Question about so-called "Luck Accounts"

[kharmin.7683]

1 hour ago, mtpelion.4562 said:

Every INDIVIDUAL roll has the same chance of getting a particular number.

However, when I roll a dice 100 times and you only roll it once, I have more chances than you do, so my odds of getting any particular number is higher. 

But you are still constrained by each, individual roll.  If you roll 1-100 twenty times, each individual roll has a 1% chance of rolling a particular number.  You would have twenty chances to roll a number, which would still be 1% per roll -- the same as my 1% for only rolling once.

But again, I'm no expert in these matters.  /shrug

[mtpelion.4562]

25 minutes ago, kharmin.7683 said:

But you are still constrained by each, individual roll.  If you roll 1-100 twenty times, each individual roll has a 1% chance of rolling a particular number.  You would have twenty chances to roll a number, which would still be 1% per roll -- the same as my 1% for only rolling once.

But again, I'm no expert in these matters.  /shrug

It has been awhile since I last did these calculations, but if you roll a 1-100 once, trying to get a 100, you have a 1% chance to eventually succeed across all your attempts.

 

If I roll a 1-100 twenty times, still trying to get a 100, I will have a 18.21% chance to succeed across all my attempts. 

 

The individual rolls still each have a 1% chance to succeed, but I (not the rolls, but me the person doing the rolls) have an aggregate chance to succeed that is higher, because I had more rolls than you.

 

You will note that the odds are not additive. I don't have a 20% chance because each time I roll, the result is independent of the prior results. The "loss" of 1.79% represents the inefficiency of "starting over" each time you roll.

Edited November 23, 2021 by mtpelion.4562

[Kurrilino.2706]

1 hour ago, kharmin.7683 said:

But you are still constrained by each, individual roll.  If you roll 1-100 twenty times, each individual roll has a 1% chance of rolling a particular number.  You would have twenty chances to roll a number, which would still be 1% per roll -- the same as my 1% for only rolling once.

But again, I'm no expert in these matters.  /shrug

 

I'm pretty sure the argument is if you have 1% chance and you roll 100 times you should by statistic have at least 1 hit.

Which of course doesn't apply in reality

[Astralporing.1957]

31 minutes ago, Kurrilino.2706 said:

 

I'm pretty sure the argument is if you have 1% chance and you roll 100 times you should by statistic have at least 1 hit.

Which of course doesn't apply in reality

No. The argument is that if you have 1% chance per roll, and roll more times than 1, your chance increases. At 100 rolls it actually will be around 65%, if you're wondering. Which is not 100%, of course, but is still way, way higher than 1%.

[Random Wax Orc.7695]

4 hours ago, Sobx.1758 said:

You were using a comparison with no direct correlation to the discussed situation (due to not having the access to the same informations between the two). Astral responded with trying to explain to you exactly that, translating your example into the situation that's discussed here ("You don't see the ball. You don't know when it was released. You don't even know the order of the pockets on the wheel. You cannot predict the end result because the starting point is, in practice, pretty much random (oh, wait, "beyond calculation", is it?)."), to which you respond with "The trope of using a mobile phone app to calculate probable roulette outcomes is common", seemingly still stuck in discussing that literal roulette situation, which still doesn't have a direct translation to what's being discussed here.

Lucky zones in roulette is a bit like lucky zones in a series of random numbers.  It doesn't matter if you can predict them or not.  Astral didn't ask what I meant, but simply attacked.  In roulette you can understand why there are lucky zones, and that is what made it a reasonable example - to illustrate a concept.  If you think it is a flawed metaphor, then fine, but it still has some utility.  All the rest is just two movies on one screen.

 

Quote

If you don't understand what you're doing here, then not sure if you're in place of talking about discussing anything in good faith. And yes, I know what's the context of this comment chain -it's just that you're not discussing anything within the realms of the game, but instead took a hard turn into literally discussing literal late bets on literal physical roulette. And, agian, this just doesn't translate to what is being discussed here.

