It's not that Democrats get more than proportional, it's that any majority party gets more than proportional. I understand that it's a goal that one might want. But this goal does not arise naturally in a single-member district system.
The natural outcome depends on how variable voters are from district to district. The more variable they are, the closer one gets to the ideal slope of 1, which you are basically suggesting. But historically (and this includes other countries as well) the slope is more like 2 or 3.
One could conceivably force such an outcome from artful drawing of districts. But this was not the point of the graph. My goal was to show that this year we have a normal and majoritarian outcome.
It's not that Democrats get more than proportional, it's that any majority party gets more than proportional. I understand that it's a goal that one might want. But this goal does not arise naturally in a single-member district system.
The natural outcome depends on how variable voters are from district to district. The more variable they are, the closer one gets to the ideal slope of 1, which you are basically suggesting. But historically (and this includes other countries as well) the slope is more like 2 or 3.
One could conceivably force such an outcome from artful drawing of districts. But this was not the point of the graph. My goal was to show that this year we have a normal and majoritarian outcome.
There is a more mathematical way to prove all of this. It is hinted at in a diagram and in footnote 100 of my article from sometime back in SLR: http://www.stanfordlawreview.org/wp-content/uploads/sites/3/2016/06/3_-_Wang_-_Stan._L._Rev.pdf#page=24
For a thread getting just a little more into the math, see this: https://twitter.com/Labrigger/status/1592578806770176000
Also excellent work by Gudgin and Taylor way back when.
thanks!