Guild Wars 2

The real question for the devs is, should players have negative outlier accounts?

We all know that statistically, negative outliers would exist, players who just tend to get terrible luck. It’s not something wrong with the system, it’s just how statistics works, at least some people will be very lucky, and some will be very unlucky in a perfectly fair system.

But they devs don’t have to allow that. They always have the option of cheating in the players’ favor. They always have the option of including streak breakers and other mechanisms that track bad luck and offset it with bonus rewards. Should they?

What you all want is not true randomness, it is equal distribution instead.

True Randomness has no memory, so whenever the rng is rolled, it has no knowledge of what happened before. If it rolls the dice for a drop, it doesn’t know if the character already received 5 precursors as drop and should be ineligible to get another one until every other character in the world also got 5 precursors.
This is fair in a mathematical sense, because for every roll in the whole game, every player has the same chance of success. All the time.
This is considered unfair at the same time, because without memory the dice may roll 2 successes for the same player and omit another player. The rng doesn’t remember that one one player already got his success and should be ineligible for another.

Instead, you want equal distribution of good rng results. You want that every player get equally good results within a time frame like any other player. If there are 5 precursor drops, they should not go all to the same player, they should all go to 5 different players.
This requires memory! You have to manipulate the previously random result. The rng has to check if a candidate already got his success, and if he actually got it, it should not give him one again.

This is unfair in a mathematical sense, because the probability of a success depends on previous rolls. The more successes you got in the past, the less successes you get in the future. Every player has his own drop probability, not equal to that of every other player.
This is considered fair at the same time, because it ensures that within a time frame every player gets his equal share of drop successes.

So true randomness within drop results is not as desirable as it seems to be. Players want a manipulated rng. A rng that smooths out too many losses and too many successes, so in the end everyone gets the same share of success within his play time. A true rng smooths out within infinite play time, but the player wants that a game should smooth it out within a much shorter time – his play time, which is not infinity but only some 100 or some 1000 hours.

Yes, definitely, accounts have an associated RNG, better in some and worst in others.

RedBaron.6058:

Yes, definitely, accounts have an associated RNG, better in some and worst in others.

But only insofar as the Luck Stat is concerned. Two accounts with identical Luck Points have identical RNGs

Ayrilana.1396:

Nash.2681:

John Smith.4610:

Every account has the exact same chance on RNG.

Granted and accepted. But this doesn’t explain why you guys still stick to this kitten concept. I remember this very thread where we were called to discuss about rng as a concept, still everything is tied to RNG. Even more so when it’s possible to get some of the most valuable items (precursors) for the least challenging content (killing trash mobs in open world).
An no, I do not consider precursor crafting as a relief here. A mindless grindfest that simply doubles the cost for a legendary is no compensation for those cursed by rngsus.

You would either have RNG or a method that would be a grind. If items were any easier then everyone would have items sooner and there would be nothing for them to work towards. Imagine if you were able to obtain every skin in the game through casual play within a year and every skin released after that with not so much effort.

This is an MMO, and like other MMO’s, there’s RNG and grind. Fortunately, in GW2, this grind isn’t required as it’s just for cosmetic items only as no content is locked behind a grind except for high level fractals. Even then, you’re not experiencing anything different than you did for the previous levels.

Well, you got some points, but the thing is- at one the only thing GW2 offers is just cosmetics since the overall content quantity is quite limited and for almost everyone further cut down by personal preferences (e. g. some don’t like to PvP, others won’t do world completion/story on every character, etc.).
I’m all for having to put some effort in obtaining desired/rare skins, but to me, the effort should be skill based. Unfortunatly, the current situation is almost exactly what you describe- every skin in the game can be obtained through casual play, even legendaries since the hardest part to get (precursor) can be obtained by everyone, either by luck (killing yellow creatures, 4 rares in the mystic toilett) or numbing grind (precursor collection).
Even the Ad Infinitum backpiece ain’t that hard to obtain as long as you’re either lucky (asc. drops to salvage for balls) or grind (to craft asc. gear to salvage for balls).

Only somewhat exception of this rule might become the legendary armor since it’s tied to raids.

Silmar Alech nailed it down (which was also part of the before mentioned rng discussion thread):

[…] Players want a manipulated rng. A rng that smooths out too many losses and too many successes, so in the end everyone gets the same share of success within his play time. […]

A slight variation of this, coupled with the requirement of actual player skill for the more rare/desireable skins, would be a better distribution system then the current one imo.

