Who do you tip to win the US presidential election

[Jon Miller]

Polling techniques may have improved in the past decade, but some points noted from Cornell University:

In a paper published in the International Journal of Forecasting, professors Berg, Nelson and Rietz of the Tippie College of Business, University of Iowa argue that prediction markets are notably more accurate than polls at predicting presidential election outcomes. The authors compared the Iowa Electronic Market predictions to 964 polls over five presidential elections. They had four major findings. First, the prediction market gives a noticeably different picture of the race as a whole, and is often significantly different from polls for long periods of time. During these periods, the market tends to be much closer to the final outcome than polls. Second, the prediction market does not experience “convention bounce,” where one party rises in the polls during its convention and falls after it ends. Third, the market tends to predict outcomes better than polls. Finally, polls are significantly more volatile than the prediction market, and much more subject to day-to-day shifts in voter sentiment.

The third point above is perhaps the most significant. Analyzing this data, the authors found that the prediction market is closer to the actual outcome than are polls 74% of the time.

Yes, the prediction markets include lots of priors which do better than polls without analysis. That seems obvious.

JM

[Jon Miller]

To be clearer:

Polls only would give you 99% or something Biden.

If you just use the uncertainty information from 538 you would predict 95% Biden or something by looking at Arizona, Michigan, Wisconsin and Pennsylvania.

If you start including more priors (economy, etc) then you get the Economist's number or something like it, 82% Biden.

The betters on the betting markets (Which are not the masses and are not the big companies/predictors like Sam and Nate) are using their priors and reading the Economist/538/etc. But they aren't acting algorithmically and their priors are not going to be as successful just as the market became dominated by algorithms and not traders going by their gut.

JM

[Ted Striker]

In an interview several days ago with a radio station in Holland, where he grew up and received his doctoral degree, Kapteyn had predicted that “Clinton is going to win, but I think it’s going to be a lot tighter than people think,” he recalled.

That prediction, he said, highlighted the problem with most efforts at political analysis.


“When you look at pundits and their predictions, the correlation is zero” between what they forecast and how things actually turn out, he said, citing work done by Daniel Kahneman, the Nobel Prize-winning behavioral economist.

“You have to trust the numbers,” he said. “Don’t get distracted by all the things you think about plausibility.”

“What you think personally doesn’t matter,” he added. “I thought Clinton would win. But that shouldn’t change the numbers.”

The tracking poll was not perfect, of course; it projected Trump to win in the popular vote by slightly more than 3 percentage points, but in reality Hillary Clinton seems set to gain a slender popular vote majority, currently about 0.2%. Her margin could expand as late ballots are counted in heavily Democratic California.

That result, however, was well within the poll’s margin of error. The more crucial point was that the poll correctly detected Trump’s appeal to a key bloc of voters: conservative whites who had sat out the 2012 election but intended to vote this year. That group strongly favored Trump, the poll found.

The poll’s ability to pick up those voters, Kapteyn said, stemmed from its approach, which differs notably from the one used by most major surveys.

Instead of asking people to simply choose between the candidates, the Daybreak survey asked respondents to rate, on a scale from 0 to 100, their chance of voting for Trump, Clinton or some other candidate. The poll also asked people to use the same 0-100 scale to rate their likelihood of voting.

That method, which Kapteyn had used four years ago to accurately forecast President Obama’s reelection, “is the most important part” of what the poll demonstrated, he said.

By asking people to give a probability, the poll avoided forcing voters into making a decision before they were truly ready. As a result, it may have more accurately captured the ambiguity many people felt about their choice.

Moreover, by asking participants to rate their chance of voting, the poll could take advantage of information from everyone in its sample group, rather than cast aside those who do not meet a test for being a “likely voter,” as most traditional surveys do.

“One of the ways in which other polls may have gone wrong is that they have a hard time defining who is going to vote,” he said. Polling firms “should look at their likely voter model” and think about whether they are excluding too many potential voters, he said.

[Jon Miller]

Thank you.

Even if it is the best poll, analyses like 538 are going to be better.

