38 pres wards in a row still did not refine an AE at 10%

frontliner · August 2017

I spammed through all 38 of my pres wards trying to get a shield up, it was only supposed to be a 10% chance, but in reality it still didn't happen and it is now down to 1.8% chance. This is a repeating issue, and while we all understand there is inherent probabilities, this is stepping into "extremely unlikely to happen" probability. This makes you really frustrated with the game and start to feel like you're trying to bleed us for money to just keep buying more wards. Several of my guild mates have also been seeing this with even higher percentage gear (etc).

randomdigits · August 2017

I actually enjoy observing this randomness and my (incorrect) intuition. The most remarkable part is the gambler's fallacy - which is basically thinking that once you failed (several times) you now have a higher chance to win soon. The truth is if you failed - your odds are back to 10% on you next try.
Specifically there is a 1.8% that someone would experience a 38 losses in a row for a 10% chance trial.

kalina311 · August 2017

pretend you put a quarter(10 pres wards ) in a gambling machine for a 1 in 10 chance to win
and you lost 4 times

that is essentially what happened
pretend you used 40 instead of 38 ..

nothing remarkable here ... moving on

how many times did you get something on your first try and then run to the forums to complain that something was obtained too few attempts and you were too lucky ? -

randomdigits · August 2017

kalina311 said:
pretend you put a quarter(10 pres wards ) in a gambling machine for a 1 in 10 chance to win
and you lost 4 times

I like your example, but to be accurate, the chance for that quarter to bring you a win in this case is actually 65%. So the chance of 4 attempts failing is 0.35 * 0.35 * 0.35 * 0.35 ~ 1.5%

zebular · August 2017

It is not a cumulative factor.

In this case with a 10% chance, it is:
Try 1, 10% chance.
Try 2, 10% chance.
Try 3, 10% chance.
Try 4, 10% chance.
Try 5, 10% chance.
Try 6, 10% chance.
Try 7, 10% chance.
Try 8, 10% chance.
Try 9, 10% chance.
Try 10, 10% chance.
and so on...

Your previous attempt has absolutely no bearing on your following attempt. You could try a million times and still not get it. You could try once and get it.

armadeonx · August 2017

After a few failures, zone to an area with very few players and wait a minute before trying again.

I used to refine in PE and never had a good average on rng but now I always do well, usually in less that 1/3rd of the stated chance (and I do a lot of refining).

Some people will tell you that this is the equivalent of 'wearing your lucky socks' or whatever, but the only concrete thing I can tell you is that I always beat the numbers now.

mordekai · August 2017

Yesterday...
Three Rank 10 enchantments, 72 pres wards, each giving a 5% chance for rank 11.
1st one popped at try number 17, good oh... that put me about three ahead of standard.
42 tries later (number 59) the second one popped. Started to get that "here we go again" feeling and began figuring what I had left to sell...
Third one hit three goes later on ward 62, so despite one being almost on the nose, one wildly over and another vastly good luck, the balance was pretty close to spot on.

Last double rp I did two refines at 10% (can't remember exactly what items) first took just shy of fifty wards, second popped on the very next go.

As mentioned above, we often only remember when the RNG goes way past the typical odds, and forget all the times it comes in in our favour amd just a few under.

dsn1118 · August 2017

@zebular it may not be cumulative but as @randomdigits#2166 said there is a 1% chance to fail 38 times with 10% success chance.

I think game has a design that every instance in the game has shared pool of RNG rates. I mean most of long time players choose to do their refinement or open their lockboxes in private instances like stronghold PVP tutorial

frontliner · August 2017

I don't think I clearly stated the point of my post. I did not say there was specifically a bug, etc, just that the current system in place may have some flaws in it, which can sometimes cause major frustration to the point where people do not want to deal with it. The fact that players have to go to isolated instances to open or level their gear, if that is actually true, is an issue that needs to be addressed. It is in ARC's best interests to keep the game fun and rewarding, and it is in our best interest to let them know when that is not reasonably happening.

armadeonx · August 2017

You're right, the current rng system is bad, there's no point in denying it. However the game is built entirely around the concept of rng so I doubt they are going to change it - the only thing you can do is 'manage' it.

thefiresidecat · August 2017

I worry about openly acknowledging how we "manage it" here. I hate and fear their nerf hammer

*the first rule of the refinement club is that we don't talk about the refinement club

randomdigits · August 2017

I'm embaressed to say

that I never go to any special instances, and usually do all of my refinement in PE where mailbox, bank, salvage and shops are all nearby (and yes, I've noticed that some unfortunate developer removed the Sell button from the Trade bar merchant).

