Author Topic: Betting as prediction  (Read 3104 times)

Crunch

  • Members
    • View Profile
Betting as prediction
« on: December 19, 2019, 02:11:28 PM »
You guys may find this interesting. I really liked InTrade but it got regulated out of existence. This one fills the gap: electionbettingodds.com.

There’s lots of reasons to expect predictive betting like this to be more accurate than polling (reasons given at site) and I found InTrade to be much better than just polling or a mix of betting and polling. I think especially now this may be more accurate.

Interestingly, Trump has broken above 50% for the first time today. It fluctuates as bets are made so it could move a bit but he’s clearly trended up in the betting.

Biden remains the front runner for the Democratic nomination by quite a margin and he’s also trending up.

Anyway, have fun with it.


TheDrake

  • Members
    • View Profile
Re: Betting as prediction
« Reply #1 on: December 19, 2019, 03:25:14 PM »
I think such a platform is dangerous. What's to stop George Soros from manipulating it to make a candidate look better?

TheDrake

  • Members
    • View Profile
Re: Betting as prediction
« Reply #2 on: December 19, 2019, 03:44:03 PM »
Quote
It is based in the U.K., where regulations are less onerous. It does not accept American traders due to regulations

Well that sort of solves the Soros problem, but introduces another. Is potential accuracy affected by the elimination of people most likely to be very knowledgeable on the topic? I do know that prediction markets can be quite accurate and useful, just wondering about this one.

LetterRip

  • Members
    • View Profile
Re: Betting as prediction
« Reply #3 on: December 20, 2019, 02:03:55 AM »
The number and size of bets are small enough that they are currently easy to manipulate over the short term, though longer term trends tend to be more robust.

see this article at freakonimics,

http://freakonomics.com/2008/10/02/manipulation-in-political-prediction-markets/

Crunch

  • Members
    • View Profile
Re: Betting as prediction
« Reply #4 on: December 21, 2019, 09:39:53 AM »
Interesting
Quote
They find that manipulation may knock market forecasts off course for short periods, but they are rarely off base for long. And even despite these missteps, prediction markets have amassed a pretty impressive record, out-predicting pollsters and pundits. That is, prediction markets are clearly imperfect, but I remain convinced they are the least imperfect tool we have for tracking the political race.

So manipulating these markets is not easy and only makes short term impacts. I tend to agree with freakanomics that these are better than polls.

election betting odds does this:

Quote
One tricky thing is the rare occasion when liquidity is very low for a candidate. For example, one could imagine a candidate who has no chance -- so low that almost nobody is bothering to trade on the person's chances. In that case you might see a wide bid-ask spread. In the most extreme possible case, the bid might be 1% and the ask 99%. In that case the average would compute as 50%, which would be pretty clearly misleading for a candidate so obscure that nobody bothers to trade. It would not be good if this site suddenly put that candidate in the lead at 50% just because nobody was trading that person. The solution this site uses is: In cases where the bid-ask spread is greater than 10%, the bid alone is used, because it provides a solid lower-bound. In cases where the bid is greater than 50% and the bid-ask spread is greater than 10%, the ask alone is used. (These cases are extremely rare; the average bid-ask spread is around 1% and nearly all the time all of the candidates' spreads are well under 10%.)

Looking at their prediction accuracy, it’s very good. It’s accuracy slightly breaks down around 40%-50% predictions (coin toss).

Crunch

  • Members
    • View Profile
Re: Betting as prediction
« Reply #5 on: February 06, 2020, 09:56:44 AM »
So following the traders, it really does a good job of reflecting expectations at this specific point in time.

Sanders has really surged and Biden has gone into free fall. Sanders is a massive favorite in the New Hampshire and Nevada primaries - currently 75.2% and 66.6% respectively.  He’s at 38.9% on the overall primary win but if NV and NH play out as expected that will probably blow over 50%. Sanders has risen to 16.1% on the general election (Biden under 6% and Bloomberg holding at about 10%).

I’m looking forward to these next primaries and see how the predictive power of this works out.

Crunch

  • Members
    • View Profile
Re: Betting as prediction
« Reply #6 on: February 14, 2020, 12:22:50 PM »
Wow, in the last week Bloomberg has rocketed past Sanders in the market. Bloomberg is up 12.5 points while Sanders has fallen 6 and the majority of that was in just the last day.

