Networks

https://www.nytimes.com/interactive/2018/11/06/us/elections/results-dashboard-live.html

After the 2016 elections, where the prediction markets were wrong about who would take control of the presidency. Bettors and pollsters were confident in who would take control of the house and senate during the 2018 midterm elections. On the PredictIt market, the scenario where the Republicans gain less than or equal too 217 seats was $0.65. (Gaining less than or equal too 217 seats would result in the Republican’s losing the house. For the senate race there was much higher chance of the Republicans keeping the Senate. Republican’s needed at Least 50 seats to maintain the majority. The PredictIt market had the Republican’s getting less than or equal too 49 seats at $0.12. So the odds for the Republican’s losing the senate were much lower than losing the house.

Now that we know the results of the 2018 midterm elections we can see that the PredictIt market was correct with their predictions. Republicans lost 32 seats putting them at 198 (10 still undecided), meaning they lost control of the house, whereas in the senate race they gained 1 seat (2 still undecided) putting them at 51 meaning they keep the senate majority.

These prediction markets are often very accurate with what they are predicting. There are some cases like the 2016 presidential election where they get it wrong, but 2018’s midterm elections show how accurate these prediction markets can be.