Nash.2681:

Ayrilana.1396:

Nash.2681:

John Smith.4610:

Every account has the exact same chance on RNG.

Granted and accepted. But this doesn’t explain why you guys still stick to this kitten concept. I remember this very thread where we were called to discuss about rng as a concept, still everything is tied to RNG. Even more so when it’s possible to get some of the most valuable items (precursors) for the least challenging content (killing trash mobs in open world).
An no, I do not consider precursor crafting as a relief here. A mindless grindfest that simply doubles the cost for a legendary is no compensation for those cursed by rngsus.

You would either have RNG or a method that would be a grind. If items were any easier then everyone would have items sooner and there would be nothing for them to work towards. Imagine if you were able to obtain every skin in the game through casual play within a year and every skin released after that with not so much effort.

This is an MMO, and like other MMO’s, there’s RNG and grind. Fortunately, in GW2, this grind isn’t required as it’s just for cosmetic items only as no content is locked behind a grind except for high level fractals. Even then, you’re not experiencing anything different than you did for the previous levels.

Well, you got some points, but the thing is- at one the only thing GW2 offers is just cosmetics since the overall content quantity is quite limited and for almost everyone further cut down by personal preferences (e. g. some don’t like to PvP, others won’t do world completion/story on every character, etc.).
I’m all for having to put some effort in obtaining desired/rare skins, but to me, the effort should be skill based. Unfortunatly, the current situation is almost exactly what you describe- every skin in the game can be obtained through casual play, even legendaries since the hardest part to get (precursor) can be obtained by everyone, either by luck (killing yellow creatures, 4 rares in the mystic toilett) or numbing grind (precursor collection).
Even the Ad Infinitum backpiece ain’t that hard to obtain as long as you’re either lucky (asc. drops to salvage for balls) or grind (to craft asc. gear to salvage for balls).

Only somewhat exception of this rule might become the legendary armor since it’s tied to raids.

Silmar Alech nailed it down (which was also part of the before mentioned rng discussion thread):

[…] Players want a manipulated rng. A rng that smooths out too many losses and too many successes, so in the end everyone gets the same share of success within his play time. […]

A slight variation of this, coupled with the requirement of actual player skill for the more rare/desireable skins, would be a better distribution system then the current one imo.

You only need to salvage one piece of ascended armor or one ascended weapon to get enough dark energy to complete the first collection. Every collection after that can be completed by salvaging the backpack you craft and using the dark energy obtained.

Other than that, I’m not against gating skins by skill level. That would actually be something I hope they start doing more of and specfically in regards to individual player skill.

Ayrilana.1396:

[…]

You only need to salvage one piece of ascended armor or one ascended weapon to get enough dark energy to complete the first collection. Every collection after that can be completed by salvaging the backpack you craft and using the dark energy obtained.

Other than that, I’m not against gating skins by skill level. That would actually be something I hope they start doing more of and specfically in regards to individual player skill.

Just a slight correction to not spread missinformation about the Ad Infinitum thingy- you actually need 2 balls to start off, one for the collection itself to trade in at the aetherblade guy and another to build the first backpiece itself
Since we seem to agree on the other stuff, no need to argue here

Nash.2681:

Ayrilana.1396:

[…]

You only need to salvage one piece of ascended armor or one ascended weapon to get enough dark energy to complete the first collection. Every collection after that can be completed by salvaging the backpack you craft and using the dark energy obtained.

Other than that, I’m not against gating skins by skill level. That would actually be something I hope they start doing more of and specfically in regards to individual player skill.

Just a slight correction to not spread missinformation about the Ad Infinitum thingy- you actually need 2 balls to start off, one for the collection itself to trade in at the aetherblade guy and another to build the first backpiece itself
Since we seem to agree on the other stuff, no need to argue here

Yeah. I may have salvaged an amulet or gotten more than 1 from the weapon which probably is rare.

John Smith.4610:

Ironically, the randomness of the system is significantly contributes to these posts continually popping up.

With all due respect, I disagree.

It is not the randomness of the system. It is the homeopathic low drop rates.

If I have a chance for a drop that averages at 20%, you will not very likely see me complain about RNG, since even an unlucky streak will not be that terrible.

