Really really informal simulation of the election

Presidential Simulation.R

library("hash")

# States treated as going democratic with probability 1
DemStates <- hash()
.set(DemStates, "WA"=12, "OR"=7, "CA"=55, "NM"=5,
                "MN"=10, "WI"=10, "MI"=16, "IL"=20,
                "PA"=20, "MD"=10, "DC"=3, "DE"=3, 
                "NJ"=14, "CT"=7, "RI"=4, "MA"=11, 
                "VT"=3,"ME"=4,"HI"=4, "NY"=29)

# States treated as going republican with probability 1
RepStates <- hash()
.set(RepStates, "ID"=4, "UT"=6, "WY"=3, "MT"=3, "ND"=3,
                "SD"=3, "NE"=5, "KS"=6, "OK"=7, "TX"=38,
                "AZ"=11, "MO"=10, "AR"=6, "LA"=8, 
                "MS"=6, "AL"=9, "GA"=16, "TN"=11, 
                "KY"=8, "WV"=5, "SC"=9, "AK"=3, "IN"=11)
     
# Swing States
# First entry in each value of the hash is the electoral 
# college vote
# Second entry is the intrade probability of the state 
# going democratic
SwingStates <- hash()
.set(SwingStates, "NV"= c(6, 0.812), "CO"=c(9, 0.762), 
                  "IA"=c(6, 0.750), "VA"=c(13, 0.741), 
                  "NC"=c(15, 0.370), "FL"=c(29, 0.678), 
                  "NH"=c(4, 0.786), "OH"=c(18, 0.779))

# electoral college votes needed to win
kCountNeeded <- 538/2

Run.Simulation <- function () {
  # Returns TRUE if simulation predicts democratic win
  # ... and FALSE otherwise
  # treats each state as independent (I know - not true...)

  dem.count <- sum(values(DemStates))
  for (i in keys(SwingStates)) {
    result <- runif(1)
    state  <- SwingStates[[i]]
    if (result < state[2]) {
      dem.count <- dem.count + state[1]      
    }
  }
  if (dem.count > kCountNeeded) {
    return(T)
  } else {
    return(F)
  }
  
}

Run.Simulations <- function(count) {
  # returns portion of simulations that returned a
  # democratic win 
  i <- 0
  total.results <- c()
  while (i < (count + 1)){
    single.result <- Run.Simulation()
    total.results <- append(total.results, single.result)
    i <- i + 1
  }
  return(sum(total.results)/count)

}

Run.Simulations(10000) #=> 0.9894