What if electoral votes were awarded proportionally?

[brucejoel99]

• State:               GOP DEM LIB GRN
• Alabama               7      2
• Alaska                  3
• Arizona                 7      4
• Arkansas               5      1
• California              17    35    2     1
• Colorado               3       6
• Connecticut           2       5
• Delaware                       3
• Florida                  16    13
• Georgia                 10     6
• Hawaii                           4
• Idaho                     4
• Illinois                    7     13
• Indiana                  8      3
• Iowa                      5      1
• Kansas                   5      1
• Kentucky                6      2
• Louisiana                6      2
• Maine                     1      3
• Maryland                3      7
• Massachusetts        3      8
• Michigan                 9     7
• Minnesota               4     6
• Mississippi              5      1
• Missouri                 7      3
• Montana                 3
• Nebraska                4      1
• Nevada                   2      4
• New Hampshire       1      3
• New Jersey              5     9
• New Mexico             1     4
• New York                10   18    1
• North Carolina          9     6
• North Dakota           3
• Ohio                       11    4
• Orgeon                    2     5
• Pennsylvania           11    9
• Rhode Island                   4
• South Carolina          6     3
• South Dakota           3
• Tennessee                8     3
• Texas                      21   16    1
• Utah                        4     2
• Vermont                         3
• Virginia                    8     5
• Washington              4     8
• Washington, D.C.            3
• West Virginia            5
• Wisconsin                 6     4
• Wyoming                  3                   
• TOTALS                 276  257   4  1

Result: Donald Trump still elected 45th President of the United States

[Antonio the Sixth]

What allocation method are you using? I find it a bit strange that Hillary gets all 4 in Hawaii and T***p all 4 in Idaho. PR would suggest a 3-1 split.

[brucejoel99]

What allocation method are you using? I find it a bit strange that Hillary gets all 4 in Hawaii and T***p all 4 in Idaho. PR would suggest a 3-1 split.

The allocation that OP suggests: awarding 2 EVs to the pop. vote winner, w/ the remainder allocated based on the percentage of pop. vote earned.

[Antonio the Sixth]

What allocation method are you using? I find it a bit strange that Hillary gets all 4 in Hawaii and T***p all 4 in Idaho. PR would suggest a 3-1 split.

The allocation that OP suggests: awarding 2 EVs to the pop. vote winner, w/ the remainder allocated based on the percentage of pop. vote earned.

I see. That makes sense then.

[vote for pedro]

Note how swing states with an even number of CDs only net 2 EVs to the winner.  While swing states that have an odd number of EVs would be more important and net the winner 3 EVs.  Very interesting.

• State:               GOP DEM
• Arizona                 7      4 net 3
• Colorado               3       6 net 3
• Florida                  16    13 net 3
• Maine                     1      3 net 2 (same result IRL)
• Michigan                 9     7 net 2
• Minnesota               4     6 net 2
• Nevada                   2      4 net 2
• New Hampshire       1      3 net 2
• North Carolina          9     6 net 3
• Pennsylvania           11    9 net 2
• Virginia                    8     5 close enough to act like a swing state net 3
• Wisconsin                 6     4 net 2

Other than the odd/even deal it makes all the swing states equally important.

[dercook]

Would be rare to see landslide victories, but after all, it isn't like we're getting much of those these days.

[IceAgeComing]

The thing with even number districts behaving differently is known in PR: since with an even number of seats any election near 50/50 (assuming a two party race here, which isn't exactly the case in any country that uses PR outside of Malta and they use STV which is different although the same thing does tend to happen) the seats split equally while with an odd number of seats someone must come out with the most.

What it probably does is actually empower bigger states more than a full national popular vote would do: since .  For example using D'Hondt PR (and not factoring in the majority bonus in this; the same principle applies though) in order to swing an electoral vote in a three vote state you need to swing a hell of a lot of votes (in a straight two-candidate race the quota to get an electoral vote in a three vote state would be 25%; so outside of DC they'd all split 2/1 unless one candidate had a total landslide and got over 75% of the vote) while in California you'd need 1.8% of the vote to get a seat; so its much easier to pick off a few seats with a relatively small swing of the vote): so in small states that are 60% for one candidate why would the parties bother campaigning since they know what the distribution is going to be?

Alexander Hamilton's version of the 12th Amendment would have provided that presidential electors be elected by electoral district drawn by Congress (e.g. a state with one representative would have three electoral districts) and that the mode of election be specified by Congress, and that the electors designate presidential and vice presidential votes.

Hamilton's proposed 12th Amendment

Had this been adopted, there wouldn't be any questions about national popular vote, since nobody would be adding up the votes, anymore than they total the national popular vote for Congress.

A modern version would provide that:

Electors be apportioned among the United States and their territories based on the Citizen Population over age 18. An elector would represent between 20,000 and 50,000 persons.

Electors be chosen by the voters eligible to vote for the larger house of a legislature, with time, manner, place regulations set by the legislature, subject to a congressional override (e.g. same rules as apply to the election of Congress).

