Pollsters try to predict elections for a living while gamblers do it for fun and profit. But who does it best?
The value of election polling is in doubt.
In an analysis published last week, ElectionBettingOdds.com founder Maxim Lott ran the numbers on how often betting markets are right about U.S. elections compared to professional polling organizations.
Lott focused on the 687 betting market predictions about election candidates aggregated on his website.
- He checked how often those predictions were right versus the corresponding polls aggregated by FiveThirtyEight.
- The upshot?: There was no significant difference between the accuracy of the two sites.
Overall, Lott concluded, “Betting odds slightly outperformed 538 in accuracy for the period 2016-2020. However, in 2022, the betting odds did less well, and 538 now holds a narrow lead.”
- Lott called out FiveThirtyEight founder Nate Silver for tweeting in 2020 that “political betting markets are populated by people with a sophomoric knowledge of politics.”
- Yet, at a time when even some pollsters have started to question the value of their profession, Lott offered an olive branch.
- He ended his analysis by citing new research that suggests election polls and betting markets are most accurate when averaged together.