Well here we are again. As with the UK election, as with Brexit, as with many other voluntary voting elections we have an unexpected result with the election of Donald Trump as the next President of the USA. Pollsters are in disrepute because most had Clinton with a modest popular-vote lead, but overconfident modellers deserve their share of the blame for the level of public surprise at the result.
A few days ago, Nate Silver of fivethirtyeight was the target of a terrible Huffington Post article and an argument broke out about whether it was more accurate to say Donald Trump had about a one in three chance of becoming President, or virtually no chance at all. HuffPo was to double down with this rather pretentious piece by a stats prof accusing Silver of overstimating Trump's chances - a piece that has proved to have an exceedingly short shelf life indeed. Silver's model might not look crash hot in the wake of what has happened, but it still looks a great deal better than those that were saying Trump had only a 1% chance of winning.
Showing posts with label USA. Show all posts
Showing posts with label USA. Show all posts
Thursday, November 10, 2016
Wednesday, January 2, 2013
2012 Ehrlich Award For Wrong Predictions Goes To ...
Introducing the annual Ehrlich Award, which at the start of each year will be given to the "wrongest" published prediction I observe of or relating to the previous calendar year. (Of course, if I make a really stupid prediction of my own pertaining to this or any other year, I will happily self-nominate, and even if necessary self-award. )
There are a few groundrules - for instance the predictions need to be vaguely meaningful (in terms of being able to assess whether they have happened - this year's winner stretches that one sometimes), and secondly predictions that carry a stated assessment of chance of falsehood are not included unless that assessed chance is ludicrously low. After all, even odds-on favourites do get beaten sometimes.
The Ehrlich Award is named in (dis)honour of Paul Ehrlich whose lifetime achievements in the field of publishing wrong enviro-scare predictions (oh no, you're an "idiot" if you say they were "predictions"; they were actually only "scenarios") are almost as staggering as the excuses trotted out on behalf of all his false apocalypses. Apparently, not only is it a hit predictively for Ehrlich fans if something vaguely like the prediction happened, or if it could still happen later, but even if nothing like it happened, that's a hit too, because it means people must have listened to the prophet's warnings and, for that reason alone, averted their ways from their sins.
There are a few groundrules - for instance the predictions need to be vaguely meaningful (in terms of being able to assess whether they have happened - this year's winner stretches that one sometimes), and secondly predictions that carry a stated assessment of chance of falsehood are not included unless that assessed chance is ludicrously low. After all, even odds-on favourites do get beaten sometimes.
The Ehrlich Award is named in (dis)honour of Paul Ehrlich whose lifetime achievements in the field of publishing wrong enviro-scare predictions (oh no, you're an "idiot" if you say they were "predictions"; they were actually only "scenarios") are almost as staggering as the excuses trotted out on behalf of all his false apocalypses. Apparently, not only is it a hit predictively for Ehrlich fans if something vaguely like the prediction happened, or if it could still happen later, but even if nothing like it happened, that's a hit too, because it means people must have listened to the prophet's warnings and, for that reason alone, averted their ways from their sins.
Friday, November 16, 2012
As Gillard recovered, so can Abbott?
Advance summary:
1. The view has been advanced that since Julia Gillard has become much less unpopular, then so can Tony Abbott.
2. This view does not automatically follow, because Gillard and Abbott occupy different positions and became unpopular for mostly different reasons.
3. The record since the late 1960s shows that the average recovery in net satisfaction for both Prime Ministers and Opposition Leaders between their worst rating and the next election is about 20 points.
4. However, very unpopular Opposition Leaders are usually removed, and those who make it to elections have so far lost them even if they had ceased to be unpopular by election time.
5. There is one case of an Opposition Leader recovering from a rating worse than Abbott's to a positive net rating within two months, but without winning the election. (Andrew Peacock, 1984)
6. However, the general historical trend is against the likelihood of Tony Abbott recovering to a positive or even near-zero rating before the next election.
1. The view has been advanced that since Julia Gillard has become much less unpopular, then so can Tony Abbott.
2. This view does not automatically follow, because Gillard and Abbott occupy different positions and became unpopular for mostly different reasons.
3. The record since the late 1960s shows that the average recovery in net satisfaction for both Prime Ministers and Opposition Leaders between their worst rating and the next election is about 20 points.
4. However, very unpopular Opposition Leaders are usually removed, and those who make it to elections have so far lost them even if they had ceased to be unpopular by election time.
5. There is one case of an Opposition Leader recovering from a rating worse than Abbott's to a positive net rating within two months, but without winning the election. (Andrew Peacock, 1984)
6. However, the general historical trend is against the likelihood of Tony Abbott recovering to a positive or even near-zero rating before the next election.
Tuesday, November 6, 2012
Thoughts on Forecasting and the US Pres Election
UPDATE 4 pm: Nothing has changed with only Virginia and Florida genuinely close and this is looking like it will be a triumph for state-polling-based models for the second election in a row.
UPDATE 2:40 pm: At present only Virginia and Florida counts are looking really close (as well as NC where Obama has done better than expected but still looks like falling just short). Even if Romney wins all this he still falls short so there is no sign at present of a path for him to victory.
UPDATE 1:40 pm: Obama has in my view won Ohio based on similar methods to those used to model Australian counts in progress.
--------------------------------------------------------------------------------------------------------------
Apparently, some 8% of this site's readership so far is US-based, ten times more than any other non-Australian country to this stage. That inspires me to say a few quick things about the massive logistic exercise unfolding over there at the moment and the debates about what will very soon happen.
For those who are not familiar with the US system, making sense of the endless history-laden data-drenched arguments about how to predict who will be President can be a daunting task. Hopefully the following comments will be useful in informing people about some of the pitfalls when it comes to what to take seriously.
Beware Overfitted Models
No we're not talking catwalk stuff here, but these things are certainly overdressed and each particular one will go out of fashion very quickly. When you see someone claiming to have found a model that predicts a certain candidate will win based on a shopping list of items that have supposedly "predicted" (despite being made after) every presidential election since the year dot, run for the hills. The chief offender I noticed this time was the Uni of Colorado study that claimed that Romney would win, but there were plenty of equally shoddy examples calling it for Obama.
UPDATE 2:40 pm: At present only Virginia and Florida counts are looking really close (as well as NC where Obama has done better than expected but still looks like falling just short). Even if Romney wins all this he still falls short so there is no sign at present of a path for him to victory.
UPDATE 1:40 pm: Obama has in my view won Ohio based on similar methods to those used to model Australian counts in progress.
--------------------------------------------------------------------------------------------------------------
Apparently, some 8% of this site's readership so far is US-based, ten times more than any other non-Australian country to this stage. That inspires me to say a few quick things about the massive logistic exercise unfolding over there at the moment and the debates about what will very soon happen.
For those who are not familiar with the US system, making sense of the endless history-laden data-drenched arguments about how to predict who will be President can be a daunting task. Hopefully the following comments will be useful in informing people about some of the pitfalls when it comes to what to take seriously.
Beware Overfitted Models
No we're not talking catwalk stuff here, but these things are certainly overdressed and each particular one will go out of fashion very quickly. When you see someone claiming to have found a model that predicts a certain candidate will win based on a shopping list of items that have supposedly "predicted" (despite being made after) every presidential election since the year dot, run for the hills. The chief offender I noticed this time was the Uni of Colorado study that claimed that Romney would win, but there were plenty of equally shoddy examples calling it for Obama.
Subscribe to:
Posts (Atom)