And man, do I refine... must have refined rank 3=>7 and sold about 40 of brutals last time they were cool (no, now they are cheap). In the long run all odds add up as expected. How to avoid frustration? Be prepared to lose an use the law of big numbers.

thefiresidecat · August 2017

> @randomdigits#2166 said:
> I'm embaressed to say

that I never go to any special instances, and usually do all of my refinement in PE where mailbox, bank, salvage and shops are all nearby (and yes, I've noticed that some unfortunate developer removed the Sell button from the Trade bar merchant).
>
> And man, do I refine... must have refined rank 3=>7 and sold about 40 of brutals last time they were cool (no, now they are cheap). In the long run all odds add up as expected. How to avoid frustration? Be prepared to lose an use the law of big numbers.

Try it

randomdigits · August 2017

thefiresidecat said:
Try it

I'm all for trying various fun things, like checking if fashion gear gives protection from lava...
Though refining is a bit costly to test. How about I refine 2 batches of 100 rank 2 to rank 5 - one in PE and one in an empty instance of ToS or eSoT or IG, and I would also leave that ToS/eSoT/IG instance after a failure. Would that test be conclusive enough for you if I run it?

thefiresidecat · August 2017

siege tutorial and portal to tuern are the things I generally do. not saying you need to test it for my benefit. just saying for your own benefit. do or don't try it. just sayin. it does seem to make a difference. it isn't magic.. but the luck does seem to be slightly better for preswards that way. I haven't seen any difference for lockboxes but for refining it does seem to work better in single instances and low population instances.

omegarealities · August 2017

Once it took 50 tries at a 5% chance.
The next it took 5 times at a 5% chance.
I guess ir's broke both way.

greywynd · August 2017

frontliner#7980 said:
I don't think I clearly stated the point of my post. I did not say there was specifically a bug, etc, just that the current system in place may have some flaws in it, which can sometimes cause major frustration to the point where people do not want to deal with it. The fact that players have to go to isolated instances to open or level their gear, if that is actually true, is an issue that needs to be addressed. It is in ARC's best interests to keep the game fun and rewarding, and it is in our best interest to let them know when that is not reasonably happening.

...coalescent wards...

armadeonx · August 2017

Personally I believe that refining in low/no population areas makes refining work as intended and that the impact of too many other players skews the results negatively.

randomdigits · August 2017

thefiresidecat said:
Try it

I'm ok with the outcomes I'm getting and would only run the test if that would help to convince some of you, refiner club folks, that everything is WAE (or, if it isn't that would prove me wrong). Thus I would only run the test under preagreed conditions, as long as it wouldn't cost me over, say, 200k. So if some of you say get a new instance after a few failure, please determine if this is required. And if some say preservation ward must be used - again does it?

P.S. What you see is what we, mathematical statisticians, call confirmation bias.

thefiresidecat · August 2017

yeah, it may be confirmation bias. I think most of us are aware that could be it. but I also dont' think there is any testing that can prove it one way or another because it is still rng. i've done a lot of refining in both high population zones and low and it really does seem to take less wards on average in a lower instance.

and enough people have experienced this that it's a pretty much agreed upon phenomenon. in most circumstances i'd be thinking that it was probably just in our heads but given this game and it's quirks and it's track records and the way things rarely work as you'd think or as described... I don't think it's on par with believing the aliens walk among us lol

kalina311 · August 2017

randomdigits#2166 said:

kalina311 said:
pretend you put a quarter(10 pres wards ) in a gambling machine for a 1 in 10 chance to win
and you lost 4 times
I like your example, but to be accurate, the chance for that quarter to bring you a win in this case is actually 65%. So the chance of 4 attempts failing is 0.35 * 0.35 * 0.35 * 0.35 ~ 1.5%

yup was actually trying to put it in a way for a less mathematically minded player//
each toss (is not self aware or quantum entangled to subsequent tosses ) it has no way of knowing if its previous toss is meaningful to any statistic/trend either execpt if you are tracking patterns trends or streaks

randomdigits · August 2017

So I gave the Siege Training map a try recently running all of my Brutal rank 10 to 11 upgrades there, remembering that they run at a 5% chance. At first it started very promising with one ranking up taking 5 attempts and another taking 15. But then I've got into a full 71 and counting attempts, which is of course depressing but not unexpected.
As I am saving AD for another 50 attempts (I don't like running short of wards all the time) I just wanted to let all of you "Refiner Club folks" known that there is no magic map instance or lucky rabbit foot. It's OK if you believe there is one, some people are just luckier in some things, but in the long run (remember the big number law) the overall stats bring no surprises.

38 pres wards in a row still did not refine an AE at 10%

Comments