If otoh I have a chance for a drop of about 1%, it is quite likely to see me complain – simply for the fact that a bad streak can be completely demotivating up to the point where I believe that the item in question does not even drop at all (reclaimed metal plates anyone? I did not have a single drop in 2016 so far, and I am playing on a daily basis).

Solution: rather than have a single complete drop of 1% or less, require 10 drops of 10% each and let players advance towards the final reward.

Halvorn.9831:

John Smith.4610:

Ironically, the randomness of the system is significantly contributes to these posts continually popping up.

With all due respect, I disagree.

It is not the randomness of the system. It is the homeopathic low drop rates.

If I have a chance for a drop that averages at 20%, you will not very likely see me complain about RNG, since even an unlucky streak will not be that terrible.

If otoh I have a chance for a drop of about 1%, it is quite likely to see me complain – simply for the fact that a bad streak can be completely demotivating up to the point where I believe that the item in question does not even drop at all (reclaimed metal plates anyone? I did not have a single drop in 2016 so far, and I am playing on a daily basis).

Solution: rather than have a single complete drop of 1% or less, require 10 drops of 10% each and let players advance towards the final reward.

Not all items are meant to be common. Some are meant to be rare.

Increasing the drop rate makes items more common. Maybe more common than ANet wants them to be.

OK, so I misphrased that.

Suppose now you have to have 3 reclaimed metal plates to buy one of the weapons at DS. They have an abnormally low drop rate. Now what if the drop rate was increased by a factor of 10, but on the other hand you needed 30 to buy the same weapon? On average it takes the same amount of attempts, but you feel progression.

That’s what I meant.

Halvorn.9831:

OK, so I misphrased that.

Suppose now you have to have 3 reclaimed metal plates to buy one of the weapons at DS. They have an abnormally low drop rate. Now what if the drop rate was increased by a factor of 10, but on the other hand you needed 30 to buy the same weapon? On average it takes the same amount of attempts, but you feel progression.

That’s what I meant.

Ah, ok. That I wouldn’t have an issue with. At least for things that already operate on a token system like that.

Silmar Alech.4305:

What you all want is not true randomness, it is equal distribution instead.

Exactly. But it is also easy to see why. So many MMOs, this one included, always talk about how they are doing their best to make the game fair, to make the game balanced. Well, RNG is just about the opposite of that.

RNG is not fair at all. So it is no wonder that people have this perception that isn’t correct.

They put in a direct way to get Precursors (while really fluffing the heck out of it due to lack of checking for quality in development. Seriously, the developer/s should have got each and every one of those issues if they actually bothered to properly test it.) But when you consider “perceived” value vs. cost to make yeap… fairness pops its ugly head up again.

shrugs

RNGs arnt fair, but they are what players want, indirectly.
The alternative to an RNG is not to have one.
This then means that if a group of players contribute to doing something that yields a very good reward, they should all get the reward.
This is fair , but players dont want this because it causes gear inflation in the game, ie suddenly everyone has the best stuff because its being handed out to quickly.
Its a bit like the real world.
Is it fair that everyone in the world isnt equally rich?
Players are indirectly elitist, ie getting good gear is only sensible if everyone else in the game doesnt also have the same gear.

Sartharina.3542:

RedBaron.6058:

Yes, definitely, accounts have an associated RNG, better in some and worst in others.

But only insofar as the Luck Stat is concerned. Two accounts with identical Luck Points have identical RNGs

No, no…what I mean is that accounts have their own initial RNG when they are created plus the lucky points they get during play time.

I was getting one ascended drop a month for six months, then I questioned RNGesus and my drops stopped. Coincidence? Or conspir——<no carrier>

RedBaron.6058:

Sartharina.3542:

RedBaron.6058:

Yes, definitely, accounts have an associated RNG, better in some and worst in others.

But only insofar as the Luck Stat is concerned. Two accounts with identical Luck Points have identical RNGs

No, no…what I mean is that accounts have their own initial RNG when they are created plus the lucky points they get during play time.

No. This theory has been disproved time and time again.

mauried.5608:

RNGs arnt fair

They’re fair in the sense that everyone is treated exactly the same. Those players who get 20+ precursors, everyone else in the game has exactly the same chance of being those players. The same can be said about those who play for 10K+ hours without a precursor drop. Everyone has the same chance of being one of them too.

Ayrilana.1396:

RedBaron.6058:

Sartharina.3542:

RedBaron.6058:

Yes, definitely, accounts have an associated RNG, better in some and worst in others.

