🏅 Top traders

Alas. The sun did not rise at 88.135 latitude, 157.587 longitude, for it never set in the first place, as it is currently perpetual daytime in the Arctic Circle.

A complete transcript of the script will be posted once it finishes running in 1-2 hours. For now, we may resolve as No, as per the description.

an honor working with you all. o7

@AndrewQuinn The sun rose in 808/1000 locations on August 9, 2023. Full transcript output at: https://gist.github.com/hiAndrewQuinn/48a80663cd34c6b3a52e52b72155b59e?permalink_comment_id=4657461#gistcomment-4657461

@AndrewQuinn wow that's a lot, what is the share of no-sunrise-today-surface to total earth surface? I just ball parked when betting, curious about the real expected value

@LeonardoKroger It surprised my wife and I as well. My original Fermi estimate was as follows:

• At the summer solstice, June 21, about 1/3 of the planet gets kicked into one phase. (I don't know the exact latitudes where the cutoffs begin , even though I apparently live quite close to one here in Finland.)

• At the autumn equinox, September 21, this 1/3 falls to 0.

• A linear extrapolation seems reasonable, so the day halfway between 2023.06.21 and 2023.09.21 is August 6, which is close to August 9, so I was originally guessing about 1/6 of the Earth would not see the sun rise on this day. Clearly an underestimate, but more in the ballpark than I was expecting to be.

@AndrewQuinn No hardcore EV calculated for you, though. Not good enough at geometry to do so quickly :(

@AndrewQuinn you sampled points wrong, it's not a uniform distribution. Please google the formula for sampling from uniform distribution on a sphere

@Lavander https://manifold.markets/AndrewQuinn/will-the-sun-rise-tomorrow-august-9#ph4PONqvLTrUeGaSKOLh

Will the sun rise tomorrow, August 9, 2023?

Resolved NO. To stay unbiased, I will not bet in this market. In order to decide the fate of this question, a Monte Carlo method will be employed. We will generate 1,000 random latitudes and longitudes, and check to see whether the sun indeed rose at all 1,000 points. Random seeds from the fairlyrandom bot and Python source code will be provided tomorrow so you can replicate the experiment yourself. Market will resolve N/A if I cannot find an adequate API to determine whether the sun rose xor if the sun explodes. Edit: The script works! We are using https://sunrise-sunset.org/api and the script itself is preserved along with its chosen random seed 12345 at https://gist.github.com/hiAndrewQuinn/48a80663cd34c6b3a52e52b72155b59e

@AndrewQuinn okay, i see.

I am having difficulty figuring out how to add this market to the FairlyRandom group, so I will do the next-best thing and choose an obvious seed so you can have reasonable certainty I'm not gaming the results: 12345.

The script in question: https://gist.github.com/hiAndrewQuinn/48a80663cd34c6b3a52e52b72155b59e, running now.

Please resolve

@JosephNoonan apologies, scripting now

@AndrewQuinn how's the casino going about its business?

@LeonardoKroger just got back from work my bad, gonna set to writing the script now

placed M1 bet, so i can give 1 star review afterwards

@Lavander why?

@LeonardoKroger why not? Like, it's pretty shitty behavior.

title: "Is 2 + 2 equal to 4?"

description: "To stay unbiased, I will ask my cat and try to guess its opinion and resolve accordingly"

sure, yeah, cool. But i will give it 1 star review

@Lavander let me politely disagree, based on a couple of months of Manifold experience:

To start: Is this market based on a trick question, or "Fangfrage", as we say in German? Well, yes, kind of.

There's an important difference here though. You will find markets that engage in splitting hairs about English grammar. There are markets that manipulate you by contradicting description and title. And most of the time, you'll get frustrated with markets that are just plain sloppy when setting minimally clear criteria. All this is not fun and often really what this market surely is not: They deceive you, intentionally or not. Not to speak of markets that simply resolve wrongly.

This market is not deceiving and I don't expect the resolution to be wrong; let me explain.

Is this market actually a prediction market? Yes, there's the Monte Carlo simulation and Earth's physics to be studied in order to get to a reasonable estimate.

Is the question misleading by seemingly having an obvious answer, yet only revealing important facts if you read the description? No, as Manifold is a globally accessible website the "obvious" answer is based on implicit assumptions.

Does the question contradict the description or vice versa? No.

Could you reasonably answer the question without reading the description? Yes, the fact that it seems to be so obvious should make you think twice and evaluate your own assumptions.

"Oh you're such a pedantic! It's shitty behavior!1!!11!!"

Why not take a different approach? People loose mana, not real money by the way, on this kind of market but could learn valuable things and surely not fall into the (same) trap again (even outside Manifold):

• Frustration management

• The world around you is bigger than your own world (or: it's not about you)

• Read the description (or: be careful with rushed and implicit assumptions)

• Analytic thinking (which inputs do I have, and which ones are missing maybe?)

• Math, Statistics, Geodesy (?)

I mean, all this for free and in a fun way. C'mon, it's better than loosing real money to some Coin Scam in order to learn a lesson.

We all do stupid things sometimes all the time, me too and you too. Smile more, 1-star-review less.

@LeonardoKroger

Could you reasonably answer the question without reading the description? Yes, the fact that it seems to be so obvious should make you think twice and evaluate your own assumptions.

This doesn't make sense. You would never reasonably expect that this market is going to resolve the way it does without reading the description. I would expect it to resolve YES if the Sun rises anywhere on Earth from the title alone. Though, to be fair, people should always be cautious about betting on seemingly obvious things without reading the description, especially when the probability doesn't match the obvious answer.

😭 😭 😭 I lost M400 on predicting whether the sun will rise

@SavioMak We'll see, maybe it will rise after all

At least one of the random points will probably be far enough into the Antarctic Circle that the Sun won't rise there tomorrow.

@JosephNoonan Actually, it could be in the Arctic Circle as well, since, if the Sun is out for the whole day, it never actually rose.

@JosephNoonan Lol I hadn't thought of that. That decreases the odds to even less than the 1 in ~10^-46 that I had. Probly gotta nearly square that

@chrisjbillington Was 10^-46 the probability of none of them being in 24-hour darkness? I wasn't sure what the probability was because I don't know exactly what latitude is required for 24-hour night or darkness tomorrow, but that sounds about right.

@JosephNoonan Correct. It's very rough, I just eyeballed the edge of civil twilight on this map and got the approximate latitude from google maps, arrived at 10^-46 and realised it didn't matter if the estimate was grossly wrong.