Sam Trenholme's webpage
Support this website

Saturday election prediction

 

October 27 2012

In today's prediction, I am using Nate Silver's numbers from yesterday which can be seen at 538.vk.tj to see how likely each candidate is to win, and by how many electoral votes.

To make these numbers, I ran 100,000 simulated presidential votes; this is a type of simulation called a "Monte Carlo" simulation.

No correlation between states

Assuming there is no correlation between the states, we get the following numbers. In other words, I am assuming that Obama getting Ohio does not affect his chances of getting Florida, and vice-versa.
Overall
Romney    8.86%
Tie       0.64%
Obama    90.50%
These numbers are show a considerable bounce for Obama compared to my last numbers.

By state

In these numbers, I have determined a given candidate's chances of winning based on them winning or losing a given state.
Obama gets OH
Romney    3.46%
Tie       0.43%
Obama    96.11%
Romney gets OH
Romney   26.40%
Tie       1.31%
Obama    72.29%
Obama gets FL
Romney    0.28%
Tie       0.03%
Obama    99.70%
Romney gets FL
Romney   13.88%
Tie       1.00%
Obama    85.12%
Obama gets VA
Romney    3.46%
Tie       0.55%
Obama    95.99%
Romney gets VA
Romney   15.27%
Tie       0.75%
Obama    83.98%
Romney gets FL and VA
Romney   23.85%
Tie       1.17%
Obama    74.99%
Obama gets FL and/or VA
Romney    2.78%
Tie       0.43%
Obama    96.80%
In the above blog, we look at the election probabilities based on various conditions. For example, if Obama wins Ohio, Romney has only a 3.46% chance of winning the election.

By electoral vote

Here are the chances of the incumbent (Obama) getting a certain number of electoral votes (EVs):
Obama EV
230-239   0.2%
240-249   0.6%
250-259   2.6%
260-269   6.0%
270-279  12.3%
280-289  17.8%
290-299  19.9%
300-309  14.8%
310-319  12.7%
320-329   7.5%
330-339   4.2%
340-349   1.2%
350-359   0.1%

Mean   295.5
Median 294
"Mean" is adding up all of the possible EVs together and dividing by 100,000. Median is sorting by the number of electoral votes and choosing the middle one.

Alternate point of view

In contrast to this prediction, Dan McLaughlin thinks "Obama is toast".

In order to reduce spam, comments for this entry are now closed