But only insofar as the Luck Stat is concerned. Two accounts with identical Luck Points have identical RNGs

No, no…what I mean is that accounts have their own initial RNG when they are created plus the lucky points they get during play time.

No. This theory has been disproved time and time again.

mauried.5608:

RNGs arnt fair

They’re fair in the sense that everyone is treated exactly the same. Those players who get 20+ precursors, everyone else in the game has exactly the same chance of being those players. The same can be said about those who play for 10K+ hours without a precursor drop. Everyone has the same chance of being one of them too.

So either be lucky or be kittened to no return ?
Doesn’t seem fun; in fact it isn’t when you are in the negative outlier.

Btw reading in what you are writing; there should be a chance for me to get out of that outlier.
Except that this chance still haven’t come by in 3 years of playtime but for my buddy that is drown in prec; he still get them.

mauried.5608:

RNGs arnt fair, but they are what players want, indirectly.

Again, I don’t agree. It’s the low drop rates that make it appear unfair and no, it is not what I want and I AM a player.

Look, if I have the choice between “healthy economy” and “feeling rewarded and motivated”, it is an easy choice for me. I don’t care the crap for the economy. I play because I want to enjoy playing, full stop.

I have lost track of the amount of times I did Tequila just to get a hoard. It must be somewhere around 200 or more. Then I finally got it. Guess what? I didn’t feel " oh my, look at that! How good does it feel." I felt “ok, I’ll never do that again. Tick it off, I’m done with it, no way I will try to get another of those weapons”. And I have not been there ever since.

And they don’t even influence the economy, they’re account bound anyways. The same holds true for those cursed reclaimed metal plates. They are only used for the elite spec accended weapon collection, which you only can get once per weapon per account.

See I know: go do a gold farm and buy the rare weapons to salvage them.

Just no. I don’t want to play a game where all you are reduced to is always do the same optimized gold farm then buy stuff from the AH rather than play the content that should give you the rewards just because the drop rates are so pathetic.

In my opinion the drop rates need to be changed. No single drop should have a drop chance of less than 5%. If you want to do something more rare, make it require a larger amount of drops – like you did in HoT in some places, but way not often enough.

What do you mean by a drop chance of less than 5%?
RNG has no memory.
You could do an event 1000 times and get no drop and the next 50 players who have only done the event once all get a drop.
Still drop rate is 5%.
Hardly fair though.

The other thing I’m wondering about is how common these “OMG awesome luck” people are. Most players come to the forum with this complaint and assume that their account is a bad luck count.

Fact is, I’ve never met a person in-game who’s had multiple precursors drop in a short amount of time. In all the time I’ve played, I think I’ve seen a precursor drop on a full map maybe once. So, after thousands of hours… that’s it. One precursor per the hundreds of people constantly around me.

When I say “luck is for other people”, I mean it. When a person is lucky it is a big deal and they announce it proudly. When someone is unlucky, it is business as usual and nobody says anything. So while we may know that guy who’s had 5 precursors fall from thin air since launch, chances are we know 10 times as many players who’ve never received a precursor drop, period.

Now, if it was the norm to receive several precursors as drops from enemies by now, and it just happens that your account has none while everybody else is showered with them, then we might have a case of accounts being utterly unlucky. But, with how things look, it seems more like the RNG is working as intended, and it just happens that a few players fall outside the norm.

Now that most people seem to have accepted the mathematical facts… Let’s try the psychological ones (Well as factual as you can ever get with psychology).

There is a HUGE number of people who believe they have one these so called “Outlier” accounts, have a look at how often these threads come up and how many people chime in with confirmation from their own point of view!

They think this because they overestimate other players success and underestimate the number of ‘spins’ players with lots of successes have taken.

And once they think they have a poor chance of success, they simply give up on chasing it or chase it at a much, much slower rate than those that have had some success early on.

papry.8096:

So either be lucky or be kittened to no return ?
Doesn’t seem fun; in fact it isn’t when you are in the negative outlier.

Agreed, but this isn’t a discussion about whether pure RNG based system is good, but whether it is bugged/manipulated.

papry.8096:

Btw reading in what you are writing; there should be a chance for me to get out of that outlier.

