eightyonekilograms:

Oh hell yes, poll results are now to tenths of a percent. But @staff can we pretty please have the raw counts?

eightyonekilograms:

wildgifthorses: I think you can get raw counts from the API

Yeah, and you can also get the vote percentages to six decimal places in web debugger, thanks to a truly wild web design choice:

fbfed457ff06f272e1cde6d57c8e4c5a7e7e80a1

Sticking the raw percentage directly into the width attribute is just a delightfully bonkers way to implement that, and I love it. And with six decimal places that’s enough to derive the raw count unless you go really mega-viral (and maybe even then it just adds more decimal places).

But it would be nice to have a way I can see right on the dash without having to tinker.


Tags:

#The Great Tumblr Apocalypse #fun with loopholes #fun with statistics #the more you know

keynes-fetlife-mutual:

one of the top 5 things I wish average people had a better grasp of is that the size of a number is contextual. a number’s bigness or smallness is not a fixed property of the number itself, it depends on what that number represents.

when you were a kid, a thousand seemed like a very big number. did you ever count that high? pretty tedious, isn’t it? it’s a massive number of toys to own, or dishes to wash, or people to meet. for most of the countable things you interact with in a normal day, a thousand is a big number. but in other contexts, a thousand can be a small number. It’s a tiny number for the population of a town. it’s a smallish number of dollars – if you only make a thousand dollars a month (in a wealthy country), you’re poor. it’s an utterly minuscule number of bacteria in your body, or dollars in a government budget, or grains of sand on a beach. a thousand of something might be a lot, or might be a little – it depends on what you’re counting, in what context.

most of the time, 0.1% is a pretty small number. if 1 in 1000 eggs has a double yolk, you probably won’t eat very many double-yolk eggs in your lifetime, unless you seek them out deliberately or eat vastly more eggs than the average person. if 1 in 1000 homes in your town have garbage disposals, garbage disposals are very rare where you live. but if a cosmetic surgery had a 0.1% risk of lethal complications, most people would see that surgery as unacceptably dangerous – 0.1% risk of death is a big number.

we’re not very good at thinking at scale. we’re especially bad at thinking about our fellow humans at scale. suppose that 0.1% of the population has some trait X. if you assume that none of the people you meet have trait X, you’ll be right 99.9% of the time. when you crack a joke about trait X at a party, there probably won’t be any X-people around to be offended by it. but 0.1% is 1 in 1000 – how big is that number, really?

it’s about 335,000 people in the US, for starters – the population of a small city. throughout your education, you probably had multiple classmates with trait X. there’s a handful of them at any large school. a medium-sized company will probably have several trait X employees. you might not know someone with trait X personally, but it’s a virtual certainty that someone else you know does. if you have a modest online following, several of your fans have trait X. if trait X were a disease, it would be too common to count as “rare”.

the next time you see a number you think of as “big”, like a million, I want you to stop and consider what’s being counted: does this number represent a large quantity here? compared to what? when you encounter a small proportion, think: how frequent is that, really? what’s the denominator (0.1% of what)? can you think of something comparable that happens about as often? is this number surprising?


Tags:

#yes this #death tw? #fun with statistics #(”fun” isn’t quite the right word‚ but that’s the category tag)

jadagul:

dagny-hashtaggart:

Ah yes, “do we really live longer than our ancestors” articles, the theses of which always seem to be “well once you eliminate deaths by disease, violence, and really every factor other than natural causes in old age, and probably only consider the upper classes (because they’re all we have data for in many cases), the difference in life expectancy from adulthood is only 5-10 years.”

And like, I get that infant and child mortality doesn’t reflect what most people intuitively conceive as life expectancy (though it bears noting that youth mortality on such a scale that it substantially distorts life expectancy figures is also really bad), but I’d argue that things like reduced frequency of wars and other homicides, the presence of antibiotics and other modern medicines, and lower rates of extreme poverty, are exactly what most people think of when they think of factors that impact life expectancy, and it’s blatant cherry-picking to say we ought to treat them as inadmissible.

