I was asked on the Donald Trump thread to provide a data sheet on political betting, which I'm happy to do, since I find the subject very interesting. Here's the first (more general) part, the second (on the actual process of betting) will come in an hour or so:
Political betting vs polls & political experts
Having a look at the political betting market can be interesting even if you don’t plan on risking any money because of the delta that the betting markets often show to the polls and the opinions of political experts. For example, here are two potentially misleading factors that betting markets are more immune to than experts and polls:
Social desirability
When somebody is asked questions by a pollster, maybe on the street or on the phone, there’s a certain probability that there will be an acquaintance in the vicinity. For the average guy, when his girlfriend is near him, it does make sense to answer the question of who should be the next president with „Hillary – it’s time for a woman to be president!“
When that same guy has to bet his own money or cast his own vote though, meaning when it actually counts and nobody else will know, that’s when he might just go for another candidate.
The bubble
How often do we talk about the media’s leftist and politically correct bias? I don't want to offend anyone, but it can be that forums like RVF are just as much of a bubble as an SJW's news feed, just with a different slant.
On betting markets however, the people who actually put something at risk are much more motivated to look at a question like „who will be the next president“ from all kinds of different perspectives and integrate many more different bits of information into their opinion than what is necessary in a comment section.
It is psychological factors like these two why statisticians regularly come to the conclusion that political betting markets are closer to the end results than polls or the opinions of political experts. For example british professor Leighton Williams on the last presidential election:
Snapshot vs Forecast
It’s also important to understand that the questions that pollsters ask invite the interviewed people to provide a snapshot of their current opinion, for example: „If the election were held tomorrow, who would you vote for?“. Even if the interviewer doesn’t frame the question in this way, the interviewed people understand the question in this way. They will answer what they currently think.
But the election will not be held tomorrow – the U.S. presidential election will be held on November 8th, 2016 – 14 months to go. People betting on the betting market will take that into account, people answering a phone call will not.
For example, in spite of Angela Merkel’s very extreme decision to grant unlimited asylum, just last wednesday a poll came out that shows her in the lead with her party getting 41 percent of the (virtual) vote
My guess is that a result like this comes because of the time delay between her decision on the refugee/immigration decision and the results of that decision. Let one year pass, and as the true cost of her decision will become a little more clear to the voters, I'm sure her approval ratings will drop – political bets are highly motivated to take these future developments into account and provide an actual forecast leading up to the point oft he next election – people who are polled will just give their current opinion.
There are many more aspects to it, but I think this should already be enough to at least take a look at the political betting market and why they may see the odds a little differently.
Political betting vs polls & political experts
Having a look at the political betting market can be interesting even if you don’t plan on risking any money because of the delta that the betting markets often show to the polls and the opinions of political experts. For example, here are two potentially misleading factors that betting markets are more immune to than experts and polls:
Social desirability
When somebody is asked questions by a pollster, maybe on the street or on the phone, there’s a certain probability that there will be an acquaintance in the vicinity. For the average guy, when his girlfriend is near him, it does make sense to answer the question of who should be the next president with „Hillary – it’s time for a woman to be president!“
When that same guy has to bet his own money or cast his own vote though, meaning when it actually counts and nobody else will know, that’s when he might just go for another candidate.
The bubble
How often do we talk about the media’s leftist and politically correct bias? I don't want to offend anyone, but it can be that forums like RVF are just as much of a bubble as an SJW's news feed, just with a different slant.
On betting markets however, the people who actually put something at risk are much more motivated to look at a question like „who will be the next president“ from all kinds of different perspectives and integrate many more different bits of information into their opinion than what is necessary in a comment section.
It is psychological factors like these two why statisticians regularly come to the conclusion that political betting markets are closer to the end results than polls or the opinions of political experts. For example british professor Leighton Williams on the last presidential election:
Quote:Quote:
the balance of opinion among so-called political experts was pretty evenly divided between those thinking, like Fox News pundits Dick Morris and Michael Barone, that Romney would win by a landslide, and statistical analysts like Nate Silver, of FiveThirtyEight, Sam Wang of the Princeton Election Consortium and Drew Linzer of Votamatic, who believed the opposite.
Meanwhile, a glance at the betting/prediction markets showed what they had always shown — that the president would be re-elected by a handsome margin, and that nothing that Mr. Romney or his team could do was likely to prevent that happening.
Snapshot vs Forecast
It’s also important to understand that the questions that pollsters ask invite the interviewed people to provide a snapshot of their current opinion, for example: „If the election were held tomorrow, who would you vote for?“. Even if the interviewer doesn’t frame the question in this way, the interviewed people understand the question in this way. They will answer what they currently think.
But the election will not be held tomorrow – the U.S. presidential election will be held on November 8th, 2016 – 14 months to go. People betting on the betting market will take that into account, people answering a phone call will not.
For example, in spite of Angela Merkel’s very extreme decision to grant unlimited asylum, just last wednesday a poll came out that shows her in the lead with her party getting 41 percent of the (virtual) vote
My guess is that a result like this comes because of the time delay between her decision on the refugee/immigration decision and the results of that decision. Let one year pass, and as the true cost of her decision will become a little more clear to the voters, I'm sure her approval ratings will drop – political bets are highly motivated to take these future developments into account and provide an actual forecast leading up to the point oft he next election – people who are polled will just give their current opinion.
There are many more aspects to it, but I think this should already be enough to at least take a look at the political betting market and why they may see the odds a little differently.