Great Vault, Random? No Way

[huth]

What do you mean 1 in 10^15? 140 000 still sounds like it'd take years... The one million is more like decades.

Yeah the 1 in 200 million was winning twice in a row. It was a small scratchie. Not a real lottery where just winning once is closer to one in ten million. That twice in a row gets to impossible levels.

Of course almost everyone spends more than they win with gambling. If winnings was bigger it'd be a charity and not gambling.

Depends on whether you mean winning the top prize or just winning anything. And 1 in 10,000,000 twice in a row is actually just 1 in 10^7*10^7 or 1 in 10^14, so another order of magnitude smaller. Many lotteries also allow multiple entries per player. Given that lotteries have existed for centuries, the odds aren't that bad that it happened repeatedly already.

[kukkamies]

Depends on whether you mean winning the top prize or just winning anything. And 1 in 10,000,000 twice in a row is actually just 1 in 10^7*10^7 or 1 in 10^14, so another order of magnitude smaller. Many lotteries also allow multiple entries per player. Given that lotteries have existed for centuries, the odds aren't that bad that it happened repeatedly already.

It does sound kinda possible. But haven't seen in any records that it has happened. Only twice overall but never in a row.

But yeah. The original 50 coin tosses and all heads doesn't sound probably at all for a single person. Not even if he kept doing it his whole life. Similar to generating crypto wallets. There is a miniscule chance you can generate another persons keys and can steal their coins. But it's so small it's not taken as a risk to crypto.

[Azerate]

The problem with understanding randomness is that it requires some knowledge, and internet discussions on topics that requires that are not doing well in general.

The base chance and the statistical probability of an event occuring are two different things. This is something that a lot of people get wrong and argue about. As a collector I mostly hear that in discussion about rare mount drops. For example if the drop rate is 1%, it is 1% whenever you do the action that potentially grants the reward. Based on that some people would argue that it is as likely to get such item on your first attempt as it is on your 1000th attempt, which is obviously not true. By making more and more attempts you're making it statistically more probable to get the result you seek, that's just how it works - simply because you have more attempts.

There are actually calculators on the internet for rng drops and giving you probability of getting such a drop. I just found this one by quick google search. You input the % chance and you get some trivia about probability, chance, likelihood etc. https://xplainthegame.com/dropchance-calculator/ So if something has 1% drop chance, at 229 attempts you have 90% statistical probability that it happens at least once. At 458 it's 99% which means that if you don't get a 1% drop by 458 try you are either extremely unlucky, or there is a hidden mechanic added to the rng and it's not just a simple 1 out of 100 roll.

Another issue. As some people here point out, randomness and probability created by software / coders is not really randomness, but pseudorandomness. This is well explained on the website dedicated to giving you truly random results (True Random Number Generator - check it out here https://www.random.org/) to quote the explanation:

Perhaps you have wondered how predictable machines like computers can generate randomness. In reality, most random numbers used in computer programs are pseudo-random, which means they are generated in a predictable fashion using a mathematical formula. This is fine for many purposes, but it may not be random in the way you expect if you're used to dice rolls and lottery drawings.

So yes, everyone arguing that the randomness in games is not real randomness is right. Randomness in games is done by formulas, and what's more, it can be easily augmented by additional values and modifiers.

[Skillslam]

Stitchflesh's Misplaced Signet 6 weeks in a row crew

[Shaetha]

I know how probability works, and yes - random is random.

The problem is that people don't fathom how randomness interacts with large sample sizes.

They assume that one person describing a rare event isn't random because they forget that there has been millions of experiments generating the output.

People need to head to random.org and check out how random works.

I hope that helps with your error.

Confirmation bias is fun.

[gutnbrg]

my only bad item was a slightly better item level of a trinket i already had, and it was a crappy trinket. Everything else has been good

also im prob a little older than u but if ur ever at a casino and ur down $2k and u finally get blackjack and u feel amazing and then the dealer gets black jack too, your mindset on rng and life will change drastically

[Shaetha]

It's possible in the same way winning the lottery twice in a row is yeah. Don't think anyone expects it to happen

The law of large numbers dictates it will happen and someone will bitch about it on a forum because they cannot accept they were just unlucky and want to spread hate because of it.

