Wednesday, March 4, 2020

it's worth pointing out that the united states has 158 cases and 11 deaths from coronavirus, and canada has much fewer cases, but zero deaths.

germany has 322 cases and zero deaths. switzerland has 93 cases and zero deaths; the uk has 87 cases and zero deaths. spain has 2 deaths out of 228, and france has 4 deaths out of 285. there is one death in san marino. while most countries have seen few cases, the only other country with deaths is italy, which has a very high poverty rate.

if you look at just france, the mortality rate is 1.4%. in spain, it's 0.8%. put together, and it's 6/513=1.1%. add the uk & germany & switzerland, and it's 6/1015 = 0.5%. i'm not cherry-picking countries in western europe - the more countries i add, the lower the death rate gets, because there aren't any more deaths.

while there are probably unreported cases, the official numbers put the death rate in the united states at nearly 7% - 15x as high as western europe.

why is that?

could it have something to do with the health care delivery system in the united states?

this is the best argument for medicare for all that's ever existed....just look at the numbers. if you're an american, you should be embarrassed - your mortality rates are comparable to those of developing countries, not those of the industrialized world.
so, i took a nap, and i'm in for the night.

i decided yesterday that i wanted to get something to eat while i was watching the results come in, so i got most of my grocery shopping done for the month last night (and made some eggs when i got back). i finished that up with a last run this morning, before crashing when i came in.

i need to take a shower, and hopefully the laptop comes out of hibernation nicely...

the weather forecast has taken a turn towards the crappy tomorrow night, and i'm wondering if i want to go out at all. we'll see tomorrow.
i'm sorry to say it, but i watched most of the debates and...

trump is going to rip biden to little pieces. all of those pauses, trail offs, etc - trump will mercilessly annihilate him as a senile nincompoop, and even lifelong democrats will be forced to agree that, while trump's policies may be scary, biden is just not fit for the job.

it's going to be one of those elections.
bob dole is actually still alive, though.

woah.
"bob dole doesn't think he's much like joe biden."
but, with bloomberg gone?

bernie is going to just get smashed in some of these remaining states, like it's 2016 all over again.

and, he's not doing well in the north, like he was then, either.

i do not want warren to drop. clearly, she should drop. but, it's not going to help sanders the way he thinks it will..and it's going to set up this disastrous choice, from a general election perspective.

say terrible things about clinton if you want, i have and will again, but she was a smart, educated, capable woman that had the ability to attract all sections of the upper class. she did well with the nouveau riche, with old money, with educated professionals, etc - she was just awful on actual policy and it freaked out a lot of wonkish types, on the left.

biden is a friendly, apish dunce - a village idiot, more in the style of a gerald ford and, kneejerk reactions aside, he's just not going to maintain the support of these necessary swing demographics, even if he somehow managed to shock and awe them into it for the short term. he can't speak their language. he's really not one of them.

if the party walks down this path, where it forfeits the educated professional in favour of appealing to the low information voter, it's going to lose all of the most important districts in november. it won't even be close.

if warren reminded me of dukakis or adlai stevenson, and ran the risk of being smeared as a liberal egghead, what biden reminds me of is a bob dole or a john mccain, and runs the risk of getting completely shunned by actual smart people.

they need to keep the warrens and the buttigiegs kicking around because they need to have a way to appeal to these people.
bloomberg dropping does not surprise me.

he has better things to do than waste his time talking to voters.
i can imagine there were probably some people that had no idea, and got hit with an unexpected hydro increase, from a government that campaigned on decreasing hydro costs. if they voted for them, that would have been a nice shit sandwich.

but, i actually understood what they did and adjusted.

and, now i should get compensated for that....
so, remember when i was freaking out about the electrical, and how they changed the formula to undo the rebate?

i've mostly avoided the issue by not doing laundry, which has let me kind of balance the usage out to roughly zero. it's been a dollar or two one way or the other.

well, i got a letter in the mail letting me know that this actually wasn't a nefarious ploy but an "honest mistake" (this is the doug ford government and, for all his phony populist rhetoric, the tories are not known for caring much about poor people - it's actually not hard to believe that it never crossed their mind), and they've:

1) fixed the formula to apply the discount before the rebate
2) will recalculate the last several months worth of usage and send me a lump sump credit

given that my electrical has legitimately gone way down, this could be almost enough to wipe out the balance. and, if i understand the change correctly, i should actually now have much more space to consume without it hitting the limit.

which would be great.

as i want to start using some of this gear soon....

now, if i could get a similar letter from the cra about the carbon tax...
so, what's the summary, then?