Yes. Chance (meaning, based on luck – and i don’t mean any in-game hidden stat, but a normal, run of the mill, real world luck). Of course, you can help your chance. By doing content with greater drop chances (high fractals for ascended drops, forging for precursors, SW farming in general, for example) or just actively playing more.
The way you play can significantly impact your visible rewards. For example, salvaging/selling rares will likely give you better income than forging them, but in a way that’s mostly invisible to the average player. Forging them on the other hand will give you occasional exotics, named exotics and possibly a precursor, which, while less efficient for wealth acquisition, are more visible, and tend to have a greater psychological impact. Fractal chests are best opened in larger quantities – you are almost certain to get at least one-two good ascended drops one opening session that way (while opening them as soon as you acquire them, while having the exact same drop results total, might seem as less rewarding, as most open sessions will not give anything special). The same doesn’t hold true for champion boxes anymore, unfortunately (the chance to drop anything good from them has been nerfed so much, you’d need whole stacks for it).
Also, remember that old jewish joke about a man praying to God for fortune, and God answering “first buy that lottery ticket”? Content skipping may be faster and bring more gold per hour, but killing everything that moves gives you more “precursor chances”. Both my open world precursors came from trash mobs most people avoid. It may very well be that you have run past your “precursor drops” without noticing them. Perhaps this even happened many times by now.

All that’s assuming you are in an outlier at all – you should understand, that the drops for average people aren’t that good either, and most of the complainers here likely are no outliers at all, but dropwise belong well in the middle of the game population.

papry.8096:

Except that this chance still haven’t come by in 3 years of playtime but for my buddy that is drown in prec; he still get them.

If you are doing open world almost exclusively, and do not use forge much, that’s unfortunately highly possible. Remember, mere playtime will give you no drops at all. It’s how you play that influences that.

Astralporing.1957:

If you are doing open world almost exclusively, and do not use forge much, that’s unfortunately highly possible. Remember, mere playtime will give you no drops at all. It’s how you play that influences that.

Here come the assumption that “people just sit in Lion’s arch”.
The part that you don’t seem to get is that aside that I have more playtime than my prec looting buddy; I do almost everything like him as we often play together.
Most of my playtime (at least 95%) is from actual play & not waiting at lion arch.
That what’s frustrating; playing in the same way & don’t get anything while your friend with less hour played is drowned in prec.

Fun facts :
- Everytime I had excess T5 & mithril; I was doing greatsword to throw in the forge.
- All my yellow weapons from anything have gone in the forge for the last 3 years.
- I was putting so much in the forge than I buyed a mystic conduit & ended up convincing my prec looting to also buy one (which he loved) as we had the same intensive usage.

So no that thing isn’t right.

papry.8096:

Astralporing.1957:

If you are doing open world almost exclusively, and do not use forge much, that’s unfortunately highly possible. Remember, mere playtime will give you no drops at all. It’s how you play that influences that.

Here come the assumption that “people just sit in Lion’s arch”.

Where exactly did you get “people just sit in Lion’s Arch” from my “doing open world almost exclusively”?
Also, what is that “so much”? Do you forge stuff every day? Are your numbers of attempt daily in one, two, maybe three digits? (If it’s any less, you still are at below 50% chance of getting a SINGLE precursor out of it since the beginning of the game).

Ah, seen you have clarified that in the meantime. Yes, that would make you a low outlier indeed. And yes, that is going to be frustrating. Unfortunately, it doesn’t mean that the RNG is not working as advocated, only that possibly it’s not the best system to use for drop distribution. That is however something for a separate thread.

Astralporing.1957:

Where exactly did you get “people just sit in Lion’s Arch” from my “doing open world almost exclusively”?
Also, what is that “so much”? Do you forge stuff every day? Are your numbers of attempt daily in one, two, maybe three digits? (If it’s any less, you still are at below 50% chance of getting a SINGLE precursor out of it since the beginning of the game).

Okay, my bad there.

Astralporing.1957:

Ah, seen you have clarified that in the meantime. Yes, that would make you a low outlier indeed. And yes, that is going to be frustrating. Unfortunately, it doesn’t mean that the RNG is not working as advocated, only that possibly it’s not the best system to use for drop distribution. That is however something for a separate thread.

I actually think that this is a perfect thread to lash at it.

Seera.5916:

DGraves.3720:

Seera.5916:

Then get a decent number of players (20-30) together to do tons of testing on RNG. Have everyone note down EVERY possible variable that your group feels may affect the results. Magic find, the boosts to Magic find, what enemies they fight, what level and what type of items they throw into the Mystic Forge, etc. Have everyone record EVERY drop. Even note down the trash or when no rewards are given (if you go after drops from enemies). In the order received. And note down chests and then what was in the chests And have everyone then share the data on a public site and link to it so that everyone, including ANet, can look at the raw data and the variables accounted for. And make sure everyone has at least 250 items (trash or better, so keep going if they receive nothing for drops from enemies) on their list.

