Posts by David Hood
Last ←Newer Page 1 2 3 4 5 Older→ First
-
Hard News: The Righteous Humour, in reply to steve black,
Nate Silver before Nate finished telling the Republican strategists the list of things they needed to change to win the next election
Well, Nate Silver has described himself as basically a libertarian/ Mitt Romney kind of guy personally (though choosing not to vote because of his profession), it is just he is primary a quantitative analysis kind of guy.
-
Hard News: The Big 2012 US Election PAS Thread, in reply to George Darroch,
Actually running these is slightly beyond my abilities, but I like the concepts.
They are fairly easy to do in excel (anyone keen could google monty carlo excel) but the real trick is correctly judging the probabilities you assign to potential outcomes, for which you generally need in depth knowledge of the area.
-
Here is an interesting article about how back in 1952 the press were saying the election was too close to call while the computer (The Univac) kept spitting out results that it was going to be a landslide.
-
While it is pretty thouroughly explained, I might as well offer my take as well.
538 is based on a kind of modelling generally called a Monty Carlo method, where you run the same test (election) lots of times with different random numbers (voter support) to see how likely each potential outcome is. In 538's case this is running mock elections many, many times see how changing the support in each state affects the final outcome.
The expert knowledge comes in knowing what range of numbers to use as inputs, in this sense how far the polls and other input data should be randomly shifted to reflect potential election day results. Basically, assessing how much the polls are potential wrong by. I suspect this is where the republican data modelling broke down, in the expert judgement about how far wrong (and in what direction) the polls are.
There is also some model building expert judgement in the way in which all the inputs are not independent- For example, if California had gone to Romney, then it is much more likely most other states would also have gone to Romney than it is that California would be an isolated incidence.So by assessing how often an Obama result came up in simulations, and the strength of the input data, Nate Silver derived his headline figure of 92%. This figure is technically an oversimplification, as no indication of the confidence of the parts of the input was made (Nate Silver's model is a private one, unlike some of the University based ones which were predicting an Obama win as more likely but also providing the estimates of the strength of the conclusion and the data assumptions ).
I think at one point Nate Silver did say he was taking the more conservative options- essentially the Republican analysts were so certain of victory he reduced Obama's chance of winning on the assumption that they might know something he didn't. This is one reason that until close to election day Silver was rating Obama lower than some of the other such sites.
With a headline statistic like 92% you can say that given enough elections under the conditions set in the model, Romney would win 8% of the time (in that sense like the coin toss analogy). This is what the Electoral Vote Distribution graph on 538 shows- the distribution of potential results from the model.The percentage headline figure ultimately derives from an estimate of how different the polls are from final turnout. If you assumed the polls were a totally accurate representation of the results of voting day, then Obama had a 100% chance of winning when giving him 51% support. If you assume there is some uncertainty (and it is in the professional polling analyst's judgement here) in some of that uncertain time Romney will win, and some Obama (and this is where the repeated analysis of possible outcomes comes in). Because of the uncertainty, If I run 9 elections with these assumptions (92% chance of victory) in the normal course of events I would expect my assumption to wrong somewhere between 0 and 9 times, but typically 1 time, more likely between 0 to 2 times.
-
While very much still in the anger stage of loss (I think the republican establishment genuinely was certain of victroy) the line among house Republican's seems to be forming that "as Obama ran with no policies, he has no mandate and we are free to oppose him". Now, I thought last week Republicans were clear Obama was running to increase taxes on the rich. But not acknowledge their own personal history has been a characteristic of recent behaviour.
Interesting result in California, the Democratics look like they have a super-majority, which means they can actually do things like change land taxes beyond the 1% specified in 1978 and other things involving governing effectively.
At the moment the Republican 2016 field seems to be being talked about in terms of Ryan (tea-party heir), Rubio (we need a latino), Christie (currently hated but America sees him as a moderate, Bush (a Bush), or some combination of the above. -
Hard News: The Big 2012 US Election PAS Thread, in reply to John Armstrong,
I am thinking mainly about polls for things like preferred Prime Minister or current voting preference, which perhaps don’t add much to anyone’s understanding of the policy platforms that the prime Minister or the various parties are standing on.
In the reading I have been doing this evening, it has been suggested that polls convey an element of the wisdom of the crowds- people are aware that their knowledge is imperfect, and and a bunch of people supporting something suggests there may a a reason for them doing so. Now that reason might be that people are judging the preferred Prime Minister by that he looks nice to them on TV, but it does show that there is a consensus about something.
This could either lead to people wanting to take the next step in answering the why question (why do a lot of people support x) and filling in their known imperfect knowledge, or it could lead to judgements about the people who support x if one is certain in ones knowledge on a matter. -
Hard News: The Big 2012 US Election PAS Thread, in reply to John Armstrong,
But people only have so much time and energy to absorb information. With so much media content focusing on poll results, it surely must be less likely that the electorate is well informed about policy
I'm actually kind of hopeful here, for reasons I think showed in the U.S. campaign. I would tend to say that Romney pitched his message a the level of the media soundbite, but so many people are now used to the ability to follow up detail on the internet that the inconsistencies in his "budget" were well known pretty quickly. In this respect I think both Labour and, even more so, the Greens did well in providing readily accessible policy detail at the last election via the internet, and it is a trend I hope continues and flourishes.
-
Hard News: The Big 2012 US Election PAS Thread, in reply to Bevan Shortridge,
That states are being projected for a candidate in the east while people might still be deciding whether to bother going to vote in the west.
A (very) quick bit of digging into the results suggests that it is not as much of a difference as people worry about (and they do worry about it). Possibly because U.S. elections are mega-elections by NZ standards where you decide everything from ballot initiatives (normally many per an election, California being known for them) through state government (including many jobs that NZ would think of as civil service ones). To save on costs of running elections a lot of places also run their local elections at the same time. Election forms can stretch to many pages. While the Presidental election is the headline act, most of what people are voting on is local.
That said Hawaii is known as the state that doesn't vote, but a certain amount of that seems to be about entrenched power within the state rather than the President being already decided. When the National Popular Vote Interstate Compact is eventually triggered, I would expect it to have little effect on Hawaii's turnout, though they could be far more significant in a truly close election. -
Hard News: The Big 2012 US Election PAS Thread, in reply to John Armstrong,
maybe that’s fair enough. But in the context of a proportionally elected parliament
Yes, and in a proportionately elected parliament, the information about other people's opinions is less useful. However, in New Zealand we do not have pure proportionality, and if I may highlight the plight of the ACT voter (or indeed anyone not supporting National/ Labour/ Greens). If there are so few people willing to declare support for the tiny party that they are nowhere near the threshold, then my vote may be better served (in the calculus of ideological purity versus pragmatic gains) by voting for a major party. On the other hand, if my party of choice* is nudging the threshold, this fact would seem to be a strong motivator. There are even more value to polls for voters in tactical voting around geographic seats (particularly where there are coattails flow on MP effects).
*I wish to make it clear ACT is not the party of my choice.
-
Hard News: The Big 2012 US Election PAS Thread, in reply to Bart Janssen,
There are a whole bunch of psych studies
There are all kinds of individual level psych studies showing you can affect people in the short term in all kinds of ways, I'm thinking more of acutal evidence at a population level with the real world of competing influences. A bit of googling suggests having poll information in a first past the post American election affects a statistically insignificant proportion of voters (it affects a small proportion of undecided voters (and the truely undecided are a small proportion of the voting population)).