- the exceedingly compact movement of buttigieg & klobuchar voters to biden in the fourth northern states that voted today raises some serious questions about the sanctity of the results. however, sanders' showing was fairly weak, anyways, after months of neglect - which is kind of exactly why clinton lost there, isn't it? biden may have cheated, but he's only going to get away with it because sanders has lost enough support in his actual base to allow him to.

- while biden won the south, which was widely expected, he did so with much weaker levels of support than clinton did, partly due to bloomberg showing up and splitting the vote. bernie was not able to take advantage of that anywhere within the south. worse, he squandered the opportunity by polling poorly in the north. the dropped candidates are not likely to have performed very strongly in these states, so i didn't look at the consistency of the results very carefully.

- the west does not appear to be suffering from the same anomalies that are appearing in the north, and the results are more or less in line with expectations.

after being beaten in the north, fairly or not, bernie should be crushed and defeated. however, bloomberg has split the vote up enough to prevent biden from putting him away (fairly or not), and this will carry on a while longer.

however, bernie had better win a few of the states that propelled his 2016 run, and by substantive margins, or he's going to run out of states to win in and have to concede.
california would also appear to be roughly in line with the polling, including for the two recently dropped candidates. was there a lot of early voting? or is this actually hiding a last-minute spike for these candidates?

if buttigieg was flirting with viability right before he dropped, then a showing of 7% would be about in line with expectations. and, i guess these voters end up back where they were.

or, the small amount of support he had might have been very dedicated....

but, it doesn't seem like they cheated in california, or, not right now, anyways.
utah is actually...not screwy. how 'bout that.

buttigieg got about half of his polling numbers - roughly what you might expect from somebody that just dropped. biden got a small bounce, not 90% of buttigieg voters. and, bernie got a little bounce, too. warren & bloomberg are roughly in line with their polling numbers.

are the mormons the truly independent thinkers, here? 

or did the biden campaign overlook this?

colorado seems to be kind of weird in a lot of ways.

first, the polling was actually pretty accurate, with the caveat that buttigieg and klobuchar are not being tallied, apparently. these were apparently all mail-in ballots...

so, i'm going to withhold analysis.
all things considered, if sanders had won the important blue and purple states that he was supposed to win - maine, massachusetts, minnesota - then he'd be in a decent position, right now - because bloomberg came through and split the vote in the south.

it shouldn't matter if biden cheated or not, because sanders should have never ceded that much ground to buttigieg in the first place.

sanders didn't lose because of his policies, he lost because of poor tactical planning.

i've had a running commentary on this, and i hope it is useful to future candidates that run into this same problem of trying to find a way around these red states in the south.
what happened to bernie is kind of like what happened to the last romanov - he got involved in a pointless war brought on by his mindless ethnic solidarity for the serbs, and then had to face a revolution at home.

was the czar morally right to stand in solidarity with the serbs? well, you're ignoring the calculation of his own ambitions, as the romanovs had been seeking to swallow the slavic speaking areas of eastern europe for quite some time. standing in solidarity with the serbs over the death of the archduke was really just a cynical ploy. of course.

but, let's hope he keeps his head on for the rest of this and figures out how to get out of it.
i don't think they sing cardi b songs in church very often, bernie.
will biden win these northern states in the general?

no.
regarding the states in the south, this is nowhere near as bad as last time, and it might not be insurmountable for sanders - if he can figure out how to win in states like minnesota.

but, if michigan looks like minnesota, he's fucked....

my skepticism about the sanctity of the results aside, it had been clear for a while now that sanders's decision to focus on the south instead of shoring up support in the north was a major tactical mistake. he was never going to get all of these religious fundamentalists to vote for him, and he should have had the intuition to key in pretty quickly on the centrality of religion in the south. i heard him say things like "we have a big lead in the north and can afford to focus elsewhere", but that's just taking them for granted. you have to tend to your base, or you lose it.

he won all those states by huge margins four years ago. even with fishy results, why was a coalition of centrist candidates able to beat him in these states? why couldn't he get above 35% in states he dominated four years ago?

i've pointed this out before: he's running for commander in chief. that's a tactical mistake that a smart general just doesn't make. it's called "thinning out your troops". it's akin to fighting wars in foreign lands, instead of fixing the problems at home.