It turns out that, after controlling for life expectancy, moderns live about as long as Medieval peasants did.


Tags:

#anything that makes me laugh this much deserves a reblog #history #death tw #fun with statistics

transgenderer:

i love how even the studies that support psi effects being real have really small effect sizes, like can you imagine if psi was real but gives you like, a 51% chance of guess correctly at a 50/50 chance

 

transgenderer:

@triviallytrue said:

do you mean ESP? i’ve never heard the acronym PSI

for some reason parapsychology researchers call it psi, even tho it should really be psy. its definitely not an acryonym. it might be like, a…translation thing? spanish for psychology starts with psi

 

sigmaleph:

means you can abbreviate it as Ψ

wikipedia on psionics mentions a claim that that’s actually the source:

In 1942, two authors—biologist Bertold Wiesner and psychologist Robert Thouless—had introduced the term “psi” (from ψ psi, 23rd letter of the Greek alphabet) to parapsychology in an article published in the British Journal of Psychology.[7] (This Greek character was chosen as apropos since it is the initial letter of the Greek word ψυχή [psyche]—meaning “mind” or “soul”.[8][9]) The intent was that “psi” would represent the “unknown factor” in extrasensory perception and psychokinesis, experiences believed to be unexplained by any known physical or biological mechanisms.[10][11] In a 1972 book,[12] Thouless insisted that he and Wiesner had coined this usage of the term “psi” prior to its use in science fiction circles, explaining that their intent was to provide a more neutral term than “ESP” that would not suggest a pre-existing theory of mechanism.[13]

but it could just be people liked “psionics” more than “psyonics” and the abbreviation came from that, idk.


Tags:

#(the language bit is neat too but I’m primarily reblogging for the OP) #yes this

{{previous post in sequence}}


sigmaleph:

sigmaleph:

sigmaleph:

anyway, fantasy universe where reincarnation is real but you keep absolutely none of the memories of your past lives and the only relevance is that your magical power is directly proportional to how old your soul is

secret cabals of wizards fighting over population growth because of their ideological commitments to particular distributions of magical power

“Do you know of Praidib’s law, Firem?” She was standing, talking, as if there was nothing more interesting going in the world

“Praidib…? What does that have to do with anything?!”

“A soul does not grow in power when not in a living body. It was quite ingenious, how Praidib proved it. I’m sure your classes at the university would have covered it eventually”

“If you hadn’t murdered them all? Yeah, maybe I’d have a more complete education.” I had nothing better to do than engage her, I suppose. I could not escape my bindings. “What’s your fucking point, Hillah?”

“Think of the consequences, Firem. The archmage’s soul is ten thousand years old. After the population explosion of the Blue Renaissance, two-thirds of the people in this world have souls less than a hundred years old. Less than one percent of the power that will be wielded by whichever lucky child happens to inherit that soul. And as long as that soul is embodied, it will continue to accumulate power and have a ten thousand year head start on the vast majority of the world. You have seen what people with power do to those without”

“His power certainly didn’t stop you from killing him”

“Nobody should have that kind of power, my dear. Not me, and not him, and not you. But how do you stop it? How do you even begin to slow down a soul’s accumulation of power? Why, Praidib’s law, of course.”

“So you think you’ve solved soulcaging? Is that your big plan?”

“No, of course not. Soulcaging is impossible. If you want a soul unhoused… you deny it a body. There’s a billion souls in the world today. Soon, there won’t be a billion bodies to house them. Or a hundred million. Or even twenty million. I’ve run the numbers. I know how long it took civilization to build up to its current numbers. I have given us time to catch up”

Twenty million. That was what she was planning? That was what her weapon would do? Wipe out hundreds of millions of lives? I could not say anything

“The vast majority of the souls embodied will be, why, the vast majority,” she continued, seeing my lack of reply “The children of the renaissance, with less than a century’s worth of power to them. But they can even out. They can age. The problem will not be solved, not entirely, but…”

“But nothing! In another millennium, those souls will be lucky to have aged another century, and the archmage’s soul will still be ten thousand years old! And every body it has, it will still be an unmatched wizard. You’ve accomplished nothing except mass murder.”

