The proportional Electoral College

Politics & Elections
1

Oft discussed threads in this forum ponder alternative voting systems and their effect on the success of third/minor parties.

So for the sake of the exercise I have calculated what the EC might look like were the delegates allocated in proportion to votes cast rather than winner takes all. If you like, apply the Maine logic to all states.

The result in 2016 is close, very close and therefore the precise manner that is used to determine how quotas are calculated and how surplus votes are reallocated or exhausted is crucial and likely vary the result.

However, FWIW my calculations for Election 2016 show a hung EC!

Clinton 268
Johnson 4
Stein 1
Trump 265
Full state results are below:

Alabama
Clinton 3
Johnson 0
Stein 0
Trump 6

Alaska
Clinton 1
Johnson 0
Stein 0
Trump 2

Arizona
Clinton 5
Johnson 0
Stein 0
Trump 6

Arkansas
Clinton 2
Johnson 0
Stein 0
Trump 4

California
Clinton 34
Johnson 2
Stein 1
Trump 18

Colorado
Clinton 5
Johnson 0
Stein 0
Trump 4

Connecticut
Clinton 4
Johnson 0
Stein 0
Trump 3

Delaware
Clinton 2
Johnson 0
Stein 0
Trump 1

Dist. Of Columbia
Clinton 3
Johnson 0
Stein 0
Trump 0

Florida
Clinton 14
Johnson 0
Stein 0
Trump 15

Georgia
Clinton 8
Johnson 0
Stein 0
Trump 8

Hawaii
Clinton 3
Johnson 0
Stein 0
Trump 1

Idaho
Clinton 1
Johnson 0
Stein 0
Trump 3

Illinois
Clinton 11
Johnson 1
Stein 0
Trump 8

Indiana
Clinton 4
Johnson 0
Stein 0
Trump 7

Iowa
Clinton 3
Johnson 0
Stein 0
Trump 3

Kansas
Clinton 2
Johnson 0
Stein 0
Trump 4

Kentucky
Clinton 3
Johnson 0
Stein 0
Trump 5

Louisiana
Clinton 3
Johnson 0
Stein 0
Trump 5

Maine
Clinton 3
Johnson 0
Stein 0
Trump 1

Maryland
Clinton 6
Johnson 0
Stein 0
Trump 4

Massachusetts
Clinton 7
Johnson 0
Stein 0
Trump 4

Michigan
Clinton 8
Johnson 0
Stein 0
Trump 8

Minnesota
Clinton 5
Johnson 0
Stein 0
Trump 5

Mississippi
Clinton 2
Johnson 0
Stein 0
Trump 4

Missouri
Clinton 4
Johnson 0
Stein 0
Trump 6

Montana
Clinton 1
Johnson 0
Stein 0
Trump 2

Nebraska
Clinton 2
Johnson 0
Stein 0
Trump 3

Nevada
Clinton 3
Johnson 0
Stein 0
Trump 3

New Hampshire
Clinton 2
Johnson 0
Stein 0
Trump 2

New Jersey
Clinton 8
Johnson 0
Stein 0
Trump 6

New Mexico
Clinton 3
Johnson 0
Stein 0
Trump 2

New York
Clinton 18
Johnson 0
Stein 0
Trump 11

North Carolina
Clinton 7
Johnson 0
Stein 0
Trump 8

North Dakota
Clinton 1
Johnson 0
Stein 0
Trump 2

Ohio
Clinton 8
Johnson 0
Stein 0
Trump 10

Oklahoma
Clinton 2
Johnson 0
Stein 0
Trump 5

Oregon
Clinton 4
Johnson 0
Stein 0
Trump 3

Pennsylvania
Clinton 10
Johnson 0
Stein 0
Trump 10

Rhode Island
Clinton 2
Johnson 0
Stein 0
Trump 2

South Carolina
Clinton 4
Johnson 0
Stein 0
Trump 5

South Dakota
Clinton 1
Johnson 0
Stein 0
Trump 2

Tennessee
Clinton 4
Johnson 0
Stein 0
Trump 7

Texas
Clinton 17
Johnson 1
Stein 0
Trump 20

Utah
Clinton 2
Johnson 0
Stein 0
Trump 4

Vermont
Clinton 2
Johnson 0
Stein 0
Trump 1

Virginia
Clinton 7
Johnson 0
Stein 0
Trump 6

Washington
Clinton 7
Johnson 0
Stein 0
Trump 5

West Virginia
Clinton 1
Johnson 0
Stein 0
Trump 4

Wisconsin
Clinton 5
Johnson 0
Stein 0
Trump 5

Wyoming
Clinton 1
Johnson 0
Stein 0
Trump 2

2

You have Georgia split 8-8. Clinton’s share of Georgia’s vote entitled her to 7.34 delegates, not eight, so a fairer apportionment would be 9-7. It’s hard to work out, I guess, because the fractions don’t perfectly add up to 16.

3

Nebraska does use the same system as Maine and it went 5-0 for Trump.

4

This is fascinating, thank you for your work. I’ll be saving a link to this thread for future reference.

But as a nitpick: Maine and Nebraska don’t assign EC votes proportionately. They assign them by congressional district, 1 for the winner in each of them plus 2 to the statewide winner, tracking how EC votes are allocated to each state in the first place.

5

Yes, it’s relatively silly to talk about “proportional” EC voting using strictly proportional methods. That’s tantamount to using the popular vote, with rounding errors. :smiley:

6

I haven’t been able to find any breakdown of the presidential results by congressional district for this year. Back in 2012, the Daily Kos crunched the numbers and posted the results about two weeks after election day.

