How about a simple dot product as the measure of alignment between Voter(i) x Politician(j) on their respective position vectors?In this framework, "get out the vote" efforts can be seen in terms of politicians trying to amplify the magnitudes of their position vectors so that their voters feel more strongly about the discrepancy between their preferred choice and the alternatives (indifferent voters being less likely to vote).

Such cynicism for one so young!You could have a "Cone of Principles" meaning that there is a limit to how far they would go.P

The Principal Vector of Principles?P

Elector Vector?

The [$]i[$]th voter has 'Elector vector' [$]\textbf{v}_i[$], a vector representing his views. The [$]j[$]th politician has principles [$]\textbf{p}_j[$], a vector. Note that these are not necessarily unit vectors. The view and principles vectors can include dimensions for hairstyle, making this model particularly important in the US.The [$]i[$]th voter will vote for the [$]j[$] that maximizes [$]\textbf{v}_i.\textbf{p}_j[$], as long as it is positive otherwise they don't vote (*). The politicians can adjust their [$]\textbf{p}_j[$]s (PJs, pyjamas, geddit?!) to make them more electable, i.e. to maximize number of voters according to (*) above.Each politician is constrained by his Principal Principles (the vector [$]\textbf{P}_j[$]), such that [$]|\textbf{P}_j-\textbf{p}_j|\le H_j[$] where [$]H_j[$] is the politician's level of hypocrisy. P

QuoteOriginally posted by: Traden4AlphaIn the US, we see even stranger patterns in which some voting districts vote 60:40 (or 600,000:400,000 to clarify that it's not a random error issue) and other voting districts vote 400,000:600,000 and the combined total is suspiciously 50:50. The underdog in one geography is the alpha dog in another.Suppose it costs me $101 to move 1 vote in district A, and $99 to move 1 vote in district B. This is at the moment I have 50% of votes in both districts. My opponent is facing the opposite situation. Generally we will both move votes in districts where it is cheaper. Diminishing marginal returns will kick in when we both get 60% in our cheap district.Now suppose it is expensive to move votes. But at the start it is something like 55-45. The one you are higher in is probably the one where it will be cheaper for you to move the first vote. So any original differences will be magnified. If it is very expensive to move votes, the effect will be even greater, to where we both focus 100% of our energies on areas where our cost-benefit is slightly better. If they start out with similar money, and similar polls - say 51-49 - it will be magnified in each district to still end up with 51-49.If such an effect exists, it could be caused by the social nature of candidate selection and turnout. I only four in 10 people are driving to the polls for me, then it costs more to get that fifth person. Because it will be harder for him to find a ride. If I already have 60% of people in a district voting for me, it will be a small push to get the seventh person into a car with his best friend and voting for the same person.

QuoteOriginally posted by: farmerQuoteOriginally posted by: Traden4AlphaIn the US, we see even stranger patterns in which some voting districts vote 60:40 (or 600,000:400,000 to clarify that it's not a random error issue) and other voting districts vote 400,000:600,000 and the combined total is suspiciously 50:50. The underdog in one geography is the alpha dog in another.Suppose it costs me $101 to move 1 vote in district A, and $99 to move 1 vote in district B. This is at the moment I have 50% of votes in both districts. My opponent is facing the opposite situation. Generally we will both move votes in districts where it is cheaper. Diminishing marginal returns will kick in when we both get 60% in our cheap district.Now suppose it is expensive to move votes. But at the start it is something like 55-45. The one you are higher in is probably the one where it will be cheaper for you to move the first vote. So any original differences will be magnified. If it is very expensive to move votes, the effect will be even greater, to where we both focus 100% of our energies on areas where our cost-benefit is slightly better.The cost of vote shifting is not at all symmetric for at least two reasons:First, although both candidates may have an equal cost to shifting an undecided voter sitting in the position-space boundary between the two candidates, any attempt to recruit a contested voter may have different effects on the turn-out of the candidates' core voters. If a liberal candidate flexes on gun control to attract a contested semi-conservative voter, that liberal might lose a different number of their core voters than if the conservative candidate flexes on gay marriage to attract a contested semi-liberal voter.Second, get-out-the-vote efforts can have a different cost-per-vote for different candidates to the extent that the different groups of voters reside in distinctly different types of geographies. It may be easier for liberals to "get out the vote" in dense urban areas than for conservatives to "get out the vote" in rural areas.

The percentage of intellectual voters is tiny. If Bill Clinton buys an ad saying he will "build a bridge to the 21st century" nobody is alienated.

QuoteOriginally posted by: farmerThe percentage of intellectual voters is tiny. If Bill Clinton buys an ad saying he will "build a bridge to the 21st century" nobody is alienated.Except the Fundamentalist nutters who think "bridge to the 21st century" is code for gay abortionists confiscating everyone's guns.

QuoteOriginally posted by: Traden4AlphaQuoteOriginally posted by: farmerThe percentage of intellectual voters is tiny. If Bill Clinton buys an ad saying he will "build a bridge to the 21st century" nobody is alienated.Except the Fundamentalist nutters who think "bridge to the 21st century" is code for gay abortionists confiscating everyone's guns.Very few swing voters are that smart.

QuoteOriginally posted by: farmerQuoteOriginally posted by: Traden4AlphaQuoteOriginally posted by: farmerThe percentage of intellectual voters is tiny. If Bill Clinton buys an ad saying he will "build a bridge to the 21st century" nobody is alienated.Except the Fundamentalist nutters who think "bridge to the 21st century" is code for gay abortionists confiscating everyone's guns.Very few swing voters are that smart.You're no doubt right. And yet the hyper-partisan pundits on both sides are observant and will punish a candidate who "abandons their base" by compromising (==hypocritizing) to expand their voting base.

QuoteOriginally posted by: PaulThat sounds like 1990's Nonlinear Dynamics with economic agents and noise traders and whatnot!!! Great fun was had by all, but nothing useful emerged...Any model is just a slight translation of information to go from inputs to outputs. Anybody who downloads an open-source robodialer can make better election predictions in two hours than someone who thinks for an entire year. If you add information - which seems to be the goal of that hot new trend trader who has an "information czar" - it is better than trying to twist the existing information around a little more. And new information is incomparably superior to developing a new theoretical model for which the input information is not actually available. The ticker is what we have to work with, mostly.The ticker theoretically has near infinite information. I found that the same trend-trading system that worked in 1-minute timeframe also worked in 1-day timeframe. In both cases, it took me months to fully understand what I was actually looking at. And the two things observed in the same way were completely different; the agents or dynamics or whatever were not even remotely similar. What was similar was the peephole, which nobody else was looking through. Because presumably they were using the same tools, with the same blind spot, on 1-minute and 1-day tickers.