Astral said you can't see the ball in roulette, not a metaphorical ball in a computer.  That seemed literal and I made a reasonable response that.  I wasn't arguing but trying to convey a concept.  You are not trying to understand; you are trying to argue... that is the bad faith.  I didn't enter an argument; I tried to enter a discussion.  Hopefully someone somewhere can take away from this is a slightly larger perspective on such issues.

 

Quote

As a reminder, it more-or-less went on from the claim that having "late rolls due to hardware differences could meaningfully influence your chances of getting rare drop". But in reality it doesn't, because in the end you don't know when exactly you roll, you don't know what the potential roll queue is, you don't know what is the next outcome, nor do you know which "square" is the winning one. Even if it influences the results, it influencesthem without a direct possibility to meddle with said result, so you're just arguing aboout random shots being taken at random times and why that's unfair. It's not unfair, since you don't knowingly influence the chances with your hardware, you're just flowing with the current anyways.

Meaningfully influence does not mean you have any knowledge or control.  Injecting this meddling narrative seems like shifting the goal posts.  In my first post, I directly said "Anyway, there is nothing you can do about it. [...] It is simply an academic issue... that essentially everyone strawmans one way or the other, based on their biases."  I don't see how one can take that and get to being able to intentionally manipulate outcomes.
Just because there are unknowns doesn't make it random.  Just because it is beyond your control doesn't make it random. Just because you can't determine your effect on it doesn't make it random.  There is a strong possibility that the distribution of individual player results does not conform to a random distribution.  Patterns of results are not necessarily more or less fair than random results.  The Universe if full of emergent patterns in complex systems.  It is reasonable to suspect there might be emergent patterns in an MMO.

Edited November 24, 2021 by Random Wax Orc.7695

[mythical.6315]

4 hours ago, kharmin.7683 said:

But you are still constrained by each, individual roll.  If you roll 1-100 twenty times, each individual roll has a 1% chance of rolling a particular number.  You would have twenty chances to roll a number, which would still be 1% per roll -- the same as my 1% for only rolling once.

But again, I'm no expert in these matters.  /shrug

This should help. This is something that trips up many people. 
 

 

[Astralporing.1957]

8 hours ago, Random Wax Orc.7695 said:

Lucky zones in roulette is a bit like lucky zones in a series of random numbers.  It doesn't matter if you can predict them or not.  Astral didn't ask what I meant, but simply attacked. 

Oh, i actually did understand what you meant. Probably way better than you did. Which is exactly why i pointed out the problem with your example.

Hint: the existence of "lucky zones" has absolutely no impact on individual accounts.

 

8 hours ago, Random Wax Orc.7695 said:

Astral said you can't see the ball in roulette, not a metaphorical ball in a computer. 

No, i was talking about the in-computer RNG system. In case of roulettes you can try to "game" the system placing late bets based on what you see. In case of game system you cannot do that, because the whole system is completely invisible to whomever is playing that "roulette, MMORPG version"

8 hours ago, Random Wax Orc.7695 said:

That seemed literal and I made a reasonable response that.  I wasn't arguing but trying to convey a concept.  You are not trying to understand; you are trying to argue... that is the bad faith.

In context of you clearly not trying to understand what others tried to explain to you, that's quite ironic, don't you think?

 

I will try to explain in very simple way where the problem in your reasoning is.

Let's use, for illustration, your examples of "Slow Ron", "Fast Harry", and do the comparison with roulette

Let's just say, that Harry's results fall within the +4 to +12 pockets range. Ron's within +10 to +30 range.

You seem to think, that this will give an advantage to one of them, because one of those ranges will contain better potential set of values than the other. There are two problems with this.

First, the starting point for Harry and Ron (the "zero" point) is not the same. It's X+4 to X+12 versus Y+10 to Y+30. Second, not only the starting point is not the same, but at the moment each of them is rolling the values in the pockets on roulette board may be different (as the RNG thread might have already been restarted with different starting seed).

So, at one specific situation, this might give an advantage to Harry - but next time that advantage might shift to Ron. So, while in theory the system might introduce a bias, the bias itself will be shifting and changing constantly, becoming in practice just another random element in the mix.

