How the law of large numbers works.

NicWester

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

ExarTheKun

I think that one pretty large issue with these types of statistics, is that people seem to believe one bad roll increases your odds of a decent roll the next time around.

For instance, let's say you have a 10% chance to recieve a full character from a chromium, rather than just shards. That doesn't mean nine failed attempts later, you are guaranteed a character on your next pack. You still have a 10% chance, even if you've gotten squat on your last thirty tries. RNG really is random.

Loose_Lee

Im laughing out loud...

Your thread inspired me to write it out completely.

aeaswen

NicWester wrote: »

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

I'm not sure any Sane person thinks rng doesn't work against them or is rigged. It's why you have so many 0/8 1/15 in gear farming. Gets you to buy crystals.

Vorgen

@ExarTheKun, you do realize your example doesn’t really reflect what the op is expressing, yes? By your example, the similar question in op’s post would have been “what is the odds of the ship dodges on the 8th attack”? The answer would have been 44% doesn’t matter whether it has dodged the previous 7 attacks in a roll. The % op calculated (not checking his math), would be comparable to your example only if you rephrase your question to “what’s the odds of me getting at least one full character in any of say 8 pulls of chromium”? Then the answer would not be 10%.

ewqewqc

NicWester wrote: »

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

lol...
you dont understand some stuff aswell. your basis is that if someone played vs this ship, he can be taken into consideration of this "8 tries to attack". he's not. if he killed the ship with 2 turns, with 3 turns, with 4...7 turns, he shouldn't be in the Math of "attacking 8 times". (and i failed to understand this logic for long long time, my university math lecturer explained this specific kind of thing for me for a WHOLE HOUR)
therefore, its not one in a thousand. if you got to the point you missed 7 attacks, the chances of miss the 8th is just STILL 44%. therefore, 4 of 10 people who missed 7-in-a-row will miss the 8th.
your point is still valid, and the RNG whining is out of control and people have very narrow sight.. but you are welcome to read about it aswell it obviously dont make you dumb, different people have different strengths and weaknesses, and your thread feels like really bad taste aswell.
BUT HEY, my comment is also salty, so humanity is just doomed because we are all bad people

NicWester

aeaswen wrote: »

NicWester wrote: »

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

I'm not sure any Sane person thinks rng doesn't work against them or is rigged. It's why you have so many 0/8 1/15 in gear farming. Gets you to buy crystals.

Except any time anyone actually tracks results with a statistically significant number of results they find that the expected values are played out--roughly 33% for shards, roughly 15% (I think?) for purple gear, etc.

aeaswen

NicWester wrote: »

aeaswen wrote: »

NicWester wrote: »

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

I'm not sure any Sane person thinks rng doesn't work against them or is rigged. It's why you have so many 0/8 1/15 in gear farming. Gets you to buy crystals.

Except any time anyone actually tracks results with a statistically significant number of results they find that the expected values are played out--roughly 33% for shards, roughly 15% (I think?) for purple gear, etc.

Sure but not in a regular pattern. Designed to have hot and cold streaks that lead to spending. Not to mention how bad arena rng is...

OnEMesSduPKiD

There are not 100,000 people on this forum though.

Mullato

But you are the only person playing your game. Ha.

NicWester

aeaswen wrote: »

NicWester wrote: »

aeaswen wrote: »

NicWester wrote: »

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

I'm not sure any Sane person thinks rng doesn't work against them or is rigged. It's why you have so many 0/8 1/15 in gear farming. Gets you to buy crystals.

Except any time anyone actually tracks results with a statistically significant number of results they find that the expected values are played out--roughly 33% for shards, roughly 15% (I think?) for purple gear, etc.

Sure but not in a regular pattern. Designed to have hot and cold streaks that lead to spending. Not to mention how bad arena rng is...

You are correct that the RANDOM NUMBER GENERATOR is not set to a REGULAR PATTERN.

aeaswen

NicWester wrote: »

aeaswen wrote: »

NicWester wrote: »

aeaswen wrote: »

NicWester wrote: »

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

I'm not sure any Sane person thinks rng doesn't work against them or is rigged. It's why you have so many 0/8 1/15 in gear farming. Gets you to buy crystals.