so, you can follow my arguments or reject them, but it doesn't change the reality that sanders' decision to put all of his efforts into winning in the south has backfired very, very badly. not only did he not win in the south, but he lost states in the north that he should have won very easily. this should be deeply concerning to him, as it may reflect badly on his ability to compete in a slew of other states he's expected to do well in. we may find out very soon that he's done, and this is over - not because biden cheated, although it seems as though he did, but because he didn't tend to his base, lost their enthusiasm and in the end lost their interest.

he can't win the nomination solely on the strength of hispanics and teenagers, but he ignored everybody else, and they appear to have moved on.

so, there's going to be a narrative about how sanders got beaten badly in the south, and it's inaccurate. he seems to have improved his totals in most places, and biden is not getting numbers anywhere close to the ones that clinton got - mostly due to bloomberg, who is just aimlessly splitting the vote out of vanity. if sanders does turn this around, bloomberg may find himself a rather hated person, because he's prevented biden from really putting him away.

rather, sanders got beat badly in the north - areas he was supposed to be dominant in, and this is a consequence of his tactical error of focusing resources on winning red states in the south.

this should be a lesson for the next northern liberal that tries this - don't make the same mistakes as bernie!

for right now, i want to say "if his campaign does not realize the error and course correct, it's days are numbered", but the truth is that it's probably too late.

he'll probably win arizona, though.
it's a farce. it's theatre.

i knew that.

i'm not sure bernie does...
and, yes - it is the same thing in maine, where there was not much polling done, but you have to scrounge together all of the buttigieg/klobchar/biden voters to even get close to 34%.

buttigieg was polling ahead of biden, who you wouldn't have expected to end up viable if this was done last week.

and, the other three candidates - sanders, warren, bloomberg - were all running in line with their averages. no bumps at all.

so, that's four states, now, on one night, where you have to accept that biden swallowed virtually every single klobuchar voter and virtually every single buttigieg supporter - an event so rare, that you would consider it virtually impossible (not technically impossible.) to happen even once.

what are the chances?

(1/10^6)^4 = 1/10^24.

Yotta is the largest decimal unit prefix in the metric system, denoting a factor of 10^24 (1000000000000000000000000), or one septillion. 

these are the exit polls (conducted by major corporations.) for massachusetts, and tell me what jumps out at you about them.

men: 34% biden
women: 34% biden

white: 36% biden
black: 36% biden

democrats: 34% biden
independents: 32% biden

what do you notice about those numbers?

& check this one out - in massachusetts:

very liberal: 27%
somewhat liberal: 37%
conservative: 36%
listen, i'm the one that pointed out that sanders' numbers were an illusion, and he'd get leapfrogged once some weaker candidates dropped out.

but, i never thought for a second that you could just sum the totals together. i expected some bleeding, in a few directions - including towards sanders.

the math is too clean, it's without error, and, in the real world, everything is defined by error. when you see an absence of error, you expect human engineering - a designer, a creator. in the real world, there is no designer, no creator.

when you see the error dissipate, there's strong reasons to suspect it's because the results were fabricated - real results would be messy and full of error, not defined by these clean transfers of votes.
while the results of the vermont primary are not surprising, you see the same very curious result that you saw in massachusetts - biden got 90%+ of the buttigieg/klobuchar vote, while warren & sanders got no measurable bump at all.

this kind of groupthink is just overwhelmingly unlikely; there's, like, no split. as though somebody added the numbers up to make sure they made sense...

the results are not as extreme in minnesota as they are in massachusetts. you would expect at least some klobuchar supporters in minnesota to move to warren, and she did get a minor bounce. sanders also appears to have gotten a minor bounce.

but, you still have to accept that klobuchar convinced almost all of her voters to vote for biden, who had been polling in single digits here since he announced.

i could run through all of the same arguments, with the caveat that they're all a little weaker - it's 80% rather than 90%, and still very hard to believe, if that little bit less hard to believe. but i won't. i made my case and it's transferable....

i suspect that i'll find a pattern here with this.

what you would expect is to be able to model the movement of voters from buttigieg & klobuchar to biden using a probability distribution. so, you'd end up with some type of curve. and, there's even some artistic license here, there's no specific right answer.

but, when you do the work and realize that the probability distribution that you pull out via empirical analysis is.....uniform? that sets off alarm bells, it raises red flags.

it's not technically impossible, and rare events do happen. but, it's exceedingly unlikely to happen even once, and approaching impossible to see it happen in several states, all on the same night.

i would advise the sanders people to...i don't know what they do. conduct independent local polling. ask for readouts from the machines. i don't know. but, do something.

