I would like to know what is currently wrong with the strategic voting simulations.
In response to questions about my earlier simulation proposals... here it is. The reason that the voters vote in stages and are apparently more willing to support front runners is this: 1) This way as time goes on the situation will hopefully approach equilibrium and we can call it a day. 2) The weight of each pseudovoter grows exponentially in Vote By Result. Therefore if they support a front runner and abandon their favorite, you can be damn sure the favorite wasn't viable. *I revise my earlier idea. Instead I think the population size should grow linearly (so that, little by little, pseudovoters are forced to make compromises). Either that or have the weight grow subexponentitally. So the two things I have proposed are Vote By Result and Probability Feedback Loop. I think that in the last thread or so I explained them, so tell me if something needs touching up. The main purpose of this message however, is just to request some general ideas for what should be included in a strategic voting simulator.
---- Election-Methods mailing list - see http://electorama.com/em for list info
