Proposed method of allocating electoral votes

[pbrower2a]

This looks like a reasonably-neutral way in which to allocate electoral votes with respect to both the state entities and in proportion to the rest of the state's electoral votes. This would apply (it is arbitrary on my part) to states with ten or more electoral votes. It's the more populous states that could gerrymander districts to fit the agenda of a political party that might have an ephemeral success in a year ending in zero. States with small populations can't gerrymander districts as blatantly as can those with large populations.

two to the winner of the state overall

none to those who get less than 5% of the vote in such a state under any circumstances

the remaining popular votes allocated in proportion to the total of relevant votes, but incomplete shares going to the winner of the plurality.

Example -- Texas 2008:


   John S. McCain, III   Sarah Palin           Republican    4,479,328     55.39%   
   Barack H. Obama   Joseph R. Biden, Jr.   Democratic    3,528,633     43.63%   
                  
                  
Neither Bob Barr nor any other nominee won more than 5% of the vote, so their votes are dropped from consideration.

McCain wins both of the at-large electoral votes. President Obama could then get a whole number of votes up to his proportion of  relevant popular votes (which turns out to be roughly 44.0%) of the remaining 32, but only up to a whole number, which would be the whole number short of 14.1, or 14. McCain would get the rest. Thus Texas would have given John McCain 20 electoral votes and Barack Obama 14.

No gerrymandering could game the system.

So how would it have worked in Pennsylvania in 2008?


   Barack H. Obama   Joseph R. Biden, Jr.   Democratic     3,276,363     54.47%    21
   John S. McCain, III   Sarah Palin           Republican      2,655,885     44.15%    0


Neither Nader, Barr, nor anyone else got so much as 5% of the popular vote, so their votes are dropped from consideration.

President Obama gets the two at-large votes of Pennsylvania.

John McCain gets the whole number closest to, but not exceeding, his proportion of popular votes as a share of electoral votes in Pennsylvania. That is roughly 44.7% of the other 19 electoral votes, which would be 8.50 before it is rounded down. McCain would have gotten 8 electoral votes in Pennsylvania and Obama would have gotten 11.

No gerrymandering could game the system.

Someone could argue whether fractions should be rounded up for fractions above .50; such is a matter of taste. The mathematics of such an exercise is simple, and the system could not be gamed through gerrymandering. There would be advantages for nominees to seek out minority votes, and areas that vote "wrong" (like San Antonio in Texas or Orange County in California) would matter. States would be relevant, as there would be 100 electoral votes distributed at large among the States. 

Results would be far closer to the popular vote.

[greenforest32]

Different methods (best to worst):

1. Abolish the electoral college
2. Every state agreeing to award its electoral votes to the winner of the national popular vote (http://en.wikipedia.org/wiki/National_Popular_Vote_Interstate_Compact)
3. States awarding their EC votes in proportion to the state popular vote (60% of the statewide popular vote = 60% of the state's EC votes)
4. States awarding their EC votes by (gerrymandered) congressional districts

[defe07]

Here's an idea I've come up with.

Let voters vote for 1,2 or 3 candidates. The "set" with the most votes, wins the states Electoral Votes.

For example, in a 3-way race between Clinton, Bush and Perot, you could vote for the following options:

- Clinton
- Bush
- Perot
- Clinton/Bush
- Clinton/Perot
- Bush/Perot
- Clinton/Bush/Perot

So, if a 2-candidate set wins the most votes, both candidates split the Electors evenly. If a 3-candidate set wins, all three candidates split the Electors evenly.