[Random Wax Orc.7695]

1 hour ago, Astralporing.1957 said:

I will try to explain in very simple way where the problem in your reasoning is.

Let's use, for illustration, your examples of "Slow Ron", "Fast Harry", and do the comparison with roulette

Let's just say, that Harry's results fall within the +4 to +12 pockets range. Ron's within +10 to +30 range.

You seem to think, that this will give an advantage to one of them, because one of those ranges will contain better potential set of values than the other. There are two problems with this.

First, the starting point for Harry and Ron (the "zero" point) is not the same. It's X+4 to X+12 versus Y+10 to Y+30. Second, not only the starting point is not the same, but at the moment each of them is rolling the values in the pockets on roulette board may be different (as the RNG thread might have already been restarted with different starting seed).

So, at one specific situation, this might give an advantage to Harry - but next time that advantage might shift to Ron. So, while in theory the system might introduce a bias, the bias itself will be shifting and changing constantly, becoming in practice just another random element in the mix.

So we agree that the system might introduce bias, and that bias might be shifting and changing constantly, but I disagree it is just another random element in the mix when dealing with extremely rare outcomes.

 

This is a good description.  However, the starting points are not necessarily independent, and might even be identical simply because they are playing in the same roulette event (we don't have access to the code).  Also, all possible threads are not represented (PRNG) so an arbitrary block of 8 numbers in the thread and a block of 20 numbers in the thread will not reliably have the same typical result (even if they average out over many plays).  There might be some overpopulation of repeated high or low numbers in a larger range that is not apparent in a smaller range.  The RNG thread is also more likely to get restarted in the middle of Ron's larger range which could further complicate things (again we don't have access to the code).  Now if they both instantly repeat their play when they get a result.  It's X+4 to X +12 and 2X+8 to 2X+24 versus Y+10 to Y+30 and 2Y+20 to 2Y+60.  Harry's second play has a range of 8 within a range of 16 and Ron's second play has a range of 20 within a range of 40.  So the variability of the Harry and Ron outcomes are diverging.  They both have different ranges and different patterns of ranges that are being fed from threads.  If Harry and Ron make all possible plays (every offset for every thread), it is extremely unlikely they will get the same number of double wins.  As the number of plays is increased, the bias that in practice was a random element for simple situations, becomes the source of a non-random pattern in as the situation increases complexity.  It isn't the only source.  The bias is clearly just one small pattern among many, but these patterns interact and create more patterns.

Is that clear?  That is why I biased against tiered loot tables (the multiple RNG calls are not strictly independent), and am biased towards RNG created by white noise generators rather than PRNG (the randomness of white noise will erase other patterns).

 

Again, it is like the Game of Life.  Take objects and rules, and let them interact with one another in increasing complexity, and you can get patterns and behaviors that were not part of the initial set of the objects and rules.  Again, this should be interesting, but it not actionable... beyond the advice to use white noise generators if you create an RNG heavy MMO.

[Astralporing.1957]

1 hour ago, Random Wax Orc.7695 said:

So we agree that the system might introduce bias, and that bias might be shifting and changing constantly, but I disagree it is just another random element in the mix when dealing with extremely rare outcomes.

If it affects the very same people differently every time, and there's no real pattern to that bias (because it gets affected by many things outside the whole PRNG generation thread, at last some of whose are, for all practical reasons, random), then we cannot treat that factor any other way than as a random element.

1 hour ago, Random Wax Orc.7695 said:

However, the starting points are not necessarily independent, and might even be identical simply because they are playing in the same roulette event (we don't have access to the code). 

If they use the same RNG thread, they cannot be the same. As you have already noticed, there's a queue, so one will have a starting point past the result of the other. Which one? Well, that in itself might be random (or to be more precise, influenced by non-constant factors outside of the game).

 

1 hour ago, Random Wax Orc.7695 said:

Also, all possible threads are not represented (PRNG) so an arbitrary block of 8 numbers in the thread and a block of 20 numbers in the thread will not reliably have the same typical result (even if they average out over many plays). There might be some overpopulation of repeated high or low numbers in a larger range that is not apparent in a smaller range. 