If right now there are 100,000 players online and each one gets a loot bag on average every 5s your sampling is worthless. The problem is not that the sampling cannot be taken but that without extensive sampling it’s all too small.

Players won’t be able to conclusively say that there is something wrong with the sample size. But it may show that there may be something wrong with the system.

But until the player base that claims something is wrong with the system gets off their butt and documents findings from players who claim to be on both sides of the luck spectrum and players who are in the middle, then there’s no proof that ANet is lying. And ANet has access to the code and knows what the intent of the code is supposed to.

The person who exposed the ecto problem didn’t prove that there was a problem in their findings. They proved that there might be a problem.

. But all modeling places the burden of proof on the modeling party. If I tried a sandwich and got sick, concluding it made me sick, does that mean the sandwich was poisoned?

In many minds, yeah, but saying I got sick because of the sandwich leaves out the possibility that it could have been something else. Such as the old egg salad I had with it, homemade, but not well refrigerated.

People jump to conclusions way too rapidly. Extensive testing doesn’t prove anything. It just shows a pattern which can be made to assert something. You could be right about hidden variables or totally off. Who knows?

Halvorn.9831:

OK, so I misphrased that.

Suppose now you have to have 3 reclaimed metal plates to buy one of the weapons at DS. They have an abnormally low drop rate. Now what if the drop rate was increased by a factor of 10, but on the other hand you needed 30 to buy the same weapon? On average it takes the same amount of attempts, but you feel progression.

That’s what I meant.

Reclaimed drop rates aren’t really that rare if you’re talking salvaging. Masterwork seems to be about 1/35 (regardless of salvage kit – check the talk page on wiki) while rares are guaranteed. You can either take the ‘safe’ route with respect to rng and buy rares for ~6g buy order, or you can take the (ultimately cheaper for most people) route that is reliant on rng evening out and buy out masterwork weapons. This is what I did to finish the Machined collection – I got kind of unlucky, still cheaper than bids for rares and I was buying out masterworks, not using buy orders.

A question for the ‘anecdotal evidence is king’ folks, the ‘incredibly small sample sizes are sufficient’ gang or the conspiracy theorists among us, I’d genuinely like to know how you think this lucky account business programmed into the game. Do you think the devs just pick out random people that make nice posts about them on the forums? People that haven’t played in a while (personal fav)? Do you think they just pick people in game that look cool or what? Is the choosing random itself? When the chosen have been chosen, do you reckon a couple of devs are just like “alright lets artificially boost his MF by 500%”.

Accounts do not have “luck” but I believe players do…

Beast Sos.1457:

Ayrilana.1396:

Beast Sos.1457:

John Smith.4610:

Every account has the exact same chance on RNG.

Then how do you explain a person getting 5 precursors in 20 forges? Luck? Hell no because that same guy will get a precursor drop off a mob in WvW then the next day will get 3 ascended chests and the next day will get another pre off a random mob…. While a person like me can play for 4k hours and not get a single pre.

Time played in the game does not influence whether you will get a precursor drop. Just like someone playing the lottery for 60 years is no more likely to win than someone who has played for one year. You need to take a closer look at probability.

If RNG with the lottery worked the way RNG does in Gw2 then there would be at least a couple people in the world who win the lottery every week. Which doesn’t happen. If it does show me a link then I’ll kitten.

It does. You people just dont understand/want to understand that. Just because someone didnt get the jackpot/precursor, doesnt mean they didnt match 4 or 5 numbers or something (ie dropped a named exotic worth 100g or so).

Beyond that, Ayri is correct that the “time” you logged in game means less than dolyak feces. You could have 20k hours logged in the game, but only actively played for 100 hours. Of those 100 hours, you may have only spent 30 hours combined in the act of killing enemies and looting. Of those 30 hours, you might have only spent 5 of them at level 80, which is the only point you’ve got the highest chance of obtaining a precursor.