JM

[Dauphin]

It seems to me that in the analogy that the better markets would be one part of the market which would not contain the big players, either those that use their skill (Buffet and Sam Wang) or those who use algorithms and are new to the scene ( Bridgewater? and 538 ). Of course this market will be less successful generally than the 'expert' market.

The polls without analysis/algorithms and picking the winner based on those is the 'betting against the efficient market hypothesis'.

JM

I'm not really understanding your meaning here. I don't know who Sam Wang is, so I looked him up on Wikipedia. It says this about his Princeton Election Consortium:

In 2016, PEC predicted both a 93% chance of Clinton victory in one model, and a greater than 99% chance of a Clinton victory in his Bayesian model as seen in Wang's election morning blog post titled "Final Projections: Clinton 323 EV, 51 Democratic Senate seats, GOP House".There was a dispute in the forecasting world about how to interpret the pre-election polls. Wang believed that the polls were reliable and errors were unlikely to be correlated. Friendly rival Nate Silver predicted a much more chaotic election: he pointed to the comparatively large number of undecided voters in 2016 vs. 2012, and believed that errors in state-level polling would likely be correlated (e.g. if one state's true vote favored a candidate by 2 points compared to the polling estimate, it is likely that many other states will also favor the same candidate by around 2 points). Clinton narrowly lost the 2016 election, and Wang said that "In addition to the enormous polling error, I did not correctly estimate the size of the correlated error – by a factor of five." In response to Trump's victory, Wang subsequently ate a cricket on CNN, fulfilling a promise that he would "eat a bug" if Trump won more than 240 electoral votes.

Sam Wang got the 2016 election wrong by more than the betting market.

I don't know how well he has done vis a vis other elections and the betting markets.

[Dauphin]

I would of course note that the accuracy of the gambling or polling methodology is not which one gets closer to a single actual result, it is which one has a higher correlation in the odds quoted and their occurence.

[Aeson]

If you're going to fix an election, you might as well profit from betting on the results. The more the mainstream information is off, the better it works.

[Jon Miller]

I'm not really understanding your meaning here. I don't know who Sam Wang is, so I looked him up on Wikipedia. It says this about his Princeton Election Consortium:

Sam Wang got the 2016 election wrong by more than the betting market.

I don't know how well he has done vis a vis other elections and the betting markets.

He was the well known expert who was very wrong. The >99% guy. My point is that he was strong on priors without using smart algorithms (ala 538). He had been popular, maybe because of his strong priors, but has now obviously lost a lot of his reputation.

JM

[Ted Striker]

Here's some interesting data from a survey from the Fortune 500 CEOs. Terrible ratings for Trump's performance and the party affiliations speak for himself.

His base is beyond vulnerable right now.





https://docs.google.com/viewerng/vie...ssed.pdf&hl=en

[Dauphin]

Corporate leaders rank themselves as having the best performance in response to the crisis. Why ask someone to grade their own homework?

[Ted Striker]

That was pretty funny lol

[Uncle Sparky]

Tying in with the other burning topic, police (or retired police) review police shootings, assaults, etc. nearly 100% of the time in Canada and the US. Some boring, non political types are starting to see this as a problem. The defund/reform the police has legs.

[Ted Striker]

I agree Uncle Sparky. I'm seeing support in all kinds of forums I would have never expected. It's just sortof understood what happened was wrong and the protesters should be supported.

Dude even the USMC just issued a Pride support message.

Always respect for the Marines

[MOBIUS]

Let's ask KH who he thinks is going to win?

[Broken_Erika]

Joe Biden Insult Bot FTW!

https://twitter.com/bideninsultbot?lang=en

[Uncle Sparky]

So after pouring over the available info, I've concluded the name of the person to tip to win the US Presidential election is Yuri Psevdonim. He works out of St. Petersburg (I guess Florida, though he has a really odd area code), and Jared Kushnier has his contact info. The tip will be 10% of the US Department of Education budget, but it is easy to fudge those numbers.

[Dinner]

Given that the most extreme left candidates have won their primaries, granted those were only open to Democrat voters, it seems likely that the left will win the coming presidential election.

[Fake Boris]

People are afraid to voice support for Trump aren’t they?
According to bookmakers he’s far from beaten.