Electoral College

[RedDenver]

On 2/23/2018 at 4:42 PM, zoogs said:

Is that the case, though? 

 

Like, if you're going to say the President is meant to represent the states then  it seems fair enough to me to say, "OK, we have California going 60% Democrat; a sizable majority. CA is a very Democrat state, and CA is going blue. Texas is 60% Republican; it's similarly a quite red state, and it's going to go red." Each state throws in its lot, and they are weighted by state representation in federal government.

 

If you want to approximate the popular vote by proportional allocation but still keep the state weighing, it's an odd in-between. California has 65 times the population of Wyoming, and 18 times the electors. A state like California dividing its electors tilts the scales much more than a state like Wyoming. There are competing effects and consequences but I don't think we could say this is a means of approximating the popular vote better -- it seems more like a way of, on balance, slightly expanding Republican influence from the status quo in Presidential elections (because the biggest states by population, and therefore the most under-weighted states, are more blue or swing to begin than red).

I can't think of any case where the current winner takes all by state better approximates a popular vote than the popular vote by state. Wyoming having a bigger per voter influence than California is a problem for both cases, so it's not an argument for or against either as far as I can tell.

[zoogs]

Yeah, that's fair. I think I'd break it down this way. The EC has two notably odd features about it. First is that it gives substantially more weight to smaller population states. Second is that states are almost all winner-takes-all. 

 

In concert, there is at least an argument to be made for it: we aren't trying to approximate the popular vote result, we're trying to have a President of the (United) States. It is, effectively, the weighted yes/no's of each of the fifty states, and it's weighted in such a way that small states are not totally drowned out.

 

When only one of these features is changed but the other remains the same, you have a weird middle ground. As an approximation, it's not merely a poor one -- it's a purposefully inaccurate one, and inaccurate in a very specific direction. Even supposing we can neatly solve the additional complicating issue of non-fractional electors, it leaves us in a bad place. We would then have a President of the People, where Small State People count more than Big State People. To me, including the Senator count in each state's electoral count becomes far less justified.

 

But I think we both agree that the effects of both the status quo and these proposed tweaks are so convoluted, it's better to just have a popular vote. I just think that if you want to move towards a popular vote, you have to *really* have a popular vote, and lose the part that biases the vote significantly in one direction. National popular votes, as you might expect, can be reasonably close. The consistency of the bias hurts.

[RedDenver]

33 minutes ago, zoogs said:

Yeah, that's fair. I think I'd break it down this way. The EC has two notably odd features about it. First is that it gives substantially more weight to smaller population states. Second is that states are almost all winner-takes-all. 

 

In concert, there is at least an argument to be made for it: we aren't trying to approximate the popular vote result, we're trying to have a President of the (United) States. It is, effectively, the weighted yes/no's of each of the fifty states, and it's weighted in such a way that small states are not totally drowned out.

 

When only one of these features is changed but the other remains the same, you have a weird middle ground. As an approximation, it's not merely a poor one -- it's a purposefully inaccurate one, and inaccurate in a very specific direction. Even supposing we can neatly solve the additional complicating issue of non-fractional electors, it leaves us in a bad place. We would then have a President of the People, where Small State People count more than Big State People. To me, including the Senator count in each state's electoral count becomes far less justified.

 

But I think we both agree that the effects of both the status quo and these proposed tweaks are so convoluted, it's better to just have a popular vote. I just think that if you want to move towards a popular vote, you have to *really* have a popular vote, and lose the part that biases the vote significantly in one direction. National popular votes, as you might expect, can be reasonably close. The consistency of the bias hurts.

To *really* have a popular vote requires amending the Constitution, which is possible, but when you remember that any amendment requires being ratified by 3/4 of the states, you'll realize that the states benefiting from the Electoral College number more than 1/4, so that's never going to happen. The EC has been upheld many times by the courts, so there's no plausible way to get to a "real" popular vote that I can see.

 

However, the Electoral College does NOT have anything about it which results in winner-take-all states.

 

That's really important because it means that the lawsuits challenging winner-take-all are NOT a challenge to the Constitution or the EC, which have previously been challenged in court and the EC has been upheld. It means that the winner-take-all part can be overturned.

 

So what I'm saying is that getting rid of the winner-take-all is better than nothing, which is the only alternative I see.

[Moiraine]

It's not better than nothing if they do it the wrong way which we discussed, and you have to know they'll do it the wrong way. Nebraska is already doing it the wrong way. The way you want it is fine, the way Nebraska does it is not.

 

The EC is preferrable to geographies broken up by population (like the House is done).

 

With the Nebraska way it's possible for the party with the majority of the vote within a state to get less points. With the EC that is only possible on the national level. A change to Nebraska's method can lead to an even bigger difference in the outcome vs the popular vote.