True. But then the same is also true in a block of numbers generated by a "true RNG" generator. This is completely normal.

 

1 hour ago, Random Wax Orc.7695 said:

The RNG thread is also more likely to get restarted in the middle of Ron's larger range which could further complicate things (again we don't have access to the code).

it would complicate things for you only. Because you'd have to explain how that "reset" might introduce a bias towards one or the other. Which i have noticed you cleverly didn't do.

1 hour ago, Random Wax Orc.7695 said:

Now if they both instantly repeat their play when they get a result.  It's X+4 to X +12 and 2X+8 to 2X+24 versus Y+10 to Y+30 and 2Y+20 to 2Y+60. 

....okay, i give up. That's the point i have noticed you have no idea what you're talking about. Now you're just making up stuff completely, and, moreover, it doesn;t make sense. And you don;t even notice that.

Please, stop, that's painful to watch.

(hint: no, it does not work like that.)

 

1 hour ago, Random Wax Orc.7695 said:

Harry's second play has a range of 8 within a range of 16 and Ron's second play has a range of 20 within a range of 40. 

No. By the time the "immediate repeat play" came along, the number of positions has already shifted - and you have no idea by how much. Now the starting positions are I and J

(hint: that shift i was talking about is in no way affected by accounts of Ron and Harry)

Notice also, that their second ranges will now be different, because those ranges i mentioned were for that single roll only. Now the ranges are I+13 to I+41 (Harry's little kid started a torrent tracker right now), and  Ron's is J+12 to J+22.

1 hour ago, Random Wax Orc.7695 said:

If Harry and Ron make all possible plays (every offset for every thread), it is extremely unlikely they will get the same number of double wins. 

Again, that's the same for True RNG.

 

1 hour ago, Random Wax Orc.7695 said:

As the number of plays is increased, the bias that in practice was a random element for simple situations, becomes the source of a non-random pattern in as the situation increases complexity. 

Except, as i pointed out, the bias changes every play. In ways that are not part of the pattern.

1 hour ago, Random Wax Orc.7695 said:

It isn't the only source.  The bias is clearly just one small pattern among many, but these patterns interact and create more patterns.

Is that clear? 

Yes, it's clear. That you have no idea what you're talking about.

That's my last response to this, because i don't see any point to try to educate you more.

[Random Wax Orc.7695]

1 hour ago, Astralporing.1957 said:

If it affects the very same people differently every time, and there's no real pattern to that bias.

You can't simultaneously have a pattern and have it not be real.  "It is not practical therefore it is random" is nonsense.  There are patterns to PRNG, there are patterns to the calls to PRNG (those are called computer code ), there are patterns all the way down.  You can't quantify how those patterns interact so you claim the patterns can't interact, and everything is just magically random.  The entire structure of Universe disagrees with that conclusion.

1 hour ago, Astralporing.1957 said:

True. But then the same is also true in a block of numbers generated by a "true RNG" generator.

No it isn't, because there would be no pattern to repeat in a true RNG generator.  A true RNG generator would maintain a blank canvas.  If you have a set of true random numbers, then no pattern in how the numbers are taken from that set would matter, as long as that set isn't reused (since the reuse would not be random).  That is extremely basic probability. 

1 hour ago, Astralporing.1957 said:

Except, as i pointed out, the bias changes every play. In ways that are not part of the pattern.

In ways that are not part of just that one pattern.  You are incorrectly demanding that one pattern imposes one structure on a complex system or else you declare it to be random.  Most patterns in reality are not imposed, but emergent.  Complex systems have many non-random inputs and have patterns in outputs that are affected by all inputs in non-random ways.  Just because you refuse to wrap your head around it doesn't make it random.

1 hour ago, Astralporing.1957 said:

Yes, it's clear. That you have no idea what you're talking about.

That's my last response to this, because i don't see any point to try to educate you more.