Except any time anyone actually tracks results with a statistically significant number of results they find that the expected values are played out--roughly 33% for shards, roughly 15% (I think?) for purple gear, etc.

Sure but not in a regular pattern. Designed to have hot and cold streaks that lead to spending. Not to mention how bad arena rng is...

You are correct that the RANDOM NUMBER GENERATOR is not set to a REGULAR PATTERN.

Not what I meant and you know it.

NicWester

That's because what you said is nonsense. If it's a random number generator that's been set up to be streaky then it's not a random number generator any more. Moreover, if it's streaky then it isn't set up to reward spending money, it's set up to reward patience. Stop being so cynical, Diogenes.

aeaswen

NicWester wrote: »

That's because what you said is nonsense. If it's a random number generator that's been set up to be streaky then it's not a random number generator any more. Moreover, if it's streaky then it isn't set up to reward spending money, it's set up to reward patience. Stop being so cynical, Diogenes.

Cynical= realistic. It's obvious man. Same with arena rng. My high potency gets me what exactly? More debuffs landed? Ok...

NicWester

Your high potency means nothing if the debuff doesn't proc. If you have a 60% chance of exposing someone with Poe, that means 40% of the time you fail outright whether you have a zillion potency or zero. 60% of the time it contests potency and tenacity and gives you a result that none of us know because we have no idea what the formula is. Potency and tenacity aren't necessarily 1:1, just as Physical Attack and Armor aren't 1:1.

Nikoms565

SWGOH; you will never find a more wretched hive of confirmation bias and gambler's fallacy. We must be cautious when we use actual facts and logic.

An00byNubeNewb

Loving this post lmao<3

Olga

Nikoms565 wrote: »

SWGOH; you will never find a more wretched hive of confirmation bias and gambler's fallacy. We must be cautious when we use actual facts and logic.

What facts and logic? I've never heard a dev state the numbers in this game are decided by RNG.

sneeekypants

NicWester wrote: »

That's because what you said is nonsense. If it's a random number generator that's been set up to be streaky then it's not a random number generator any more. Moreover, if it's streaky then it isn't set up to reward spending money, it's set up to reward patience. Stop being so cynical, Diogenes.

This is an odd thread.

A few points to add:
-arithmetic mean is a very poor statistic used alone
-you can have induced variance while still using rng and maintain a particular average if that was their intent or the ending drop rate could be just the resulting average of a particular algorithm. Say set the probability to 10% for 100 samples, then 50% for the next 100. That would converge on 30% over the long run but give you very "streaky" results. Or you could even use an organic probability.
-There's bugs, lots of them. It's entirely plausible that just like toon abilities that are bugged, that drops have bugs at times. They aren't going to publish it and changes can happen at any time since item acquisition is server side.

I've tracked large samples in the game as well as worked with rand() creating data models for work in the past and drops don't seem to follow a simple probability with a simple rng. It's just so "streaky". I won't speculate as to if it's intentional or not but don't see why some find the notion that a company would use metrics to increase revenue inconceivable or even likely, like the game is some sort of social obligation. Drops are directly tied to revenue. It's no different that having 5 items in a pack as a possible outcome, but underneath have different probabilities associated with each based on revenue projections. Not only is it a freemium game, it follows a gambling model which his highly psychological.

@ExarTheKun the term you are looking for is "independent events", i.e. the previous outcome has no bearing on the next outcome like a roulette table.

aeaswen

Nikoms565 wrote: »

SWGOH; you will never find a more wretched hive of confirmation bias and gambler's fallacy. We must be cautious when we use actual facts and logic.

I have seen no facts posted here. Just blind faith by some in RNGesus.

AndrewIss

NicWester wrote: »

There was a REALLY BAD thread that got deleted but it brought up a good point: Folks don't understand numbers.

If a ship has a 44% dodge rate and you attack it 8 times, there is a 0.140482% chance of all eight attacks being dodged. That's one in a thousand. We don't know how many people play the game, but let's assume it's "only" 100,000. That means, on mathematical average, this happens to 100 people a day. In reality, the number of players is vastly higher than 100,000, so it's going to happen quite a bit throughout a day, but it will only happen to you once every few months.

So please don't make death threats because you don't understand that you aren't the only person playing this game. Even if it's a joke, it's a dumb joke and you're a dumb person. Go to a library and undumb yourself.

