Saturday, June 2, 2018

i'm generally pretty cynical about politicians passing prisoner's dilemmas, but wynne & horwath have as good a chance as any i've seen.
it's not clear at this point whether voting for the liberals will split the vote or even if voting for the ndp will, in the riding you're in.

this is very much the nightmare scenario that critics of strategic voting tend to point to. there is some data suggesting that the ndp are way ahead, but there's data suggesting otherwise as well, and it's fluctuating fairly frequently. we have imperfect methods with small sample sizes and little consistency.

i was analyzing one poll.

and, i think wynne's announcement was actually premature.

my request to horwath & the ndp still stands - be careful. i suppose that also applies to the liberals, but it's a different context. and, to the rest of the electorate, beware: if you want to play this game this year, there's a chance you're going to get burned.

try to find riding polls; do the best research you can, and prepare for the possibility that you might be wrong.

but, keep this in mind, as well - in general, the strategy promoted by critics of strategic voting is to vote with your heart.

there's still four days of polling left to come in.
so, what are we going to do about these tariffs?

well, i guess we'll have to find new markets and new importers, right? perhaps we could take advantage of that ceta that came into force recently. and, i'd still like to see a bilateral agreement with japan.

if america doesn't want to trade with the rest of the world, we can't force them to. but, the rest of the world can increase trade with itself.
and, again: the media seems to think, across the board, that having your votes distributed in the least populous areas makes your vote more efficient.

that's ridiculous.

the guy's own numbers say the ndp are running over 50% in toronto & in southwestern ontario, which together comprise roughly 60 seats. the conservatives may be running higher in more subregions, but these are the most sparsely populated areas and have the least seats.

it's maybe fitting that doug ford's people don't understand what the word 'efficiency' means, but you'd think the pollsters would get it right, even if the media doesn't.

those numbers suggest the most efficient vote is with the ndp.
but, is the idea of toronto flipping clean from red to orange implausible? we're talking 40 seats moving the same way all at once, here. is that kind of mass groupthink realistic?

it might be more expected in quebec. but, we're sibling provinces, upper and lower; it's just a language difference.

and, i think you want to think it's going to happen all at once or not at all - that there's a tipping point where it flips. certainly, wynne has been going out of her way to alienate torontonians, hasn't she?

i've been talking about the liberal base being hard to move, but i've also been talking about it being pragmatic and less tied to ideology. i just didn't want to make that choice for it without seeing any actual data. and, it hasn't moved in a very long time, either. but, if toronto has decided that it can't let ford win, and horwath is the way to do it, it's not so crazy to think the whole city could flip together at the same time, at all.

unprecedented, perhaps. historic. but not crazy.

that would leave wynne with a few seats hanging around outside the 905, and a few in ottawa.

i look forward to seeing more data framed like this. but, again, regardless of what frank says, the ndp should be happy about those numbers.
ok, this has the kind of sampling frames i wanted to see.

the first three frames are not unexpected - and it may even be a bit of good news for the ndp, at the expense of the conservatives. those are powerful numbers for the ndp in ridings they probably weren't expecting to win. they may be benefiting massively from strategic voting, and that's good - nobody thought the liberals could compete here.

i'm going to presume that the grand river valley means the kitchener-waterloo-guelph tricity area, which is an education & technology hub. it's hard to know exactly what that means, but i don't think there's a big surprise on the three-way split, either. if that's right, the liberals could maybe hold some seats there - but it's just a handful of them in play. nor have the liberals been competitive in niagara or hamilton proper for a very long time, either, although 4% seems kind of obscene. and, it's hard to know where niagara starts and the "grand river valley" ends - which is important given the difference between 30 and 4. i'm suspecting small sample sizes, here. nonetheless, if those numbers are even close to right, don't be surprised if the liberals hold some seats on the end of the horseshoe, there. but, none of this is really that earth-shattering - excluding that 4%, which may end up more like 14% in the end. the shape is not surprising.

but, those numbers in halton/peel are hard to comprehend. the races in halton are traditionally conservative-liberal. it's a wealthy suburb of toronto - the kind of place the ndp just don't do well in. peel, on the other hand, is usually an ndp-liberal race. again: combining these together is tricky. it could mean that the conservatives are running away with halton and the ndp are running away with peel, or it could be obscuring closer races in aggregate. what i'm going to say is that this is weird - and that i don't like the way it's presented, as it suggests you're looking at the pcs v the ndp over a large region, when you're probably actually looking at each of them v the liberals separately in two distinct regions - and it's not clear how close it is. frank, man what are you up to?



http://www.ekospolitics.com/wp-content/uploads/full_report_june_1_2018b.pdf

i don't have any particular rejection of the brampton numbers, but it's only 3-4 seats. brampton is the kind of demographic that would swing from red to blue under ford. and, those numbers are not particularly bad for the liberals in york/durham, either, where there are a couple of conservative strongholds and a couple of conservative/liberal swing ridings.

at this point, after the first eight frames, this doesn't look like a weird election, and the seat count doesn't look likely to flip much. what's different here is the next two frames...

i accidentally cut the bottom off, but the next two frames are central and suburban toronto and those are powerful numbers that are suggestive of an ndp majority. i don't know what frank is smoking, suggesting otherwise. if the ndp can poll close to 50% over toronto in the end, that's suggestive of a clean sweep - which is 35-40 seats. that would be a near complete absorption of the liberal party by the ndp.