Yes, it's clear.  That you have no idea what you are talking about.  You are however projecting your biases and have been downright rude since your first reply.  From the beginning you set out to put me in my place, without attempting to understand what I was describing.  Maybe my communication skills don't mesh with yours, but since you default to trying to contest instead of seeking to clarify, that's all on you.  I've tried to explain the same thing multiple ways, and you just keep proving it was a wasted effort.  Bad faith.

 

To everyone else, I ask one question.  If Skynet played GW2, how long would it take for it to figure out how to be able to reliably get the rarest drops from every blacklion chest and loot bag.  Right, everyone knows an arbitrarily smart AI could do it, because it is not actually random.  Skynet could suss out all the relevant factors, all the patterns, and game the system.

Skynet could see the patterns we can't... from the start I've merely been saying the patterns are there and they are not random.  Not controversial at all.

[Astralporing.1957]

1 hour ago, Random Wax Orc.7695 said:

To everyone else, I ask one question.  If Skynet played GW2, how long would it take for it to figure out how to be able to reliably get the rarest drops from every blacklion chest and loot bag. 

According to THIRDPARTYSOFTWAREREADME.txt GW2 uses a RNG generation code with a period ("pattern") with length of 2^19937-1. For your information, 2^60 is already greater than the estimated number of seconds that happened since the beginning of the universe.

That should answer your question rather well, don't you think?

And now i am really out.

Edited November 24, 2021 by Astralporing.1957

[robertthebard.8150]

So I see two problems with this:

1.  Are the "lucky" accounts lucky every time?  If not, why not?

2.  Skynet couldn't manipulate things that are rolled outside of what it's supposed to be able to influence, w/out having their account banned.

This means that yes, you could, theoretically, cheat the system, but the key word here is "cheat".  I get the urge to point to other players getting things more consistently and wanting to attribute it to anything other than dumb luck, the RNG gods have hated me since they were created.  That doesn't mean it's accurate though, and it surely doesn't mean that the odds have been tipped in their favor by a game dev team.

 

My legion in Aion had a running joke that I could miss an item with a 100% drop rate, my luck at RNG is so bad.

[Random Wax Orc.7695]

1 hour ago, Astralporing.1957 said:

According to THIRDPARTYSOFTWAREREADME.txt GW2 uses a RNG generation code with a period ("pattern") with length of 2^19937-1. For your information, 2^60 is already greater than the estimated number of seconds that happened since the beginning of the universe.

That should answer your question rather well, don't you think?

Why would anyone think that an out of context bit of information answers a question.  Great, it is a big pattern.  The only thing that confirms is it is not random.  It doesn't confirm the pattern is accessed in non-repeating manner, or much of anything else.  The length of the pattern is too large to be indexed, so using the entire pattern would be quite an accomplishment.  And if the entire pattern isn't being used, then the length of the pattern is irrelevant.

1 hour ago, robertthebard.8150 said:

So I see two problems with this:

1.  Are the "lucky" accounts lucky every time?  If not, why not?

Why would it be?  Imagine an account has a 1 in 100,000 chance of being 2,000 times more fortunate in 1 in 1,000,000 chance.  One in one hundred thousand is not very likely, but that 1 in 1,000,000 outcome becomes a 1 in 500 outcome every time that 1 in 100,000 chance wins.  Can something like this happen?  Theoretically, yes, but it would be much more complicated.  We also wouldn't have the data or capacity to track anything like that.  Any patterns would be multivariant, probably dynamic and almost certainly incomprehensible.  You wouldn't know the variables, and you wouldn't know why the "luck" would begin or end.

1 hour ago, robertthebard.8150 said:

2.  Skynet couldn't manipulate things that are rolled outside of what it's supposed to be able to influence, w/out having their account banned.

This means that yes, you could, theoretically, cheat the system, but the key word here is "cheat".  I get the urge to point to other players getting things more consistently and wanting to attribute it to anything other than dumb luck, the RNG gods have hated me since they were created.  That doesn't mean it's accurate though, and it surely doesn't mean that the odds have been tipped in their favor by a game dev team.

Well, Skynet is a bot which is a bannable offense... but I personally wouldn't want to anger Skynet.