7

Any analysis like this is flawed because the candidates would have campaigned differently under a different system. You can’t assume the vote totals by state would have been the same.

8

The analysis is fine, based on it’s assumptions.

Certainly if that was the way the EC was determined the candidates would (or at least should) refine their campaign strategies accordingly.

9

Noted, I didn’t have the data available to do that at the time.
On a strictly proportional vote Maine split 2:2 but I overrode my calcs to show the declared result … but neglected to do the same for Nebraska. [smeg] Sorry.

10

I’m always amazed how seldom this basic point is made in debates about the EC, I don’t think I’ve ever heard it in countless TV talking head debates about the EC.

On OP, it’s interesting math, but that’s not the Maine and Nebraska system where in both cases the 2 EV’s ‘representing’ the Senators go to the winner at large in the state, and EV’s representing the House districts go to the winner of that district. Trump won the ME 2nd Congressional district so got 1 EV, Clinton the other CD by enough more to win the state at large and got 3; a 2-2 split is impossible under the current system in ME. Trump won all three CD’s in NE so got all 5 EV’s. Johnson and Stein didn’t win any CD’s in the country so still would have been shut out if the ME/NE system was applied nationally. I couldn’t find a source for 2016 presidential result by CD nationally, yet.

Obviously there’s a political problem with additional states going to the ME/NE system if they don’t do it at once. Say CA did, then it would go from 55 essentially automatic EV’s for the Democrats to perhaps 14 for the GOP and 41 for the Democrats if the presidential race went the same way as House races. And let’s face it, the idea of ‘reform’ of the electoral system is usually to increase one party’s chances of winning, and the other resists it.

11

So if we used this system we’d have a hung Electoral College. What now? Do you think we should next have a runoff election (hey, what’s another month or so of elections?), automatic recount, switch to plurality win?

12

Nothing to do with fairness. :slight_smile:
I think your calculation of an EC quota is incorrect.

I have used the AEC formula:
(Number of formal ballots / (Number of Reps to be elected + 1)) rounded down + 1) = Rep quota

Votes

Trump 2,068,623
Clinton 1,837,300
Johnson 123,641
4,029,564

Making the Rep quota is 237,034

Therefore Trump wins 8.727 quotas, Clinton won 7.751 and Johnson won 0.522
If votes exhaust then Clinton wins the last spot.
If they don’t then you need a preference split on Johnson’s vote which is problematic.
Johnsons voters would need to split 53:47 for Trump to get the last spot.

13

The US has an existing defined process in place for that eventuality.
Why wouldn’t that be followed?

14

Well you’re proposing something that’s not being followed now so I don’t think I asked an unreasonable question. But to expand, the current system very rarely has to refer the Presidential selection to the House. A proportional EC would almost certainly send more to the House for the final decision. So would you be happy with that or would you not want to also add some other “close vote” procedures?

15

Proportional would consistently produce closer EC votes than winner take all.
I don’t think it necessarily follows that one produces results requiring a tiebreaker in greater frequency than the other.
I’m happy with the current tiebreaker process. I have no thoughts on any “improved” model.

16

It quite obviously would. It’s half the point of FPTP and winner take all systems to produce a more stark win.

17

I tend to myself prefer a “thresholded” plurality system in direct popular vote elections (most Fed and State elected officials save the Prez), where the win goes to the plurality ***if ***it’s above a certain threshold and outside a certain margin of difference; trigger the runoff if those conditions fail to be met. This allows you to still have “clear winners” most often, but leaves the runoff choice open in case you end up with a dismal 30.5%/30%/20%/17.5% type outcome.

In the EC I would not support widespread adoption of the Maine/Nebraska system of allocation in larger or more diverse states, as it is vulnerable to, and creates added incentive for, gerrymandering. I prefer Electors should be chosen at-large, they are to speak for the state, not for constituencies. If we *were *to get rid of EC WTA to a large extent, then an alternative to a truly proportional allocation, to reduce the chance of hung colleges, could be a *weighted *division that rewards the frontrunner if (again) certain criteria for thresholds and margins are met – sort of a “Winner Take Most” concept.

18

The ME/NE system clearly favors Republicans currently because of gerrymandering. I don’t think there’s a breakdown for this year yet, but I found one for 2012. Obama won in 210 districts vs. Romney’s 225. Throwing in 2 per state + 3 more for DC, gives a final result of Obama 265, Romney 273.

I 6 districts the difference was less than half a percent, so that would have been a big issue with recounts.

19

Again, not necessarily.
You are, understandably basing that view on the current electoral map where the feature is a few large population Blue states and a larger number of smaller population Red states.

But amend that paradigm in a future scenario where say California goes Red and the South goes Blue.

Under that scenario you could have winner-takes-all delivering either landslides or hung EC results needing the tiebreaker while proportion delivers on average closer but still clear EC results.

20

No, I am most certainly not basing my view on the current electoral map. I am basing it on math and the actual history of other nations using proportional type systems.

As you admit, it would inevitably bring closer elections. It’s nonsense to say it would do that but not also bring a rise in deadlocks. Yes “not necessarily!” but it probably would.

When you change the rules this way, it encourages third party candidates and voting. In fact, that’s often a prime driver/hope for those advocating proportional voting systems. Thus you would have a more and more third party/independent leaking away votes. That makes it even more likely no one will get to 270.