in the east & ottawa, again, the numbers are not exactly surprising. it's not clear what is ottawa and what is east, but the conservatives always dominate the rural east and the liberals usually do well in ottawa. with the liberals running over 30, it might suggest that they may lose a seat or two to the conservatives on the edges of the city.

i've been saying the same thing over and over again - this is about the 416.

and, despite what frank says, if those numbers are correct, you'd be looking at something like the following:

ndp ~65
conservatives ~30
liberals ~15
so, that gets me through season 6 - up to the end of may 19th. there's one more season of videos before i get to the hook-up, and roughly 100 pages of data to rebuild.

i need to stop to eat, but could conceivably get most of it done by midnight.

the master music document is now over 1200 pages.
http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/j-reacts-to-updated-data-on-sanders.html
if justin trudeau wants to win an election in alberta, he should join the conservative party.

that's his one and only option to achieve that end.
i mean, this was always the issue with trudeau, he had to prove he wasn't a moron.

having him slow down and talk like cletus the slack-jawed yokel to appeal to idiot albertans is going to reinforce the criticism that he isn't very bright and cave him in the urban centres.

canada is very big, but in terms of population density, it's one of the most urbanized countries in the world.

whoever told him to do that should be immediately fired.
why is justin trudeau all of a sudden speaking at a pace suggestive of the idea that somebody snuck into his room when he was sleeping and gave him a lobotomy?

that forty minute speech of his would have taken ten if he was speaking at a normal pace.

is he trying to appeal to rural voters, or something?

bad idea: he sounds like a retard. that's going to kill him with educated voters, who are already playing lesser evil with him.
http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/j-reacts-to-kentucky-primary.html
http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/j-reacts-to-canadian-transgender-human.html
the thing about something like this, or sabbath, or zeppelin, or iron butterfly, is that if you get the random urge to listen to it at this stage in history, you're going to go to youtube, rather than spin the disc.

http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/shit-hillary-said-vol-1-59.html
http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/i-am-suspending-shit-hillary-said.html
http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/shit-hillary-said-vol-59.html
http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/shit-hillary-said-vol-58.html
http://dsdfghghfsdflgkfgkja.blogspot.com/2016/05/shit-hillary-said-vol-57.html

this isn't your typical "third party splits the vote" type situation.

she can fuck herself over by splitting the vote.
no, i need to be crystal clear: it is in andrea horwath's self-interest to ensure that she does not hand liberal seats to doug ford by campaigning too hard in seats she can't win.
i'm not sure there's any good answers in here, but it's an interesting read, and maybe a reality check if you live in canada.

i think i'm clear about what i want, even if i don't have a good name for it. i tried to do a google search for "post-culturalism", but i didn't get any hits. culture, to me, in the sense of it being an ethnic identity (rather than a synonym for art) ought to be looked at entirely derisively, as a relic of the past. but, i want to avoid this term 'cosmopolitanism', because it doesn't come attached to any actual set of rules. it's the right idea, but i want to attach it to science and logic and reason specifically, and it doesn't do enough to make that clear.

but, i have no interest in relativism, and no interest in respecting the rules that exist in different cultures, especially if they are oppressive or require some concept of relativism to tolerate. i'm perfectly comfortable with being the anarchist that wants to rip all these structures apart, and send the religious groups the bill for the damage. i have no qualms at all with looking a religious minority in the eye and telling them their religion is wrong and they should change their views, or abandon their beliefs. and i would actually like to see more people be pro-active about it; to me, this kind of evangelical atheism is actually a valid and needed form of social activism.

it's less that i want to undo the immigration, exactly, and more that i want some more aggressive approaches to ensuring that the society remains entirely secular - and that agents of any kind that want to undo this are more vigorously and more comprehensively opposed.

http://afrasia.ryukoku.ac.jp/english/publication/upfile/WP19.pdf

when you have a small, homogeneous state like kansas or new brunswick, the assumptions in the modelling should work out fine for whatever kind of outcome you want.

but, when you have a big, heterogeneous region like ontario - which is at least four very distinct regions with totally different everything - all you're doing by aggregating and modelling is polluting sample.

data from kenora or sarnia is worse than completely useless in modelling brampton, it's actually distorting and misleading.

you can do this right using distinct and mutually exclusive sampling frames, but nobody does it. and, how accurate you are depends on how precise the frames are.
there are certain areas in toronto that horwath ought to actually avoid, as an ndp bump is more likely to swing a red seat blue than orange.

i'm sure she has better riding data than i do.

so, i would put out a call to the ndp campaign to consult it very carefully - and to avoid taking modelling data seriously.

just briefly...

what the models do is they take data from across the province and try and squeeze it into specific ridings. so, if i'm talking to lawyers here, they're taking data that is overly broad and trying to force it into very specific scenarios that ignore the specifics, like incumbency and turnout and demographics.

personally, i'm going to argue that the whole idea is flawed - that the premise of modelling is overly broad and should be struck down as unconstitutional (the constitution being the predicate calculus, i suppose). but, that defence is untested.

riding polls ask people in the riding - it's directly empirical. much better. but, watch out for small sample sizes...