For the latest state polling data:
http://www.electoral-vote.com/
For the latest National polling data:
http://www.pollingreport.com
1) State and national polls polls are adjusted assuming that
70% undecided voters will vote for Kerry.
2) State polls are weighted by population by calculating the
percentage of state to total votes nationwide for the last 3
elections.
3) Third party candidates not considered.
4) The Diebold/ES&S factor has not been considered, so
take these results with a BBV grain of salt.
IMPORTANT NOTE: The analysis is only as good as the polling
data. The 13 national polls include CNN/Gallup, Fox and IBD.
It is fair to say that they have a tendency to pull Kerry's
average numbers down somewhat.
Over the last three presidential elections:
Total Votes (millions):
Dem: 138.75 (52.60%)
Rep: 125.03 (47.4%)
.....................................................................
Electoral Vote Forecast/ Simulation
Latest State Polls (adjusted for undecided allocation):
Kerry 52.25%
Bush 47.75%
Avg Sim. Kerry EV: 324
National Polls (July):
..........Actual Adj.
Kerry 49.1 53.10
Bush 45.1 46.9
Avg. July Bush Job Rating:48.0
Simulation Using Latest State Polling
SIM Electoral Votes
1 331 Win
2 315 Win
3 336 Win
4 322 Win
5 313 Win
6 321 Win
7 346 Win
8 327 Win
9 307 Win
10 322 Win
Avg 324
Win 10
I rely on Monte Carlo simulation, a method which uses
probabilities of events and applies random numbers to generate
many outcome scenarios. For this analysis, I forecast
individual state results from the latest poll numbers to
generate a probability analysis of a Kerry win.
In this case, an "event" is Kerry's share of each
states popular vote. The "outcome" is the simulated
probability of Kerry winning the election. This is based on
the combination of states which go for Kerry and from the
calculation of his total electoral votes.
For example, if we run the simulation 10 times and Kerry's
electoral vote total exceeds 269 votes in 9 of the 10
simulations, we can estimate that he has about a 90%
probability of winning the election. I assume each poll has a
margin of error of +/-4%
Kerry has a defined probability of winning each state, for
example:
Kansas: .001
New York: .999
Florida: .807
The probabilities are determined as a function of the latest
polling numbers. For instance, if the latest Kansas poll is
60% for Bush, 40% Kerry, then Kerry has an approximate 1 in
1,000 chance of winning the state.
On the other hand, if Kerry is ahead by 5% in Florida, then
there is an 80.7% probability of winning the state.
Using Excel's random generator function, each calculation
returns a random number X between 0 and 1. If X is less than
the probability of Kerry winning the state, the state goes to
Kerry, otherwise it goes to Bush.
For instance, if X = .72 for Florida, then Kerry wins the
state. The random number X= .72 is LESS than the .807
probability that Kerry will win FL, based on the latest polls
which have him leading by 5%.
The calculation runs simultaneously for all states. The
electoral votes for those states which fall in Kerry's column
are added.
Remember, we assume that Kerry's polling percentages numbers
are increased by allocating 70% of the undecided vote in the
simulation.
For example, say Kerry is leading a state by 46%-44%. Adding
70% of the 10% undecided voters adds 7% to his total, so he is
projected to win by 53%-47%.