a lot of the story so far and allow us to think about hypothetical situations in an interesting way. The upper figure shows the prices of the three contracts in the market that won the president in PredictIt, while the one below shows the prices of the contracts referring to the winning political party:
Many important events—the debate, the pressure on Biden to step aside, his resistance to this pressure, the attempt on Donald Trump’s life, and the rallying of the party’s support for Harris, can all be seen in the words above.
In early May, markets gave Biden a slight edge in what was expected to be a close contest. By mid-May Trump had overtaken Biden in the odds of victory, but the Democrats remained ahead of the party contracts – marketers were giving Harris (or someone else) a chance to replace Biden at the top of the ticket.
At the beginning of June the contract prices of the group jumped, but the gap remained small until the debate on June 27.
During the debate the price of the Biden contract dropped significantly, as did the price of the Democratic contract, while the price of the Harris contract increased. These trends continued as pressure mounted on Biden to step aside, and Harris eventually surpassed Biden in the presidential race in early July.
The party’s chances continue to decline during this period of uncertainty, as public calls for a new candidate increase as Biden and his supporters insist that the matter is closed.
The assassination attempt came on July 13, which led to a huge increase in the chances of Trump’s victory. This probability increased a few days later, and then began to decrease in line with the increase in the price of the Harris contract. By the time Biden officially withdrew on July 21, the gap between the parties had narrowed significantly.
As soon as the decision to withdraw was announced, the price of Biden’s contract dropped completely and the price of Harris’ contract went up. The gap between the two teams’ contracts narrowed, and continued to narrow over the next few days.
Yesterday, for the first time since the beginning of June, the probability of a Democratic Alliance victory exceeded that of a Republican victory in this market.
One can use these charts to visualize the unreal situation. By the time of the debate, Biden had already trailed Trump for six weeks. If there were no controversy, this gap would continue, but remain small enough to prevent a major challenge. What the debate has done is to expose weaknesses in time for action to be taken in the face of panic, rather than resignation and paralysis. If people look back at this time, the controversy of the debate may be seen as a blessing in disguise for the party. It opened the only way for a competitive election in November.
This election season has exposed the limitations of statistical models in the face of significant deviations from historical trends. When it became clear that Biden would not run in this election, models based mainly on opinion polls were stopped and had to be reviewed and recalibrated to accommodate the change in the nominee. One of these is now back online, but the others are silent.
Markets, on the other hand, adapt quickly to changing conditions. They tell an uninterrupted story, with all kinds of twists and turns. And they establish an accurate history that can complement other sources of information when these events begin to recede into the past.
Source link