[Qnubi]

They do it on purpose, to prolong your stay in WoW. For example they could easily implement if you get Bracers XY you are illegible to get it again in the next vault week. They won't, claim it's "technical impossible" and majority of players will stop thinking about it. Also, when people ask for randomness, they only want good randomness. People hate randomness that appears broken. Any sequence of 1,1,1 etc. is perceived has faulty. The human mind does not want true randomness, it wants equal spread meaning that if you get 1 the next numbers can be anything BUT 1. Developers need to learn a thing or two from human psychology. True randomness is not desirable in games.

[froschhure]

Great vault should be renamed in triple timegate enhancer

[Shaetha]

-snip-

This is called the gambler's fallacy.

After 500 tries at a 1% chance it is very likely to have happened in the past, but it is no more likely to happen week 1 than it is to happen week 501. The chance of an event occurring doesn't change. The probability of any specific result, like the one mentioned in the OP, is no more likely than any other specific result, and the law of large numbers dictates that all possible results will happen eventually regardless of their probability.

Your chance of getting any specific random result is always the same. All you can do is try again, but no amount of attempts guarantees a desired result. In fact, "1" is the asymptote in any probability equation trying to calculate the chance a single specific result having occurred over any number of events. No matter how many times you've tried there is still a chance you never got it.

Also, a pseudorandom number generator is more than sufficient randomness for casino gaming regulations, so to mention them like it means anything to make the distinction with modern technology is being intellectually dishonest, and to say they're meaningfully different from dice rolls or lottery drawings, which also have things like mechanical flaws to make them not truly random, is just trying to distract from the point.

Although, I would be more likely to believe the dishonestly of a few people on a place people go to piss and moan than anything else.

- - - Updated - - -

True randomness is not desirable in games.

Correction: True randomness is not desirable in games for the MMO-Champion user that has called themselves Qnubi.

The vast majority of people don't think that much about it.

[Qnubi]

Correction: True randomness is not desirable in games for the MMO-Champion user that has called themselves Qnubi.

The vast majority of people don't think that much about it.

Psychology disagrees with you. Humans perceive sequences such as 1,1,1 etc. in a scenario with 10 variables as not random but will feel "unlucky", when it's not unlucky and simply random. Your psychology is not different to other humans, you will feel unlucky as well, denying your human nature is not gonna change that.

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5933241/

Your source: Your bias.

Knowing that we can't shut down this psychological effect we can simply diminish the effect it has on our-selves. Henceforth I mentioned that game devs can learn a thing or two from proven psychological facts.

[Azerate]

This is called the gambler's fallacy.

No it's not. It has nothing to do with anything I wrote, but people who have problems understanding basic concepts often attempt to claim statistical probability somehow does not exist and that it's instead gambler's fallacy.

Yes, no amount of attempts "guarantees" getting the result, but the more attempts are made, the higher the probability of eventually getting the successful hit. Open the first link I posted and read, if you care about actually correcting your misconceptions. Or don't, if you don't care. It's whatever to me.

[huth]

No it's not. It has nothing to do with anything I wrote, but people who have problems understanding basic concepts often attempt to claim statistical probability somehow does not exist and that it's instead gambler's fallacy.

Yes, no amount of attempts "guarantees" getting the result, but the more attempts are made, the higher the probability of eventually getting the successful hit. Open the first link I posted and read, if you care about actually correcting your misconceptions. Or don't, if you don't care. It's whatever to me.

Actually, you are using the gambler's fallacy, though it's likely more a case of poor wording.

Based on that some people would argue that it is as likely to get such item on your first attempt as it is on your 1000th attempt, which is obviously not true.

This is in fact incorrect; what you should have said is "in 1000 attempts". The likelyhood on the 1000th attempt is actually the same, since then you're treating them as independent events.

[Kalisandra]

Then there is a matter of fact that WoW is a piece of software. Software is run on a computer. Computers are unable to generate random numbers. In Computer Science we use instead pseudorandom numbers and just call them random. They are just a string of numbers based on a function where each previous one is feeded to next one and the resulting string of numbers have characteristics of randomness. And there are statistical ways to measure how random string of numbers is. One flaw in this is that this string of numbers will always be the same. To avoid that random number generators are usually initilized with current time in miliseconds.

If you're serious about your random numbers on a computer you can add a truly random input from an external random event, such as radioactive decay. Whether you use that for a random seed number or the random numbers themselves depends on how serious you are about your randomness (and how much money you want to spend, of course).

[Temp name]

He has been corrected many times by different people. In the very first page, he makes the same ignorant claim that rolling same numbers has the same probability of rolling different each time. That was 15 pages ago. I'd stop trying to explain, he probably isn't capable of grasping basic probability.