Anyway, if there are anomalies in the probability distribution across players, it certainly wouldn't be something intentionally done by Anet.  It would be unintentional artifacts within a complex system.

1 hour ago, robertthebard.8150 said:

My legion in Aion had a running joke that I could miss an item with a 100% drop rate, my luck at RNG is so bad.

I know someone who repeatedly got lucky drops the week after buying some black lion keys.  Complaints about bad luck after that were humorously attributed to not having bought black lion keys.  (buying black lion keys is not coded to increase your luck.)

[Leo.3428]

4 hours ago, Astralporing.1957 said:

According to THIRDPARTYSOFTWAREREADME.txt GW2 uses a RNG generation code with a period ("pattern") with length of 2^19937-1.

I'm trying to find it in the 10,846 lines of the read-me, no luck (no pun intended) so far with all kinds of keywords (EDIT: found it thanks to Khisanth's post that I had missed, it is mt19937ar in the file). Is it the Mersenne Twister, that is integrated into many programming languages or their standard libraries?

The Wikipedia article left me a little bit worried as it states that "observing a sufficient number of iterations (624 in the case of MT19937, since this is the size of the state vector from which future iterations are produced) allows one to predict all future iterations." Only a little bit worried, because no one from one game client can observe the core sequences of iterations.

That said, @Random Wax Orc.7695, I'm sticking to your advice of using white noise RNG, which however left me a little bit worried too because non-random terrestrial radio waves seem to slip into these. And I'm sure quantum randomness will be cracked one day too, so I have no hope for true randomness. I don't even need it, why am I asking all these questions?

Edited November 25, 2021 by Leo.3428

[Astralporing.1957]

7 minutes ago, Leo.3428 said:

I'm trying to find it in the 10,846 lines of the read-me, no luck (no pun intended) so far with all kinds of keywords. Is it the Mersenne Twister, that is integrated into many programming languages or their standard libraries?

Yes, indeed, it's the MT19937. The relevant part of the THIRDPARTYSOFTWAREREADME.txt has been already posted by someone few pages back in this thread.

7 minutes ago, Leo.3428 said:

That said, @Random Wax Orc.7695, I'm sticking to your advice of using white noise RNG, which however left me a little bit worried too because non-random terrestrial radio waves seem to slip into these. And I'm sure quantum randomness will be cracked one day too, so I have no hope for true randomness. I don't even need it, why am I asking all these questions?

It's even worse than that - remember, that white noise is something that can be completely artificially generated and doesn't have to include any random elements.

[Leo.3428]

4 minutes ago, Astralporing.1957 said:

Yes, indeed, it's the MT19937. The relevant part of the THIRDPARTYSOFTWAREREADME.txt has been already posted by someone few pages back in this thread.

Thanks, I found it back after posting, and edited my post but you were faster in replying.

  

4 minutes ago, Astralporing.1957 said:

It's even worse than that - remember, that white noise is something that can be completely artificially generated and doesn't have to include any random elements.

Ha, more research to do, for something I don't need. Curiosity killed the cat...

[Tytos.3076]

On 11/22/2021 at 5:21 PM, mythical.6315 said:

How are you measuring this “unevenness”?  Do you have access to some data which we do not have?  Threads like these are just another case of people not understanding randomness. 

Randomness is uneven by definition. The shoe's on the other foot, bucko.

[mythical.6315]

8 hours ago, Tytos.3076 said:

Randomness is uneven by definition. The shoe's on the other foot, bucko.

If that’s the case then you can provide a source to back that up.  Every account has the same chance of being as lucky, or unlucky, as another. 

Edited November 26, 2021 by mythical.6315

[Astralporing.1957]

7 hours ago, mythical.6315 said:

If that’s the case then you can provide a source to back that up. 

You might be misunderstanding this. I think, that what @Tytos.3076tries to say is that randomness, just by working as intended, creates a situation where some people can end up less lucky than others. Notice, that it does not mean the game favours some accounts over the others, but rather that this is an inherent feature of any random process where you can "rank" potential results as "better" or "worse".