There are lots of media pieces going out today that Bloomberg is now the safe choice for Democrats so that's probably pumping Bloomberg's valuation up quite a bit. LOL, the Democrats just really don't want Sanders! Plus, dropping a $1,000,000,000+ of his own money gives Bloomberg the appearance of support so that may influence his valuation a lot.

Klobuchar remains in a distant single-digit valuation despite recent strong showings - she's actually behind Hillary Clinton who's not even running....yet. Pete the Cheat just can't get any love in the market, he's really down there and we're almost to the "Who's Biden?" point despite being the DNC's hand-picked nominee.

Crunch

  • Members
    • View Profile
Re: Betting as prediction
« Reply #7 on: February 23, 2020, 09:20:07 AM »
So it accurately predicted Sanders win in Nevada. It now has Sanders above 50% for the first time on winning the primaries while Bloomberg struggles to recover from a complete collapse of his valuation - it shot up and back down so quickly I almost wonder of Bloomberg put a few million in to artificially pump it up.

I think they need one for Bernie winning the primaries and still not getting the nomination.

oldbrian

  • Members
    • View Profile
Re: Betting as prediction
« Reply #8 on: February 24, 2020, 11:47:57 AM »
And that is one of the things I dislike about the Dem primaries - all of the super-delegates or whatever they are called that get to decide the nomination without any input from the people.
If I recall correctly, Clinton and Sanders were very close to tied after the primaries were done, but all of the supers had already declared for Clinton, which pushed her over the threshold.

LetterRip

  • Members
    • View Profile
Re: Betting as prediction
« Reply #9 on: February 24, 2020, 11:55:03 AM »
And that is one of the things I dislike about the Dem primaries - all of the super-delegates or whatever they are called that get to decide the nomination without any input from the people.
If I recall correctly, Clinton and Sanders were very close to tied after the primaries were done, but all of the supers had already declared for Clinton, which pushed her over the threshold.

The primaries were changed so that Superdelegates play less of a role.  Now if one person gets the majority of votes, they win.  If only a plurality occurs, then superdelegates make up about 20% of the total delegates.

NobleHunter

  • Members
    • View Profile
Re: Betting as prediction
« Reply #10 on: February 24, 2020, 01:38:28 PM »
And that is one of the things I dislike about the Dem primaries - all of the super-delegates or whatever they are called that get to decide the nomination without any input from the people.
If I recall correctly, Clinton and Sanders were very close to tied after the primaries were done, but all of the supers had already declared for Clinton, which pushed her over the threshold.

Clinton got 54% of pledged delegates according to Wikipedia. Close to tied just means he closed the gap but still lost even before the super delegates weighed in.

Crunch

  • Members
    • View Profile
Re: Betting as prediction
« Reply #11 on: February 24, 2020, 02:05:18 PM »
And that is one of the things I dislike about the Dem primaries - all of the super-delegates or whatever they are called that get to decide the nomination without any input from the people.
If I recall correctly, Clinton and Sanders were very close to tied after the primaries were done, but all of the supers had already declared for Clinton, which pushed her over the threshold.

The primaries were changed so that Superdelegates play less of a role.  Now if one person gets the majority of votes, they win.  If only a plurality occurs, then superdelegates make up about 20% of the total delegates.

I seem to recall that the delegates are no longer required to vote for their pledges candidate after the second round. Is that right? Could have dreamed it.

TheDrake

  • Members
    • View Profile
Re: Betting as prediction
« Reply #12 on: February 24, 2020, 02:40:45 PM »
In a contested convention, pledged delegates can indeed switch their votes. On the first round. Pledged delegates are not necessarily bound delegates. That isn't a new rule.

TheDeamon

  • Members
    • View Profile
Re: Betting as prediction
« Reply #13 on: February 24, 2020, 03:59:15 PM »
Pretty safe bet that a plurality outcome is likely for the Democratic Convention this year.

A recent CBS news poll also found that 1/3rd of Democrats expect Trump to be re-elected this year, and Republicans are very bullish on Trump winning too.

Crunch

  • Members
    • View Profile
Re: Betting as prediction
« Reply #14 on: March 02, 2020, 07:36:04 PM »
For those interested, Bernie has gone into freefall. He’s at 41% and sliding as investors realize the fix is well and truly in.

Biden has rocketed up to 44% - he was in single digits just a week ago. It’s interesting to see what Super Tuesday will do to these valuations.

Since the market is rigged, should these investors file lawsuits?