Categories
Options
politicalbetting.com » Blog Archive » The uncertainty principle. A step by step guide to handling
politicalbetting.com » Blog Archive » The uncertainty principle. A step by step guide to handling confusing polling in your betting
The EU referendum polling is all over the shop, with a stark divide between the phone polls, which show a clear Remain lead, and the online polls, which show it neck and neck with Leave perhaps fractionally ahead. How on earth are we supposed to cater for this in our betting?
0
This discussion has been closed.
Comments
I guess the key thing is whether the probabilities one assigns are correct or not.
https://royaleleseaux.wordpress.com/2016/05/22/eu-referendum-the-choice/
It isn't a betting post; it's an opinion piece. Not claiming it's neutral, and I expect not a few to disagree, but I think it fairly reflects my thinking, and that of others where it influenced mine.
http://www.dailymail.co.uk/news/article-3603793/Why-quit-EU-Cameron-s-guru-Friend-strategist-Steve-Hilton-breaks-ranks-Brexit-say-Britain-literally-ungovernable-unless-power-self-serving-elite.html#ixzz49SH7WCRE
The difficulty with this form of analysis is that it generates a spurious precision that is simply not warranted by the underlying data. So Alastair thinks there is a 77.7% probability of remain winning plus or minus 5%. But that assumes, as he acknowledges, the heroic assumption that the probabilities he has assigned to the different possibilities at the start of his analysis are in any way meaningful. These are the known unknowns and, just like the latest forecast of a year long recession, all of the outcomes of the model are driven by these inputs and the assumptions behind them.
What the decision tree does do is force you to think about the different possibilities and identify the uncertainties which might affect the result. It also forces you to acknowledge how these different possibilities interact with one another. It undoubtedly helps clarify the issues and it may well give you a better feel for the probabilities than your gut. But not much, it is too dependent upon randomly allocated probabilities at the start.
Still a very interesting thread header Alastair. Much appreciated.
Cameron should have listened to the intellectual godfathers of Cameroonism: Gove and Hilton.
"In addition, it includes reference to a capital markets union, which will apply to all 28 EU member states as part of completing the single market. This talks about addressing “bottlenecks” like ‘insolvency law’, ‘company law’, ‘property rights’ and strengthening cross-border risk-sharing through addressing the ‘legal enforceability of cross-border claims’. It also references common standards, greater harmonisation of accounting and auditing practices, and deepening integration of bond and equity markets.
The UK will have no veto over this. In fact, we agreed to facilitate it in the recent negotiations and, in any event, deepening completion of the single market is current HMG policy."
Both could be wrong of course.
Answers should be quoted to the accuracy of the model, which in this case is probably not even 1 significant figure.
Small point, maybe, but the inability of the legal profession to understand the accuracy of statistics ("one in 73 million chance of two cot deaths”) has had very serious consequences in the past (Sally Clark).
I would assign lower probabilities to the phone polls being right and a higher one to the online polls.
Insolvency: COMI. I'd also note that we've just acceded to the Cape Town Convention for aircraft which trups UK insolvency law with a trans-national scheme.
Cross border claims: there's lots of law on reciprocal enforcement already. You're implying that there's a corpus of European law which is relevant here. Evidence, please.
Accounting/auditing. That's determined by IFRS. The US has held out with GAAP for reasons which are unclear.
Bonds: you do know that these are debt capital markets don't you. Equity, see equity CM.
We've had a tremendous amount of influence over this.
The same engineer builds another bridge. The first person to cross the new bridge does so safely.
Would you now use the bridge?
Until the polling disaster at the general election has been completely understood, there is no point in attaching any value to a series of discrepant online and telephone polls.
It is of no real consequence that the polls for the London Mayoralty were correct, just as it is of no real consequence that the first person to cross the new bridge has done so safely.
The engineer still builds bridges that can still give way suddenly and without warning.
I have some numbers btw HOFER is ahead 144006 votes with 885437 postal ballots issued and uncounted.