Even further, all you tinfoil knights have an uber-massive case of confirmation bias going on. “wah~ JS said we’re all wrong, but all my anecdotal ‘proof’ is 100% true and unassailable. That must mean JS is the one who’s wrong.” Really now guys. You would think the guy saying the same thing for over 2 years would have some ground for SAYING it right? Who am I kidding, these threads are a hop, skip, and a jump away from calling JS the next corporate mouthpiece because he doesnt confirm an obviously false conspiracy.

mauried.5608:

What do you mean by a drop chance of less than 5%?
RNG has no memory.
You could do an event 1000 times and get no drop and the next 50 players who have only done the event once all get a drop.
Still drop rate is 5%.
Hardly fair though.

A (pseudo-) RNG creates a bit sequence that looks random. In fact, if you use a true RNG it is really random (using e.g. quantum effects), on a P-RNG it is not, but if you don’t know the seed, the next bit value is still unpredictable.

Now, what you typically do is take a bunch of those bits and use them as the mantissa for a binary floating number, turning the bit stream into a list of floating point numbers between 0 and 1. If you draw a bazillion of these and plot their values, you see that they are evenly distributed between 0 and 1 (this holds only true for large sample sizes).

What you then do is the following: the value range of the random numbers is from 0 to 1. So, if you filter out the ones between lets say .4 and .5, you have taken out a smaller range that is 10% in size of the complete range. So, you draw a number and if it is between .4 and .5, it represents an item drop for an average drop chance of 10%.

If you instead limit the range to values between say .71 and .72 (because of the even distribution, the real values do not matter), you have a drop chance of 1%. That’s how it goes. It will only create a real drop ratio of 1% if you really take a lot of samples – and that is the exact problem I am talking about.

Amurond.4590:

Reclaimed drop rates aren’t really that rare if you’re talking salvaging. Masterwork seems to be about 1/35 (regardless of salvage kit – check the talk page on wiki) while rares are guaranteed. You can either take the ‘safe’ route with respect to rng and buy rares for ~6g buy order, or you can take the (ultimately cheaper for most people) route that is reliant on rng evening out and buy out masterwork weapons. This is what I did to finish the Machined collection – I got kind of unlucky, still cheaper than bids for rares and I was buying out masterworks, not using buy orders.

I tried way more than 35 masterwork weapons and did not have a single success. That’s what I meant by low drop (or salvage rates) being a problem. And no, I am not considering to give in to the gold farm/AH path. I would like to play the parts of the game that naturally shell out these rewards and be rewarded while doing so.

DGraves.3720:

Seera.5916:

DGraves.3720:

Seera.5916:

Then get a decent number of players (20-30) together to do tons of testing on RNG. Have everyone note down EVERY possible variable that your group feels may affect the results. Magic find, the boosts to Magic find, what enemies they fight, what level and what type of items they throw into the Mystic Forge, etc. Have everyone record EVERY drop. Even note down the trash or when no rewards are given (if you go after drops from enemies). In the order received. And note down chests and then what was in the chests And have everyone then share the data on a public site and link to it so that everyone, including ANet, can look at the raw data and the variables accounted for. And make sure everyone has at least 250 items (trash or better, so keep going if they receive nothing for drops from enemies) on their list.

If right now there are 100,000 players online and each one gets a loot bag on average every 5s your sampling is worthless. The problem is not that the sampling cannot be taken but that without extensive sampling it’s all too small.

Players won’t be able to conclusively say that there is something wrong with the sample size. But it may show that there may be something wrong with the system.

But until the player base that claims something is wrong with the system gets off their butt and documents findings from players who claim to be on both sides of the luck spectrum and players who are in the middle, then there’s no proof that ANet is lying. And ANet has access to the code and knows what the intent of the code is supposed to.

The person who exposed the ecto problem didn’t prove that there was a problem in their findings. They proved that there might be a problem.

. But all modeling places the burden of proof on the modeling party. If I tried a sandwich and got sick, concluding it made me sick, does that mean the sandwich was poisoned?

In many minds, yeah, but saying I got sick because of the sandwich leaves out the possibility that it could have been something else. Such as the old egg salad I had with it, homemade, but not well refrigerated.

People jump to conclusions way too rapidly. Extensive testing doesn’t prove anything. It just shows a pattern which can be made to assert something. You could be right about hidden variables or totally off. Who knows?

Modeling? No one would be modeling anything by doing the experiment. They would be using the actual system to gather data on what dropped and noting down the possible variables that might affects drops as they played. A model is a representation of the system. That would be people developing a computer program that would randomly generate drops based on what players thought the loot table was in the game. How they thought things like magic find or time spent playing in the zone affected it (we don’t know how exactly DR affects the drops or exactly when it takes hold).

