Posts by BenWilson
Last ←Newer Page 1 2 3 4 5 Older→ First
-
Hard News: The Big 2012 US Election PAS Thread, in reply to David Hood,
Is that considered a riot or just a party, in Mississippi?
-
Legal Beagle: MMP Review #1: The Party…, in reply to Brent Jackson,
They're sticking with the proposed 4% threshold, even though the simulation data showed 3% to be the better option.
I'm honestly surprised that a simulation would be necessary to show that lower thresholds generate higher proportionality. It's so damned obvious that increasing thresholds increases disproportionality*. I guess it's nice to have a number that measures how much exactly between %3 and %4 but is it really a meaningful number? How do you trade that value off against other values in the system? Perhaps if they'd also modeled "stability", that old chestnut that prevents low thresholds, then we'd at least see how much we value proportionality compared to stability. All we can do with a disproportionality mean is to compare it to other countries.
I expect the reason this isn't done is because stability is very hard to define. Indeed even considering it is part of hegemonic discourse - of course a lot of minorities would like a certain kind of instability. They would like their problems addressed, which means change, which is less stable. The very introduction of instability that came from MMP was a big part of the point of it, that it broke up existing power structures.
*That said, they claimed that the simulations of thresholds less than 2% were less proportionate. But they didn't give the data. I wish they had, that would be very interesting. It could be such a tiny difference as to be insignificant, with some logical twist in the St Laguё kicking in on the data set. I'm struggling to see how it's possible that it's less proportional to include all the parties that would be cut off by a 2% threshold. Presumably, they're either ignored, or there's some kind of granularity issue that happens with the number of party list seats available. Or the simplistic model they had for preference switching just didn't have data for the really small parties in sufficient volume. Or it could just be too simplistic a model. They said "as expected" so perhaps there is some well known theorem that shows proportionality has a maximum average point for some threshold that is greater than zero (which is where I'd expect the maximum average to be).
Interesting way of modeling, too.
-
Legal Beagle: 14 Pages of Democracy, in reply to Graeme Edgeler,
I thought I'd bump this, given how sad I'm feeling about things like lines we're hearing about in the US election :-(
I distinctly heard Obama say "Yeah, we gotta do something about that" in his victory speech.
Not sure what he can do - it's not really something in the President's hands?
-
Hard News: The Big 2012 US Election PAS Thread, in reply to Joe Wylie,
My expectations did drop, but I think that was because they were unrealistic in the first place. He has accomplished things that were important, mostly domestic. US foreign policy stance is a huge one to tackle, it has plans that span decades. To even halt certain kinds of insanity, like full scale aggressive war, is to achieve a lot. Iran is probably safe for another 4 years.
-
Hard News: The Big 2012 US Election PAS Thread, in reply to David Hood,
and that is all the full fields I'm willing to invest time in looking up on this)
Aw...there could be real money to be made here.
-
Hard News: The Big 2012 US Election PAS Thread, in reply to David Hood,
I just checked and there is (or at least was) only one horse starting with T
This time. Was that the case for the past data though? :-)
-
Hard News: The Watching World, in reply to Lucy Stewart,
Indeed. Crappy pens too, but fixing them only requires getting another one out of a box.
-
Hard News: The Big 2012 US Election PAS Thread, in reply to David Hood,
What if there's three in there with a T first? I think you need to do second letter analysis too.
ETA: You should probably also indicate how many of each initial has occurred in total amongst the non-winners. I bet T is very common, Z very uncommon.
-
Hard News: The Watching World, in reply to Russell Brown,
Meanwhile, video of a voting machine that keeps flipping votes from Obama to Romney.
It's a pity they didn't show attempting to get Obama via Jill Stein, to prove it's not just a screen calibration issue. The guy just talked about it, which is much less believable. One is a simple fault, the other highly malicious, illegal intent.
-
Legal Beagle: MMP Review #1: The Party…, in reply to Rich of Observationz,
It's possible that if Labour's decline isn't arrested, this will work in the "left"'s favour as Labour holds electorates but loses its party votes to the Greens.
Yup, but they'd have to lose a lot more party vote to start generating an overhang. On 22 electorate seats, they'd need to get less than 22/120 = 18.3% before it would work. If all that party vote went to the Greens, they'd be the bigger party.
Yes, gifting electorates works too. It's not exactly the same strategy I was talking about, but it can generate overhangs, if those parties get no party vote "stolen" from their partner party.
Last ←Newer Page 1 … 407 408 409 410 411 … 1066 Older→ First