Wittenberg

Mic'haglh Autofil and Sir Marcel Tafial are proud to announce l'Institüt del Agroïcismeu Political (the Institute for Political Nerdery). What we want to do here is basically "nerd out" about a subject we enjoy -- the various ways to analyze elections, look at different systems of voting, etc. More serious atmospheres might call this psephology (https://en.wikipedia.org/wiki/Psephology), but we're pretty sure the Talossan name is actually easier to pronounce!

To start us off, I wanted to talk about a relatively basic measurement: something called the Gallagher index (https://en.wikipedia.org/wiki/Gallagher_index). The Gallagher Index is a measurement of how disproportionate an election's results are: the greater the difference between parties' share of the vote and their share of seats, the higher the index will be.

How do you calculate it? It's pretty simple, really:

• For each party in an election, take the difference between their percent of the vote, and their percent of seats in the legislature -- in other words, (V - S).
• For each of those differences, square them.
• For all of the squared difference, add them together.
• Take that sum and divide by two.
• Find the square root of that last step's result. That square root is the Gallagher Index.

As an example, let's look at the results from our most recent election:

Party	% Vote	Seats	% Seats	Diff.	Sq. Diff.
PA	43.62	87	43.50	0.12	0.01440
URL	32.98	66	33.00	-0.02	0.00040
Green	13.83	28	14.00	-0.17	0.02890
IDT	7.45	15	7.50	-0.05	0.00250
IG/A-S	2.13	2	2.00	0.13	0.01690
				TOTAL	0.06310
				TOTAL/2	0.03155
				GI	0.17762

That's pretty low! For reference, the lowest one we can find for macronations is Guyana, which as of 2020 had a Gallagher Index of 0.52.

However, we've seen people discussing changing the size of the Cosa recently, so it's not a bad idea to look at how that could affect things. For example, if the Cosa had 20 seats as recently proposed:

Party	% Vote	Seats	% Seats	Diff.	Sq. Diff.
PA	43.62	9	45.00	-1.38	1.90440
URL	32.98	7	35.00	-2.02	4.08040
Green	13.83	3	15.00	-1.17	1.36890
IDT	7.45	1	5.00	2.45	6.00250
IG/A-S	2.13	0	0.00	2.13	4.53690
				TOTAL	17.89310
				TOTAL/2	8.94655
				GI	2.99108

2.99 is still very low -- that would put Talossa between New Zealand (2.63) and Austria (3.21). Using the same method, a 40-seat Cosa, while lacking historical precedent, would have an index of 1.22. All of these would be perfectly acceptable numbers with respect to proportionality! Any of them are in line with some of the nations considered to have extremely healthy democracies. (For reference, nations like the US, UK, Canada, etc are regularly above 5 or even 10, with the UK's 2024 election delivering a Gallagher Index of 23.73!)

It's worth looking at the Senate as well, just for curiosity, using the numbers from the four Senate races in the most recent election:

Party	% Vote (1st Rd.)	Seats	% Seats	Diff.	Sq. Diff.
PA	40.43	3	75.00	-34.57	1195.08490
URL	36.17	1	25.00	11.17	124.76890
Green	4.26	0	0.00	4.26	18.14760
Ind.	19.15	0	0.00	19.15	366.72250
				TOTAL	1704.72390
				TOTAL/2	852.36195
				GI	29.19524

As we can see, the Senate is significantly more disproportionate than even a 20-seat Cosa. This calls into question the motives and reasoning of those who may support keeping the Senate as-is, while opposing decreasing the size of the Cosa as "less representative".

Tbf, the point of the Senäts in the minds of its defenders is precisely to be countermajoritarian. It is supposed to *not* represent the majority, because "stopping the majority doing things" is seen as a benefit in itself.

Quote from: Mic'haglh Autófil, O.Be on Yesterday at 09:45:17 PMIt's worth looking at the Senate as well, just for curiosity, using the numbers from the four Senate races in the most recent election:

Party	% Vote (1st Rd.)	Seats	% Seats	Diff.	Sq. Diff.
PA	40.43	3	75.00	-34.57	1195.08490
URL	36.17	1	25.00	11.17	124.76890
Green	4.26	0	0.00	4.26	18.14760
Ind.	19.15	0	0.00	19.15	366.72250
				TOTAL	1704.72390
				TOTAL/2	852.36195
				GI	29.19524

As we can see, the Senate is significantly more disproportionate than even a 20-seat Cosa. This calls into question the motives and reasoning of those who may support keeping the Senate as-is, while opposing decreasing the size of the Cosa as "less representative".

I don't actually have any views on the size of the Cosa, but this insinuation doesn't follow. One could desire one house to be popularly representative and another wholly unrepresentative, or representative of a different set of bodies; therefore, one would object to the reduction in proportionality in the popular house, even as one objected to an increase in proportionality in the other house. In fact, this was the mainstream Anglophone theory into the late 20th century, pretty much!