So powerful do they think Turkey fits the narrative that they're still running with it despite the fact that by now everyone knows about the veto.
Normally being found telling a barefaced lie would be suicidal but on this occasion they still think it's worth hanging on to.
My guess is they think it's just too good to let go and as both sides are facing credibility questions they think it's worth the risk
The area where the polls were wrong was over the Lab/Con split nationally. Obviously important, but does not discredit the whole polling system. The phone polls were closer to the truth.
Assess the probabilities yourself, but beware the power of wishful thinking.
Interesting diagram, Mr. Meeks. May be worth adding a time factor, as there's still plenty of time for polls to be accurate today but sentiment to shift over the next month.
"The case of Turkey is interesting"
With respect Roger, I think you're missing the point. Whether we have a veto or not is now irrelevant. Brakes when you're hurtling down a hill are only useful if you can or will use them.
With Cameron lying continually, and only now showing his true Euro-fanaticism, we know he won't use it. Or at least. may not do.
Nor will Farron or the Labour Party. If it exists, it's a virtual veto.
D'you happen to have a Twitter account? I'll RT it, if so, or just post it, if not.
http://www.bbc.co.uk/news/uk-politics-eu-referendum-36355564
No, I don't twat - just blog!
Postal votes:
Total 885437
Expected returned 740000
Expected valid return 700000
% needed for Hofer
41.87
40.27
39.71
Steve Hilton gets it.
Would the UK really veto it if the rest of the EU wanted it to join, and was putting heavy pressure on the UK Government?
No doubt we'd ask for a long transition time for free movement to kick in, probably 12 years+ when the administration agreeing to it had safely left office, but that decision would be irrevocable once the ink was dry.
It's unlikely to happen because Cyprus would almost certainly veto, and the French reject in a referendum, but, if they didn't, or another way of accession was found, it might - the EU has been negotiating with them for a long time, and has already found a way to grant visa-free access.
I may read it in chunks through the day (I often take mini-breaks from writing).
Mr. Pulpstar, any timetable?
Top article on BBC news site.
But well down the list of most read articles. I don't think people are believing it any more.
The lawyers accepted the word of Roy Meadows. They didn’t, for example, think to doubt it or to obtain an expert statistical opinion in the trial.
There is another famous example of lawyers and statistics in the OJ trial, which appeared in the late David MacKay’s book.
Given postal voting, purdah should be moved forward several weeks.
Excellent blog. Have circulated your conclusions (the summary of the choice anyway) - hope you don't mind...
Included a link, of course...
But that's the point I was making. As the story unfolds it becomes ever clearer that there is no chance of them joining but the lie becomes ever more publicised.
Mr. Hopkins (2), there's another option: namely that people feel saturation coverage has been reached and, and whether they believe campaign claims or not, they're just tired of hearing them.
http://www.breitbart.com/big-government/2016/05/22/exclusive-kathleen-willey-raped-sexually-assaulted-bill-clinton/
But politicians cannot help themselves, they lie routinely. Dressing up supposition as fact is a lie in my book. The end justifies the means as always.
So holding the referendum was only a mistake because it exposes the lying toe-rags we elect. Politics is the less for it. But even more reason to hold referenda if we can't trust what our elected representatives say.
Deal or no Deal: The Cameron in Wonderland renegotiation with the EU that never happened, was never intended to happen, that never could have happened with Mad Hatter Cameron at the wheel.
Or perhaps the original story is meaningless.
https://twitter.com/LeaveEUOfficial/status/734643592208224256
An expert is no better than a computer: if you put rubbish in, you get rubbish out.
The trouble with believing experts, eh!
Now, where else might this apply?
It was in no small part because of him and Michael Gove that I originally supported Cameron for leader back in 2005.
Sure, the PM is indirectly elected by people whom we elect. That is exactly the case for the President of the European Parliament. The fact that we choose not to find the EP very interesting is a different issue.