it's like we're entering the twilight zone of election results with this.
education rankings of states so far, for reference:

iowa - 29.
new hampshire - 8
nevada - 43
south carolina - 44

massachusetts - 1
colorado - 3
vermont - 4
virginia - 6
minnesota - 10
utah - 11
maine - 18
california - 25
north carolina - 31
texas - 39
oklahoma - 40
tennessee - 41
alabama - 46
arkansas - 47
is it possible that he won more by default than due to the endorsement? that is, that a very large percentage of people said "my preferred candidates have dropped, and i have to vote because it's a civic duty, and he's the best option left, so i suppose i'll vote for him. i guess. if i have to."

potentially.

and, you'd have to assign something like that to a few voters, regardless.

it's the margins, though. that's 5% of buttigieg voters, or something. it's not 90% of them....

- sanders appears to have gotten no bounce at all, not even in the rural areas that he carried comfortably in 2016. buttigieg was winning these voters from the start, but it's an anti-establishment vote, and you'd expect them to vote against biden, not for him. sanders actually did fairly poorly in these counties.
- warren did not get a bounce, either - and you'd have expected her to pick up some educated white women, at the least, from klobuchar.
- maybe bloomberg was never going to get a big bounce from buttigieg dropping, but you'd think you'd see some minor uptick. nope. 

biden literally got the entire thing, as though somebody just reproportioned it in a spreadsheet.

....which i suspect is actually the truth.

so much for massachusetts..
The most educated state was Massachusetts, and the least educated was Mississippi, according to the report.

biden will do well in mississippi.....

there's two reasons why the education factor makes it so hard to believe:

1) educated people tend to think for themselves. endorsements don't tend to have the kinds of effect on educated people that they have on uneducated people. they don't tend to follow trends, or be easily peer-pressured. 
2) biden is a complete dunce, which is the reason he had so many problems with them in the first place.

i could believe it if the endorsement pulled 30% or even 50%. but, if you believe this, it was an exercise in groupthink.

so, i can't debunk it, but i don't believe it.
it would seem as though they passed the prisoner's dilemma in massachusetts, at least - quite well. maybe a little too well.

but, this is more of a question for a behavioural psychologist than it is for a nerd with a math degree. i can't debunk this by telling you the numbers don't make sense; i'm more skeptical about the reality that they make too much sense.
let's start with massachusetts, which i find to be the least believable outcome.

biden had been doing well in south carolina for months, so the results were obvious. the recent polling in massachusetts had biden way down the list (behind not just sanders and warren, but bloomberg and buttigeg) at around or below 10%. steyer never registered in this state, and bloomberg roughly matched his polling.

in fact, sanders and warren roughly matched their polling, too.

so, the only possible way to understand massachusetts is to deduce that 90%+ of buttigieg supporters and 90%+ of klobuchar supports both moved, en masse and tout ensemble to biden - which is the kind of thing you see in alberta or quebec, but is almost incomprehensible in what is plausibly the most educated state in the nation.

so, you can pull these numbers out - they're there. but, you have to believe that the mass of eggheads in massachusetts did what they were told, and voted for the village idiot. and, out of what, exactly? fear?

so, if you're naive enough to fall for it, this is, indeed, very easy to understand - just add biden's 10-% to buttigieg's 15+% and klobuchar's ~7% and the number's come out in the wash. easy, right?

but, it's very hard to believe it.

and, i don't quite, myself. it seems a little too tidy...

https://www.washingtonpost.com/elections/election-results/massachussetts-democratic-primary-live-results/
let's see if i can try to make some sense of these results, one thing at a time.

i need to remind you that i made a guess, based on intuition. but, this site does not exist as a platform for my gut. it has enough work to do in digesting. and i don't give it a lot of energy.

i don't take my gut very seriously, and would suggest that you don't, either.

my official prediction was "anybody trying to tell you they can predict what's going to happen is full of shit."

there was no data...you need data to predict things....

but, i'm pretty skeptical about these "results". so, let's see if i can figure out how they cheated, or debunk claims that they did.

there's some complaints about people who voted early wanting their vote back, and it does seem unfair and undemocratic.

an easy way to deal with this would be to move to preferential voting. that way, if your first candidate drops out, you can move to your second choice.

that is, if they actually count them, anyways.
the senate primary in massachusetts is not until september.

so, the argument that the chance to vote for a kennedy pulled out a lot of older voters doesn't work.
these results are really unbelievable aren't they?

i don't know why i bother. i know better.

biden winning massachusetts is about as believable as trump winning pennsylvania.

there's no polling, no way to check up on them...