The group of people who have the ability to look at the code and know the intent of the code says there is nothing about a person’s account that makes them unlucky or lucky in the drops they get in the game.

The burden is on those who claim otherwise to show that that may not be the case. That there may be a bug in the system.

The issue that the people complaining about RNG being unfair are looking at it across a very small subset of accounts rather than an aggregate of accounts consisting of the entire playerbase.

You can have 1,000 people flip a coin. On average those who get heads achieve this 50% of the time. You’ll have maybe 20 that got it 80% of the time and another 20 that got it 20% of the time. The people complaining are focusing on those 20 on either side (40 total) rather than across the entire population of people flipping coins.

The question all this comes down to (as I see it) is: Are y’all able to give the benefit of the doubt that the RNG for GW2 is accurate and working fine?

If yes: then go and play the game! Stop worrying about the code, and go experience it with all the ups and downs that come with it.

If no, but you still want to play GW2: then take advantage of the various systems that give you more control than a pure RNG. Crafting, TP, & reward tracks are all there to give you more control. Utilize them as much as possible, and enjoy the game!

If no, and you don’t want to play gw2 because of it: then you’re a Flame Legion saboteur or Inquest recruiter. :P

~EW

Scandi.5693:

John Smith.4610:

Every account has the exact same chance on RNG.

I do not believe this for one moment. There are too many instances of lucky people. Even in our guild we have two that get so many precursors, we pool our rares in the guild bank and let them do the turn ins for us. One person has had over 30 precursors since the games launch and has made every legendary (up to and including nevermore-is working on the other new ones), even underwater ones because, why not. Deny all you want, but your system is broken.

When you have an entire guild feeding one person items/money, your little system is no longer a random model.

Sariel V.7024:

Scandi.5693:

John Smith.4610:

Every account has the exact same chance on RNG.

I do not believe this for one moment. There are too many instances of lucky people. Even in our guild we have two that get so many precursors, we pool our rares in the guild bank and let them do the turn ins for us. One person has had over 30 precursors since the games launch and has made every legendary (up to and including nevermore-is working on the other new ones), even underwater ones because, why not. Deny all you want, but your system is broken.

When you have an entire guild feeding one person items/money, your little system is no longer a random model.

No, that’s your guild being conned into feeding one person items/money because they believe it.

Seera.5916:

Modeling? No one would be modeling anything by doing the experiment. They would be using the actual system to gather data on what dropped and noting down the possible variables that might affects drops as they played. A model is a representation of the system. That would be people developing a computer program that would randomly generate drops based on what players thought the loot table was in the game. How they thought things like magic find or time spent playing in the zone affected it (we don’t know how exactly DR affects the drops or exactly when it takes hold).

The group of people who have the ability to look at the code and know the intent of the code says there is nothing about a person’s account that makes them unlucky or lucky in the drops they get in the game.

The burden is on those who claim otherwise to show that that may not be the case. That there may be a bug in the system.

Analysis of the data would require some form of model. Whether it be averaging or regression or prediction or something else. That’s why the burden lies on the modeler.

DGraves.3720:

Seera.5916:

Modeling? No one would be modeling anything by doing the experiment. They would be using the actual system to gather data on what dropped and noting down the possible variables that might affects drops as they played. A model is a representation of the system. That would be people developing a computer program that would randomly generate drops based on what players thought the loot table was in the game. How they thought things like magic find or time spent playing in the zone affected it (we don’t know how exactly DR affects the drops or exactly when it takes hold).

The group of people who have the ability to look at the code and know the intent of the code says there is nothing about a person’s account that makes them unlucky or lucky in the drops they get in the game.

The burden is on those who claim otherwise to show that that may not be the case. That there may be a bug in the system.

Analysis of the data would require some form of model. Whether it be averaging or regression or prediction or something else. That’s why the burden lies on the modeler.

Yea, and the people doing the experiment can do that and present their results to ANet and then ANet can decide if the experiment was in depth enough to look at if the results show any abnormalities.

Which is what the person who discovered the ecto salvage rate bug did. They did an experiment with a large number of samples. Not enough to conclusively prove there was a problem with the ecto salvage rate, but enough to make ANet look into it and they got enough evidence to discover that there was a problem with it.