I'll admit to not really paying attention to the other responses. I've tried now at least

[Vampyr78]

If you're serious about your random numbers on a computer you can add a truly random input from an external random event, such as radioactive decay. Whether you use that for a random seed number or the random numbers themselves depends on how serious you are about your randomness (and how much money you want to spend, of course).

I know but I doubt Blizzard would use radioactive devay of some element to generete random numbers. Also those events might only look random. It is possible that we don't understand quantum physics well enough and that makes some things appear random.

[kukkamies]

The law of large numbers dictates it will happen and someone will bitch about it on a forum because they cannot accept they were just unlucky and want to spread hate because of it.

Law of large numbers doesn't dictate it will happen. It means many observations will bring the result closer to mean. Doesn't say anything about if any extreme situation will or will not happen.

I can believe OP's can happen with some really bad luck, still does feel strange. I think what's more likely is that OP remembers some weeks wrong

[kukkamies]

That's if only ONE person was doing 140,000 weeks worth of tries, which obviously isn't possible.

But with 140,000 people doing one week's worth of tries, we effectively have 140,000 attempts made - And given that WoW has millions of players, we can safely say that we passed over 1 million great vault rolls a LONG time ago.

This is what people mean by "the law of large numbers" applying here. One player might only see a handful of results and try to draw conclusions based on those results - But if we look at every player's results (since every player has the same RNG) we can easily see that the law of large numbers is precisely why OP feels so "off" about his results.

Yes, his results are unlikely. HIGHLY unlikely. If we only had his results to go by, maybe we'd have reason to believe he's actually got a point about the Great Vault's randomness.

But when there's hundreds of thousands of attempts (at minimum) a week, and dozens of weeks to go by - One in a million suddenly isn't so big, considering we've had multiple-millions of rolls against the Great Vault table. And given that not everyone is having this one-in-a-million roll, it's a little easier to understand OP as being that one in a million, amongst millions who DIDN'T get that result.

This 140 000 doesn't have anything to do with WoW. And it means 140,000 per person on Earth and 1 in 10^15 chance altogether. It was about 50 coin tosses in a row that are all same side.

Law of large numbers has NOTHING to do with will something happen or not. It only means that result will get closer to average as more attempts are made. It's one of those big words people try to use to make their thoughts sound official, but in reality it has nothing to do with these rare cases.

Of course one in a million isn't so big for millions of people. 10^15 is 1 000 000 000 000 000, lose one zero or not. Even with billions of people it's not frequent.

If we think any individual item is 1/10 then op had one any and two same = 1/10^2. And that happened 7 times ^7 which is 1/10^2^7= 1/10^14. It's pretty close to that 50 tosses actually. Even with 10 million players it wouldn't happen.
My guess is that he doesn't remember it correctly. Blizzard could have fucked up bad luck protection again but more likely is OP remembers wrong.

[Nyxblow]

Making sure numbers are random doesn't mean memorization. For rand, it requires an arbitrary seed (doesn't need to be random, most of the time, it's system time). rand() will not roll same set of numbers, even if you roll it once a week so long as its properly seeded. And we are talking about the crappiest RNG. Looks like that post has gone well over your head yet you dare to tell me I should know my shit. Your initial sentence is a facepalm material, I have been rolling the same 3 numbers for months (bracer, gloves, back).

Anyone with a half brain who did bit of programming would tell you this isn't RNG.

Dude... how naive can you get. You are taking the assumption that your roll from last week happens using the exact same seeded random number generator, on the same server, on the same process, than the one used to generate your vault this week. And that of course it was kept in memory just for your character so that it can continue the next sequence number.

And don't get me wrong, I do think that the Great Vault is not random. But because Blizzard hardcoded some items to be more rare than others, not because of some flawed basic understanding of how video game servers work.

[huth]

If we think any individual item is 1/10 then op had one any and two same = 1/10^2. And that happened 7 times ^7 which is 1/10^2^7= 1/10^14. It's pretty close to that 50 tosses actually. Even with 10 million players it wouldn't happen.
My guess is that he doesn't remember it correctly. Blizzard could have fucked up bad luck protection again but more likely is OP remembers wrong.

That's a full order of magnitude difference. That's not "close". That'd also be for 7 repeats (aka 7 weeks in a row), we're talking something like 2-3 here. You're blowing things way out of proportion.

Though yes, misremembering is the most likely culprit, with how rare these reports are.