It's definitely possible to criticise the EU for lack of direct democratic influence - at heart it still hasn't quite decided if it's a prototype country (which needs direect democratic control) or merely a practical arrangement of common interests (which is largely a matter for professional civil servants), and its arrangements reflect that. But grumbling over the lack of direct elections is a distraction.
While he may always have come across as a bit of a hipster, I don't think he fits neatly into any category of the Tory Party.
http://www.zerohedge.com/news/2016-05-22/trump-victory-odds-soar-hillary-clinton-death-cross-strikes
FutureLearn, a subsidiary of the Open University, is running a MOOC on the EU referendum starting 13 June and continuing after the result.
https://www.futurelearn.com/courses/eu-referendum/1
"The course will guide you through the referendum – before and after the vote. We’ll look at the history, break down the key issues and analyse the result through videos, interviews, articles, discussions and more."
"If you haven’t already, feel free to introduce yourself to your fellow participants and share your thoughts on the debate."
"We also have a number of resources for you to explore. You can find insight and analysis on the referendum and other topics on European Futures, our academic blog. You can also explore how people are talking about the referendum on Twitter through the Neuropolitics Research Lab’s social media tracker or read a bit more about the issues in the e-book Britain Decides."
I've signed up for this. Could be interesting. The Social Media Tracker could also be interesting.
http://www.pol.ed.ac.uk/neuropoliticsresearch/sections/remote_content
If France or any other country in the EU had a referendum denying turkey, they would be forced to have another one until the result was that which Brussels wanted.
Our Parliament has a legitimacy which the EP does not really have. And that is the problem. What is the EP for? Why should it exist at all? Whose Parliament is it? Which country? Which demos?
Who is a leftie? Anyone who relies on rational argument. It wasn't reason that won Agincourt. It wasn't reason that enabled Francis Drake to plunder the Spanish Empire. It wasn't reason that defeated the Nazis - it was rage. Massive rage.
Massive rage for ever! Hopefully the new Austrian President will kill all his opponents!!
"On leaving Lokey said, "I can't be a 24-hour cheerleader for Hezbollah, Moscow, Tehran, Beijing, and Trump anymore. It's wrong. Period. I know it gets you views now, but it will kill your brand over the long run. This isn’t a revolution. It's a joke."
https://en.wikipedia.org/wiki/Zero_Hedge
The official site is here: http://wahl16.bmi.gv.at/
I made that mistake once, but only in a minor European meeting of no real consequence. And as it was only a European meeting, it was mostly irrelevant. They're decided on politics anyway.
As politicians/administrators like to say ... 'Scientists on tap, not on top."
The invasion of our minds can be just as insidious as the physical presence in our lands. Worse in some ways.
This is because he's ahead on the the already counted votes so his opponent needs a fixed absolute number of votes from a smaller pool (and hence needs a higher percentage)
I've see some Remainers in a most unflattering light, Matthew Parris in particular. And much unexpected tolerance in those who're willing to put aside old tribal differences and share a common vision of Britain.
Whatever the result - this referendum has truly shaken the political kaleidoscope.
It seems to me that in your argument, a lot depends on the point that Remain is not a vote for the status quo but the swift development of a pan-European uberstate. Of course, the future won't be the same as the present, but it seems unlikely to me that it will develop in the way you describe. I think change will be slower, less certain and with a lot more compromise.
(But it's true that when I was typing this I was listening to Radio 4 wittering on about expert reports on the EU......)
"They believe a bit of damage to their credibility is a price worth paying."
World War Three? A 'bit' of damage.
Dr Palmer, The difference is that if the UK decides to let Turks have free admittance, that is the will of the people living here. If not, we can do something about it. How in the land of eternal sunshine with twenty seven, and counting, other states can we do something about it?
Seconded! (Or thirded) - very well set out.
I think you've got a 'primaRy' when you mean primaCy
I think his best idea was to try and replace the Tory membership with metropolitan, liberal, pro immigration, socially liberal types.
Shame he never inplemented that idea