Electors would meet as a single national body, perhaps electronically linked; and would determine a president and vice president by majority vote. If no candidate received a majority on the initial vote, voting would continue by rounds among the (up to Top 10), with one eliminated on each round.

Here's the thing: this would basically mean that America would run Presidential elections in the same way that countries - hell; in some respects this is exactly the way that most European countries run with respect of picking their head of government, except that they are electing their parliament while you're elected a nonsense chamber full of thousands of people. 

[MillennialModerate]

Isn't this basically like simplifying the popular vote?

[Anzeigenhauptmeister]

When I did this I got:
Clinton 269
Trump 265
Johnson 2
Stein 1
McMullin 1

Using the D'hondt method and this calculator with no thresholds.

I wonder what the Stein elector would have done...

[Free Bird]

Isn't this basically like simplifying the popular vote?

No it wouldn't.

Also, I found this calculator on 270toWin

https://www.270towin.com/alternative-electoral-college-allocation-methods/

276 Trump
257 Clinton
1 McMullin
3 Johnson
1 Stein

[StateBoiler]

Done this for some elections. There's different ways to do it. For example,

(not touched the numbers in awhile, so may be off 1 or 2 in some spots, this may not have the final final vote count)

Proportional by State (Highest Remainder Method)

Trump 261
Clinton 260
Johnson 15
Stein 1
McMullin 1

if you say a person has to reach 1 full candidate to receive 1 (i.e. to receive 1 delegate in a state with 10 votes, must reach 10%):

Trump 267
Clinton 267
Johnson 3
McMullin 1

if you say whoever receives most votes in a state must receive at least 1 more delegate than 2nd:

Clinton 262
Trump 257
Johnson 17
Stein 1
McMullin 1

then say that category must reach 1 full delegate to receive 1:

Clinton 269
Trump 265
Johnson 3
McMullin 1

if you want to do the D'Hondt Highest Averages method:

Clinton 268
Trump 266
Johnson 2
Stein and McMullin 1 each

if you want to do say convention-style delegate rules (receive a majority get all, if no one gets a majority greater than some percentage receives proportional, 5% limit below):

Trump 284
Clinton 251
Johnson 2
McMullin 1

I have 12 different methods in a spreadsheet going back to 1992, and in all of them there's no way to get Clinton to 270 in 2016. The only way Gore in 2000 gets to 270 is under convention-style delegate apportionment.

The current Electoral College is the only method that produces a winner in 1992.

The elections I really want to do are 1980, 1968, 1948, and 1912. 1860 would be fun in theory.

Minor parties in proportional state allocations would really focus on the big states. To get a vote in California you need a little less than 2% for instance. It would also drive a more national campaign for the two main parties.

[Progressive Pessimist]

When I did this I got:
Clinton 269
Drumpf 265
Johnson 2
Stein 1
McMullin 1

Using the D'hondt method and this calculator with no thresholds.

I wonder what the Stein elector would have done...

Probably not the right thing.

[Former President tack50]

What if instead of winner-take-all, candidates won electoral votes in each state based on the percentage of voters that voted for them?

The winner of each state would get two electoral votes, which would be the at-large votes, and the rest is based on proportion.

How would you handle Rhode Island with 4 EV?  The clear victor at 55% has "about half" of the popular vote.  Who gets the 4th EV?

I think it should go to the victor; unless an argument could be made that someone else won one of the two Congressional districts.  Then you would be describing how Nebraska and Maine handle it.

Most likely the D'Hondt rule (or something similar) is used. So for Rhode Island you'd get a 2-2 split.

Of course the main disadvantage of this is that 4 EV states would essencially become useless as they would almost always have a 2-2 split unless there's a massive landslide (you'd need a 60-40 split to get a 3-1 EV split) or a third party candidate

[StateBoiler]

What if instead of winner-take-all, candidates won electoral votes in each state based on the percentage of voters that voted for them?

The winner of each state would get two electoral votes, which would be the at-large votes, and the rest is based on proportion.

How would you handle Rhode Island with 4 EV?  The clear victor at 55% has "about half" of the popular vote.  Who gets the 4th EV?

I think it should go to the victor; unless an argument could be made that someone else won one of the two Congressional districts.  Then you would be describing how Nebraska and Maine handle it.

Most likely the D'Hondt rule (or something similar) is used. So for Rhode Island you'd get a 2-2 split.

Of course the main disadvantage of this is that 4 EV states would essencially become useless as they would almost always have a 2-2 split unless there's a massive landslide (you'd need a 60-40 split to get a 3-1 EV split) or a third party candidate

You could add a qualifier of the candidate that wins a state must receive 1 more delegate than the person that gets 2nd. For example:

You must be logged in to read this quote.

It does create situations like this however occasionally. This is the most egregious example of the elections I have done:

You must be logged in to read this quote.

If you add a secondary qualifier of "to receive 1 electoral vote you must receive 1 whole electoral vote", it gets rid of this example as for Utah to receive 1 vote you'd have to reach 20%. That rule however does lead to 3 electoral vote states sometimes where whichever party has completely given up on the state, they're under 33.3% of the vote and the winning candidate sweeps the state 3-0.