 

 

Example:

 

Let's say a state has 7 districts of 10 people each.

 

The Democrat wins 3 districts with 9/10 votes. That's 27 total votes. The Republican wins 4 districts with 6/10 votes. That's 24 total votes. The Republicans get 4 out of 7 points with 47% of the vote.

 

The reverse can happen to. Anyone who wants the presidential vote to get more democratic does not want the Nebraska method to be used in more states.

Edited February 26, 2018 by Moiraine

[RedDenver]

15 minutes ago, Moiraine said:

It's not better than nothing if they do it the wrong way which we discussed, and you have to know they'll do it the wrong way. Nebraska is already doing it the wrong way. The way you want it is fine, the way Nebraska does it is not.

 

The EC is preferrable to geographies broken up by population (like the House is done).

 

With the Nebraska way it's possible for the party with the majority of the vote within a state to get less points. With the EC that is only possible on the national level. A change to Nebraska's method can lead to an even bigger difference in the outcome vs the popular vote.

 

 

Example:

 

Let's say a state has 7 districts of 10 people each.

 

The Democrat wins 3 districts with 9/10 votes. That's 27 total votes. The Republican wins 4 districts with 6/10 votes. That's 24 total votes. The Republicans get 4 out of 7 points with 47% of the vote.

 

The reverse can happen to. Anyone who wants the presidential vote to get more democratic does not want the Nebraska method to be used in more states.

I disagree. Right now one party can get 100% of the districts (aka electors) with 50.1% of the vote. And if voter suppression continues to be allowed (similar to gerrymandering being allowed in your example), then 100% of electors can go to a candidate with less than 50% of the votes.

 

In your example, the Republicans got 57% of the electors with 47% of the votes that's a difference of 10%, which is a lot less than the 49.9% difference possible in the current system (or worse if we're accounting for other factors).

[Moiraine]

10 minutes ago, RedDenver said:

I disagree. Right now one party can get 100% of the districts (aka electors) with 50.1% of the vote. And if voter suppression continues to be allowed (similar to gerrymandering being allowed in your example), then 100% of electors can go to a candidate with less than 50% of the votes.

 

In your example, the Republicans got 57% of the electors with 47% of the votes that's a difference of 10%, which is a lot less than the 49.9% difference possible in the current system (or worse if we're accounting for other factors).

 

 

 

You're giving more points to the loser of the election, and incentivizing gerrymandering more.

[RedDenver]

16 minutes ago, Moiraine said:

 

 

 

You're giving more points to the loser of the election, and incentivizing gerrymandering more.

In your own example using my method (and allowing gerrymandering), it's the difference of 1 elector (4/7 instead of 3/7 for a popular vote). Using your example with the current method, it's the difference of 5 electors (10/10 instead of 5/10). How is getting it wrong by 5 electors better than only by 1 elector? And if we got rid of gerrymandering, it'd be even closer to a real popular vote.

Edited February 26, 2018 by RedDenver

[Moiraine]

8 minutes ago, RedDenver said:

In your own example using my method (and allowing gerrymandering), it's the difference of 1 elector (4/7 instead of 3/7 for a popular vote). Using your example with the current method, it's the difference of 5 electors (10/10 instead of 5/10). How is getting it wrong by 5 electors better than only by 1 elector? And if we got rid of gerrymandering, it'd be even closer to a real popular vote.

 

 

Winner take all sounds better than loser take more than they won, to me.

 

This is a half measure though. Popular vote is what should happen.

[RedDenver]

1 minute ago, Moiraine said:

Winner take all sounds better than loser take more than they won, to me.

That's some strange logic where you'd prefer a method that is even less representative. Also notice that it's exactly the problem you're describing at the district level but raised to the state level. For example, if party A wins 90% of the vote in a state with 10 electors and party B wins 51% of the vote in a state with 11 electors, then with winner-take-all party B would win. If there's an average of 100 people per elector for 2100 voters total in this example, then party A had 900+539=1439 votes (68.5%) and lost while party B had 100+561=661 votes (31.5%) and won.

 

1 minute ago, Moiraine said:

This is a half measure though. Popular vote is what should happen.

Of course, but until someone can explain how to get 3/4 of the states to ratify a change to the EC, that's not possible. What I'm talking about is not only possible, but there are lawsuits that could make it happen all over the country relatively quickly, which also alleviates the problem of getting each state to do it despite it disadvantaging each state until all the states do it.

[zoogs]

A half measure can be a bad measure if it moves in the wrong direction. The (arbitrary) fact that our states are drawn such that we have these big blue chunk votes is a counter to the fact (what happens to be, arbitrarily) red areas are weighted much more heavily. What happens if you take away the former and retain the latter? It seems like a serious political own goal, and it doesn't even accomplish fairness.