(PS: My numbers are probably wrong because I'm very bad at math. But I get the general idea anyway.)

Your math is right, but the law of large numbers doesn't have anything to do with this

freshprince19888

I feel a lot better now. Instead of feeling like I got screwed, now I feel like I am statistically predisposed to getting screwed.

Nikoms565

freshprince19888 wrote: »

I feel a lot better now. Instead of feeling like I got screwed, now I feel like I am statistically predisposed to getting screwed.

That's looking at the glass half full. Well done!

Carth_Onassi1973

freshprince19888 wrote: »

I feel a lot better now. Instead of feeling like I got screwed, now I feel like I am statistically predisposed to getting screwed.

That’s hysterical AND depressing at the same time.

Sawyerstooth

This game is designed to make money. Just like a car mechanic or a radish farmer, EA wants to make a profit. I’m f2p and I get that. Whatever math they use to set it up to entice players to spend is fine. Annoying sometimes, but fine.

snowhut

sneeekypants wrote: »

NicWester wrote: »

That's because what you said is nonsense. If it's a random number generator that's been set up to be streaky then it's not a random number generator any more. Moreover, if it's streaky then it isn't set up to reward spending money, it's set up to reward patience. Stop being so cynical, Diogenes.

This is an odd thread.

A few points to add:
-arithmetic mean is a very poor statistic used alone
-you can have induced variance while still using rng and maintain a particular average if that was their intent or the ending drop rate could be just the resulting average of a particular algorithm. Say set the probability to 10% for 100 samples, then 50% for the next 100. That would converge on 30% over the long run but give you very "streaky" results. Or you could even use an organic probability.
-There's bugs, lots of them. It's entirely plausible that just like toon abilities that are bugged, that drops have bugs at times. They aren't going to publish it and changes can happen at any time since item acquisition is server side.

I've tracked large samples in the game as well as worked with rand() creating data models for work in the past and drops don't seem to follow a simple probability with a simple rng. It's just so "streaky". I won't speculate as to if it's intentional or not but don't see why some find the notion that a company would use metrics to increase revenue inconceivable or even likely, like the game is some sort of social obligation. Drops are directly tied to revenue. It's no different that having 5 items in a pack as a possible outcome, but underneath have different probabilities associated with each based on revenue projections. Not only is it a freemium game, it follows a gambling model which his highly psychological.

@ExarTheKun the term you are looking for is "independent events", i.e. the previous outcome has no bearing on the next outcome like a roulette table.

Thanks.

NicWester

Handful of posts got deleted so it looks like I’m talking to myself for a bit there. I promise, though, there was a tin foil hat-wearing guy talking some real nonsense between each of those posts!

Jrarr

ExarTheKun wrote: »

I think that one pretty large issue with these types of statistics, is that people seem to believe one bad roll increases your odds of a decent roll the next time around.

For instance, let's say you have a 10% chance to recieve a full character from a chromium, rather than just shards. That doesn't mean nine failed attempts later, you are guaranteed a character on your next pack. You still have a 10% chance, even if you've gotten squat on your last thirty tries. RNG really is random.

Other EA games like Mass Effect Andromeda do guarantee you an ultra rare pull after so many failed attempts so it's not like it's unheard of.

Hakoda

NicWester wrote: »

Handful of posts got deleted so it looks like I’m talking to myself for a bit there. I promise, though, there was a tin foil hat-wearing guy talking some real nonsense between each of those posts!

What? Those things never happen here! Absurd!

Maegor

sneeekypants wrote: »

It's just so "streaky".

Ask any statistics professor to analyze lists created by human generated randomness vs computer generated randomness with the same p, and they will most likely be able to pick out the human generated list. Basically because humans avoid too much streakiness in probability, and when we see it in a computer generated event like this game, alarm bells go off, even if there isn't anything nefarious occurring.

NicWester

Hakoda wrote: »

NicWester wrote: »

Handful of posts got deleted so it looks like I’m talking to myself for a bit there. I promise, though, there was a tin foil hat-wearing guy talking some real nonsense between each of those posts!

What? Those things never happen here! Absurd!

Ehhhh...... In this case they should definitely have deleted those dumb posts