[StateBoiler]

So updated my 2016 numbers yesterday to match Dave Leip's. Here's how every election I've done looks.

National Level Numbers

Method 1: Purely Proportional (Highest Remainder Method)
Method 2: Must Receive 1 Full Electoral Vote in Method 1 in order to Receive 1
Method 3: D'Hondt National (Highest Average Method)

State Level Numbers

Method 4: Proportional (Highest Remainder Method)
Method 5: Must Receive 1 Full Electoral Vote in Method 4 in order to Receive 1
Method 6: Proportional, Winner Must Receive at least 1 more Electoral Vote than 2nd
Method 7: Must Receive 1 Full Electoral Vote in Method 6 in order to Receive 1
Method 8: D'Hondt State (Highest Average Method)

Convention-style Allocation

Method 9: Receive a majority of a state's vote, you get all Electoral Votes; if no one receives a majority, everyone above a 15% cutoff line receives proportional allocation
Method 10: Same as Method 9 except it's a 10% cutoff line
Method 11: Same as Method 9 except it's a 5% cutoff line

Method 12: Current Electoral College

2016:

You must be logged in to read this quote.

2012:

You must be logged in to read this quote.

2008:

You must be logged in to read this quote.

2004:

You must be logged in to read this quote.

2000:

You must be logged in to read this quote.

1996:

You must be logged in to read this quote.

1992:

You must be logged in to read this quote.

1980:

You must be logged in to read this quote.

My goal is to do every post-Civil War election and then write up something formal about this. If reform were to ever occur, it'd have to be done at a national level. But based on what I've done so far, I would be in favor of one of Methods 7 thru 11. I think 9 thru 11 would receive better reception since it'd be easy to explain to the vast majority of people that don't understand how math works.

[QAnonKelly]

Wouldn't this just hurt big blue and heavily minority states?

[Dr. MB]

Wouldn't this just hurt big blue and heavily minority states?

Some of the heaviest-minority states are in the deep south, which (with the possible exception of Georgia) doesn't look to be going Democratic anytime soon.

Also must mention Texas. It's essentially a tradeoff: the Dems give up some CA, NY, IL, etc. electoral votes in exchange for votes in Texas and the rest of the south.

[defe07]

I would support a proportional system where each state would allocate EV with the Hare Quota and any unfilled seats would be allocated proportionally nationwide using remainder votes and surplus votes from the states.

[Fuzzy Bear]

It's time for the popular vote to determine the President.  Period.

[muon2]

It's time for the popular vote to determine the President.  Period.

I still contend that it should require a runoff if there is no majority, as France does for its president. A plurality president would not be advisable when the executive has such strong powers. We often overlook that since the two party system is so ingrained. But there's nothing that prevents the US from having more than two strong parties (perhaps except the EC).

Parliamentary democracies don't have to worry about that since a government needs over 50% of the members to form. If there is no one majority party then the formation of a coalition government functions as a runoff. That's a built-in check on the power of a plurality executive.

[Fuzzy Bear]

It's time for the popular vote to determine the President.  Period.

I still contend that it should require a runoff if there is no majority, as France does for its president. A plurality president would not be advisable when the executive has such strong powers. We often overlook that since the two party system is so ingrained. But there's nothing that prevents the US from having more than two strong parties (perhaps except the EC).

Parliamentary democracies don't have to worry about that since a government needs over 50% of the members to form. If there is no one majority party then the formation of a coalition government functions as a runoff. That's a built-in check on the power of a plurality executive.

I don't know if I would agree to this, but I would say that if the top two candidates get over 90% combined, there should not be a runoff.

[muon2]

It's time for the popular vote to determine the President.  Period.

I still contend that it should require a runoff if there is no majority, as France does for its president. A plurality president would not be advisable when the executive has such strong powers. We often overlook that since the two party system is so ingrained. But there's nothing that prevents the US from having more than two strong parties (perhaps except the EC).

Parliamentary democracies don't have to worry about that since a government needs over 50% of the members to form. If there is no one majority party then the formation of a coalition government functions as a runoff. That's a built-in check on the power of a plurality executive.

I don't know if I would agree to this, but I would say that if the top two candidates get over 90% combined, there should not be a runoff.

Should that still be true if the candidates were statistically tied with neither over 50%. By statistically tied I mean that given the limits of our accuracy of vote counting the winner can only be determined by arbitrary means such as stopping the count on some particular day. The Franken - Coleman race in 2008 was such a statistical tie as was the presidential vote in FL in 2000.

[NewFederalist]

Actually, any state can alter the way they select their presidential electors. There is nothing to prevent them from being selected by the legislature (as was the case with South Carolina until 1860). Any form of proportional division or by congressional district or whatever. The National Popular Vote plan which has now passed 12 state legislatures is another. As long as civil rights laws are not in conflict with whatever method it appears this is one of the last areas where states' rights are not usurped by the US government.

[morgankingsley]

I remember doing this for 1984 and Mondale would end up with 206 electoral votes

[morgankingsley]

I did two of them, 1924 and 1984.


Here is the one for 1924