“I told, you dear, I run the numbers. I am well aware. There will still be some great mages being born… but we need not let them live.”

“You… fuck. That device you used earlier. You can track souls by age.”

“Indeed,” she smiled. “I can, and so can my disciples. When our dearest archmage pops up again, he or she will be lucky to make it six months. My organisation will rebuild the world, and for as long as they exist, we will be on even footing. Not me, of course. This is my last life for a while now. But humanity. And when we fail, because we will fail eventually, at least we’d have made the odds closer. I don’t know how many tens of thousands of years it will take, but… best start now”

And saying so, she threw her hands to the sky, and called upon death.

“No, sorry, OK, this just doesn’t make sense”

“Does it really? Or are you just refusing to-”

“No, it really doesn’t. Like, this is not an ethical argument against mass murder, we can hash that out later, just… I can see why you’d want a population below the number of souls, sure. You want a certain number of souls not incarnated and gaining power, and you think you can bias which souls that is with constant selective murder. What makes no sense is dropping the population to, what, two hundredths of the historical maximum? less? The rate at which total human magical power accumulates is proportional to population. If you want new souls catching up to old ones, you want them gaining more power over time, not less. That means a population slightly under a billion, but not much smaller”

“I…what?” She started rifling through some papers in a nearby desk. “I could swear… crap crap crap.”

“Are you sure you didn’t mean you actually wanted to kill twenty million people, rather than leave twenty million survivors?”

“Shut up. Maybe. Look, I outsourced this to Satrean, his notes weren’t super clear, I might’ve… shit.”

“Gods fucking above, Hillah, did it not come up at any point how many people you were going to kill?”

“Oh, I’m sorry, have you ever run a secretive organisation of assassins that’s trying to upend the world order? You compartmentalise information! You don’t have an all-hands meeting every Fireday to talk about your doomsday plans!”

“Well, I apologise for implying you should put your ability to figure out what actually are your goals and how you achieve them above your cloak and dagger roleplaying. I’m sure it’d ruin your fun to double-check.”

“Shit, shit, shit… look, yeah, OK, it makes more sense the other way, you’re right. Do you mind staying tied up to that chair a couple hours more, I need to recalibrate this whole thing”

“Are you going to let me go if I say I do mind?”

“No”

“Worth a try. Anyway, going back to that argument we tabled about the ethics of mass murder…”


Tags:

#reincarnation #storytime #anything that makes me laugh this much deserves a reblog #fun with statistics #fun with loopholes #death tw #amnesia cw #murder cw

sigmaleph:

sigmaleph:

just-evo-now:

Okay so friendly reminder that if we lived in a past-lives verse (so one where we all remember our past lives and no one shares past lives) we would remember …. big drumroll….

14 people

so that’s the average, right, all humans who ever lived divided by current population

but presumably the distribution is not even, we shouldn’t expect the total number of souls has always been ~8 billion. If, say, new souls pop up whenever a new human is born and there isn’t one going spare, you’d expect most souls to have originated in the 20th century and have had at most a couple lifetimes, so conversely some souls must have more lives?

about one average human lifespan ago (so, the 1950s) there were about 2.5 billion people. five billion less than now. That was also a global high in population, so assume at that point there were as many souls as humans, that means two thirds of all souls originated between now and then and have had… what, a life, life and a half each on average? if you say life and a half, the five billion newest souls account for 7.5 billion of all the lives ever, which means those remaining 2.5 billion souls have about a hundred billion lives split between them, so averaging like 40 lives (and probably there’s still a big skew there but this post is already not careful enough, I’m not going to extend this all the way to the industrial revolution or whatever). Still not at “I was 600 peasants” territory though.

