The US presidential market is that is a market I’ve followed since 2004. Because I’ve got data going back since 2004, I’ve always taken an interest in it. But I’ve only been actively trading it in recent years.
2008 was the first market but my activity has grown as has the market! In 2016 the matched bet turnover was just short of an astonishing £200 million. That was a record individual market on the Betfair betting exchange.
That was until this year’s market. The 2020 market will comprehensively smashed that amount having exceeded the entire amount traded in 2016, even before election night.
Profiting from early market activity
The interesting thing about these markets is that initially there’s no model that you can particularly use to price it. People will try to form an opinion around how each candidate is polling, but you can’t specifically say, it’s going to be ‘this’ price as polls are typically unreliable.
While this quote was on market research, the same can applying to Political betting and research when polling. ‘people don’t think how they feel, they don’t say what they think, and they don’t do what they say’. Other than that, I believe all stuff read on polls.
Trading a political market early on in a politics market, it’s really just a case of trading news flow. As one piece of news comes out on one side, prices will go up and down in response. In 2016, when Trump was making all of these errors and interesting remarks about his relationship with women, there was plenty of opportunity.
Curiously was not to lay him because the price would react dramatically at that point and go flying out. I was actually backing him at these points, catching the best price. Because it was almost inevitable that some bad news about Clinton would follow. The price on Trump would come back in, at which point I’d reverse my position and just wait for the next news story.
So if you look at the chart from that election, the chart was going flying around all of the time and it provided the opportunity to trade in and out consistently.
Media role in forming opinion
Modern media is such that they are always looking for ‘eyeballs’, so even the most modest of stories can get sensationalised.
It’s tough to find any truly objective view on an election and politics in general. All types of media exhibit some bias or another and some outlets are just plain biased. I’ve seen many Politcal pundits accidentally throw themselves under a bus with heavily skew appraisal of the markets and odds.
Just like when you are betting on football, you probably shouldn’t bet on your team. But in Politics, it takes a lot of effort to detached yourself from the arguments. But doing so, is very profitable.
As a story breaks and proliferates around social media the market will respond then eventually settle again as the story dissipates. You just have to jump in and out on on good and bad news.
Political betting on Election night
When we come to election night, you will be able to see some of the dialogue that going on in the early hours of the morning. You can also go back and look at previous elections for clues as to how the market will trade.
The interesting thing about the US election is that it’s held on a continent over four times zones. I went back and looked at how the votes came in state by state and how they were portioned to try and look for patterns. When the votes came in, as you go from east to West Coast, then there would be a certain pattern of activity that would tend to develop.
You tend to find in recent years that on the East Coast, there could be a number of different types of results coming in there. But as you move to the Midwest, it’s predominantly Republican. Then as we move to the West Coast, it’s predominantly Democrat.
You begin to get a feeling of where the activity was going to go and how things were going to pan out, just by looking at that and comparing against previous elections. Make sure you read up on the electoral college before making any assumptions. A popular vote does not a president make.
I’ve found this really useful spreadsheet for calculating the outcome of the overall market using data and assumptions from the electoral college, this should help you signficantly.
The best way to rationalise these markets is to imagine a football match and a team being up 1-0 with two states to go. Hopefully, that will give you a picture on how you would trade the night when results are declared.
In 2016 when Trump started to get some of those key states on the East Coast, you could tell that, in fact, there probably weren’t going to be enough left by the time it went through the Midwest and out to the West Coast for Clinton to win. That meant that you had had the opportunity to trade Trump on the basis he would almost certainly shorten or vice versa.
Watch the video on this blog post for more information as to how the night played out and how that affected the odds.
The big difference in 2020
There are a number of differences in 2020.
Incumbents have to defend a track record to get re-elected, which is slightly harder than attacking a sitting candidate. But covid-19 has thrown a spanner in the works meaning that many will be casting postal votes in advance of normal polling methods. It’s quite likely this will be somewhat contentious, as people who may not traditionally vote will vote. Their demographics are somewhat different which may influence the final outcome.
But also you shouldn’t forget the Floridian farce of the 2000 election. You can guess that if there are any close votes anywhere, things get very contentious. Ultimately it came down to 537 votes of six million cast in Florida and individual ballots were being scrutinised for errors and checked for validity. In the event of a close vote this year, things could get very messy.
Ultimately, it make take longer for the final vote to be counted and rubberstamped. So a twist in the tail of the election may be the big story this year. This could make it a monster market to trade.
Overall, I’ve noticed an inherent bias when it comes to voters declaring their support for right learning politicians. Social responsibility is a given in modern society I think (hope?), but voters also feel that people should take personal responsibility for their decisions. So that tends to skew polling to the left against the actual outcome.
With Biden polling well and at short odds, I’ll spend election night scanning to see if that is a valid view or not. Never underestimate how much personal freedom and liberty are essential to the American way.