We know that Donald Trump won the 2016 US elections by the tightest of margins. His majorities were so razor thin that had 77000 votes across Wisconsin, Pennsylvania and Michigan veered the other way, his eventual 304-227 electoral college tally would have translated as a 278-260 victory for Clinton instead.
However, the above statistic was the result of post-election data mining. It is easy for campaign strategists to identify with hindsight which states qualified as swing states. In reality, this was a statistical event that very few saw coming; predicting swing states ahead of an election is a much harder thing to do .
The list of swing states – i.e. those states which cannot be considered by either the Republican or Democratic camps as “banked” – generally undergoes changes with each election cycle. And those changes can be influenced by any range of factors from demographics to the candidates themselves.
However, with the growth of Wisdom-of-the-Crowd trading platforms offering markets for political outcomes – or what some refer to as “political stock markets” – there is an argument that the job of identifying swing states is becoming easier. And according to these markets, we may have some early insight into what will be the key battle-grounds of 2020.
For the upcoming 2020 presidential elections, one political analyst has identified nine potential swing states – Arizona, Texas, Iowa, Wisconsin, Michigan, Florida, North Carolina, Pennsylvania and New Hampshire.
The list was compiled by Taegan Goddard as a result of an analysis of aggregated data coming from FiveThirtyEight – the website of statistician and political observer Nate Silver who was among the rare media voices in the US to give Donald Trump a reasonable chance of winning the presidency in 2016 – and RealClearPolitics, a polling aggregation site.
Two states on Goddard’s list can likely be ignored, given their weight in the electoral college system – New Hampshire (four electoral college votes) and Maine (1 electoral college votes).
For the remaining states on the list, Wisdom-of-the-Crowd website PredicitIt shows high uncertainty – by definition, those with a trading price closest to 50 cents – for the states of Arizona (11 electoral college votes), Florida (29) and Carolina (15).
Florida regularly crops up as a swing state in US elections – it was the scene of the voter suppression controversies of the 2000 presidential elections which saw George Bush win by an eventual margin of less than 600 votes in a state where over 5 million showed up at the ballot box.
However, the PredictIt list of early candidates for swing state status comes with a number of caveats: firstly, trading on these markets is relatively modest – and therefore a high degree of confidence cannot be attributed to them at this stage; secondly, the Democratic Primaries are still ongoing – and, depending on the eventual winner, the list of battleground states may yet still change over the fourteen months that sit between now and November 3, 2020.