(decent chance i fucked up the math or the logic somewhere, etc)

oh also i think most of your lives should be profoundly unmemorable because you died in infancy


Tags:

#reincarnation #fun with statistics #death tw #amnesia cw #(I’m not saying this is *true*) #(but if I *were* on precisely my second life‚ that would honestly explain a lot)


{{next post in sequence}}

Criminal Georg skews recidivism statistics

nostalgebraist:

stumpyjoepete:

michaelkeenan:

Have you ever seen those concerning statistics about criminal recidivism? Like: 44% are re-arrested within a year, and 83% within nine years (source: this Department of Justice report).

I’d seen those statistics before, and been concerned. There’s a great case for shortening prison sentences for deterrence reasons, because likelihood of punishment is much more deterring than severity, but at least prison incapacitates criminals from plundering society while they’re imprisoned. Why hasten prison release if they’ll be back soon anyway? “Once a criminal, always a criminal?”, asks one headline about recidivism.

But today I learned that there’s a huge caveat to those statistics. The more often you go to prison, the more you’re counted in recidivism statistics.

Consider five people who go to prison. Four of them never commit another crime, but one of them was Criminal Georg, who is imprisoned ten times. Out of the fourteen prison sentences (ten for Georg, four for the others), nine of them are followed by recidivism (Georg’s first nine). The proportion of these people who are serial criminals is 20%, but the recidivism rate is 64%.

When considering people rather than prison releases, the recidivism rate is lower than I thought.

see this thread for more examples

Thank you, I hadn’t seen it and it’s a great resource!

I knew I’d seen this pattern before, but I didn’t have a name for it. The linked post by Elizabeth Wrigley-Field tells me it’s called “length-biased sampling.”

The mentions several examples with real-world importance, incl. the recidivism one, and argues the concept should be more widely known.

(It also makes an argument that “length-biased sampling is the deep structure of nested categories” which sounds interesting but which I am not awake enough rn to wrap my head around)


Tags:

#fun with statistics #(well maybe not ”fun”) #the more you know #prison cw

pileofknives:

60e9b70ad31cf6fcdfbc096968b31a07c3e80e26

“I bet it doesn’t hurt that bad, I don’t have time for this shit.”

 

whatifweanarchist:

“at least one”

 

sexhaver:

now this is a gender binary i can get behind

 

carnival-phantasm:

Very generous of them to describe it as the “Thinking Period”

 

retiredmahoushoujo:

ad067ad5ab66686b5e20ed26c13546a34d9bb97e

They correctly excluded the outlier Electroshocks Georg

 

shacklesburst:

everything about this is extremely underpowered, except for electroshocks georg, who, by now, must be extremely overpowered

 

rustingbridges:

you’d think they would have been able to get more than 50 undergraduates to do. it doesn’t take a long time and requires answering no surveys

Also, psych researchers have a reputation for lying their asses off to the subjects regarding [what a study is actually about] until after it’s over. I would seriously consider pressing the button once or twice just to see if they were telling the truth about it being a shock button.

(Not sure if I’d *do* it, but I’d seriously consider it, and I would definitely wonder if it was some sort of covert test that I was failing by not pressing it.)


Tags:

#the first draft of this post only had commentary in the form of the tags #”basically what shacklesburst said” #”I’m not comfortable with the gender-focused framing here but I’m reblogging for Electroshocks Georg” #but–while I’m still primarily reblogging for Electroshocks Georg–I realised I did in fact have something to say #reply via reblog #the power of science #sexism cw #this probably deserves some other warning tag but I am not sure what #anything that makes me laugh this much deserves a reblog #Spiders Georg

{{previous post in sequence}}


f90f9ea9a89069bda9d435468a2f4a65a23c4707

collapsedsquid:

Since the beginning of 2020, customer satisfaction with scented candles has been dropping at a much faster rate compared to unscented candles.

(it’s easy to overlook the Twitter source link in the Tumblr metadata and there’s some useful context in there, so here’s another link to it)


Tags:

#followup to the previous post #which I had been wavering on whether or not to reblog for a couple days #but seeing this one made me decide in favour #illness tw #fun with statistics #covid19 #oh look an update


{{next post in sequence}}