As such (again, as i understand that original post), @Tytos.3076proposes to introduce some mechanisms that would "help out" those that would end up being "the losers".

While the latter suggestion can be argued about, i don't see what could be controversial about the former assumption and why someone would have to go out of their way in order to "prove" it. It's just how randomness works - if it was guaranteed to produce even results (even when pulling those from a pool of definitely not even possibilities), it would not be random anymore.

I just saw your edit:

7 hours ago, mythical.6315 said:

Every account has the same chance of being as lucky, or unlucky, as another. 

I'm quite sure that @Tytos.3076was not talking about unevenness of chances, but of results. Or at least it seems so from the context.

Edited November 26, 2021 by Astralporing.1957

[robertthebard.8150]

5 hours ago, Astralporing.1957 said:

You might be misunderstanding this. I think, that what @Tytos.3076tries to say is that randomness, just by working as intended, creates a situation where some people can end up less lucky than others. Notice, that it does not mean the game favours some accounts over the others, but rather that this is an inherent feature of any random process where you can "rank" potential results as "better" or "worse".

As such (again, as i understand that original post), @Tytos.3076proposes to introduce some mechanisms that would "help out" those that would end up being "the losers".

While the latter suggestion can be argued about, i don't see what could be controversial about the former assumption and why someone would have to go out of their way in order to "prove" it. It's just how randomness works - if it was guaranteed to produce even results (even when pulling those from a pool of definitely not even possibilities), it would not be random anymore.

I just saw your edit:

I'm quite sure that @Tytos.3076was not talking about unevenness of chances, but of results. Or at least it seems so from the context.

The thing is, the "ritual" I posted in my initial post in this thread dates back to 2005ish.  These superstitions go back to the introduction of RNG, and it has been stated in this thread, and in countless others just like it since 2005, that there is a background "luck stat" that is applied on account creation.  This is nothing new, just as "you must do x in order to get y" superstitions.  At the end of the day, someone will try that ritual, and get lucky on their roll, and will sincerely believe that the ritual made an actual difference.  It's the nature of the internet.

[Astralporing.1957]

3 minutes ago, robertthebard.8150 said:

The thing is, the "ritual" I posted in my initial post in this thread dates back to 2005ish.  These superstitions go back to the introduction of RNG, and it has been stated in this thread, and in countless others just like it since 2005, that there is a background "luck stat" that is applied on account creation.  This is nothing new, just as "you must do x in order to get y" superstitions.  At the end of the day, someone will try that ritual, and get lucky on their roll, and will sincerely believe that the ritual made an actual difference.  It's the nature of the internet.

Are you sure you quoted the right person? Because i definitely wasn't talking about this at all.

[William.3871]

On 11/15/2021 at 3:27 PM, kharmin.7683 said:

No exaggeration here.  8+ years.  Not one precursor or other unique drop.  😞

Me too. Almost 1500 h, with big times off between them, played since day 1.

The most expensive thing I remember dropped to me was a hammer that was value 60 gold back in the day. I gave it to my brother. Now it only cost 5 gold xD

[Kurrilino.2706]

I can't believe this is still going on.

According very early information, the chance of a Mystic Forge drop was 0.3%

I assume open world would be even less then.

 

It's absolutely normal that people don't get a precursor drop.

Regardless of people playing 1000's of hours or kill millions of mops.

Those numbers are not in the equation, which make them pointless.

 

I have seen someone getting the warhorn precursor just walking out of the tutorial killing a centaur.

It's pure luck and the odds are highly against us.

If precursors are so important to you, just buy them 

[robertthebard.8150]

10 hours ago, Astralporing.1957 said:

Are you sure you quoted the right person? Because i definitely wasn't talking about this at all.

I was replying to the second quote in your post, about what you expect the poster was talking about.  I suspect that we wouldn't have this thread, if it wasn't about results, as opposed to chances to get something.

[ardhikaizecson.3697]

There is this one person in my static raid who always got an ascended or unique drop at least once in each of the raid wings every week. So yeah I believe there's a supposedly luck account existing because it is only him that always got a lucky drop. First dhuum kill? Got the bench of final judgment