Failing 16 out of 24 times at 60%+?!?!

General Discussion
06/15/2016 06:30 PMPosted by OldPro
But players don't form opinions objectively, based on hard, cold facts and statistics


Yes we do. Well, I do, but I might as well generalize to.
06/14/2016 12:23 AMPosted by lukipela
Go take a statistics course.


^This^

Also, inb4 OP comes back and whine about coin flips resulting in two times in a row on heads.
The OP acts like he never got 3 successes out of 3 at 60%. It's not supposed to happen but nobody goes running to the forums complaining about it.

So all those times you got 3 out of 3, consider this run payback. It's stats. You aren't supposed to win in Vegas, but enough people do to keep the place both busy and profitable.
Gem upgrade system is sh!t.
If you do the rift, it should be a 100% upgrade.
Blizzard can't do proper coding, just crappy RNG.
Wonder how people would react to Paragon Points being
awarded at 60% chance?
This forums would erupt into anarchy.
06/14/2016 12:09 AMPosted by Jman420
This happens wayyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy more often than not.. I fail way more when probability says I should be succeeding more...Explain please blizzard. And will someone explain what reason there is to play after getting the cosmetics if you are just going to waste your time to constantly fail on leveling gems?

Also..Why have I only got one ancient bow in 880 paragon levels? Seriously wth is going on this season?


If you think that probability will have an outcome on future rolls of rng, like it is some god. Well you are wrong. I think that you have a touch of Gambler's Fallacy. You actually think that past rolls have an affect on future rolls. When in reality you never leave the same percentage chance of getting an upgrade.

Now if this is happening and it is not some outlier cases of Gambler's Fallacy. Then the only thing that could be the cause would the empowered GRs. That is something that I would suggest for Blizz to check into. Their GR empowerment system could've caused problems with the % chance of leveling gems.
06/17/2016 12:29 AMPosted by ShadowAegis
06/14/2016 12:09 AMPosted by Jman420
This happens wayyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy more often than not.. I fail way more when probability says I should be succeeding more...Explain please blizzard. And will someone explain what reason there is to play after getting the cosmetics if you are just going to waste your time to constantly fail on leveling gems?

Also..Why have I only got one ancient bow in 880 paragon levels? Seriously wth is going on this season?


If you think that probability will have an outcome on future rolls of rng, like it is some god. Well you are wrong. I think that you have a touch of Gambler's Fallacy. You actually think that past rolls have an affect on future rolls. When in reality you never leave the same percentage chance of getting an upgrade.

Now if this is happening and it is not some outlier cases of Gambler's Fallacy. Then the only thing that could be the cause would the empowered GRs. That is something that I would suggest for Blizz to check into. Their GR empowerment system could've caused problems with the % chance of leveling gems.


but why do some "luckier" people have strong and unusual trends of being luckier than ordinary or unlucky persons? Many believe that statistics and probability are based on randoms with no input from the person. If that is really the case, then how are lucky people consistently luckier on a higher % chance than unlucky people (who are consistently lower on a % chance)?

My best mate would roll a very high number of Sixes on a D6 when we played D&D. He was considered lucky by everyone. I'd be the poor sod rolling lots of 1s. We'd quite often use the same dice. And it didn't matter what dice he used, it didn't seem to affect his luck. Neither did it make a difference for me what dice I used, I'd typically roll a One. It resulted in some absolutely hilarious situations for me though, and I enjoyed role playing the bad luck in my dice rolls. Since we were running a large group (8-10 players on any given session), myself being out of the picture for some hilarious reason didn't really affect the party to a great degree. On a small 4 man party, my fellow players would have hated me lol, since in such a small team, every player needs to be fully active. For me it was "I run forward" the DM would say, roll a D6 to see if you stumble, I'd roll a 1 (fail) and stumble. Roll a 2nd D6 to see if you catch yourself (5-6 roll) or fall (2-3-4) or fall and knock yourself out and miss the entire combat (1). Guess what I'd very often roll!

Some people are just naturally luckier than others, and unless humans are capable of influencing outcomes with their minds (science says NO), why is this the case? In a game like D3, being unlucky is incredibly frustrating, especially when compared to the players who by the most part, lucky in the RNG rolls.

I guess this is why I dislike true RNG like Blizzard employs, cos I'm always at a disadvantage it seems. And it's why I prefer a pity chance like with the pity drop for legendaries.
06/17/2016 02:29 AMPosted by mrtheoden
Some people are just naturally luckier than others, and unless humans are capable of influencing outcomes with their minds (science says NO), why is this the case? In a game like D3, being unlucky is incredibly frustrating, especially when compared to the players who by the most part, lucky in the RNG rolls.


What you are explaining here are appearances based on memory. People tend to remember streaks of bad luck. I can guarantee you any streaks of good luck that have come your way in this game will go by unnoticed by you and others like you. Think anyone complains here on the forums if they get 20-24 successes at 60%. Of course not, instead they take it as a way of getting pay back for all of the times they get bad luck on upgrades.

Your idea of people that just are naturally luckier than others. Again some is no doubt based on appearances. Unless you can follow them around for their entire lives I would say that since you remember seeing them get all good luck that is what you will remember. Also some people happen to be in the right place at the right time. As others happen to be in the wrong place at the wrong time.

The only way a person would have any power to influence any random event is either if he found a way of rigging it in his favor. Or he had help from any of Satan's demons, that is if you believe in a god. Unless he has a superpower that he doesn't know about that came from no influence of demons or anyone else doing it on purpose.
So currently having a terrible run with lgem upgrades @ 60%...so far

1st rift - 1/3
2nd rift - 1/3
3rd rift - 2/3
4th rift 1/3
5th rift 1/3
6th rift 0/3

all @ 60%...

that's 6 from 18 or 33.33%...

so to even this up, I should get 9 straight at least..that'd take it to around 57%, still below the "average" of 60%...what'd be the odds of that happening eh, Blizzard *wink wink*.

So, Blizzard, is your RNG broken or what? BTW, I don't get results in the "positive" like this, only negative (by positive I mean where "odds" go my way when they shouldn't).

For those suggesting I take a statistic course, I'll suggest that you take a long walk off a short pier if you get my drift.

RGN is BROKEN in this game. Utterly BROKEN. The above should NEVER happen. And don't give me that crap that "it's too short a run to be of any use statistically". If Blizzard are coding right, it should stay consistent over 10 runs or a 1000 runs.
The above should NEVER happen. .


sigh..
I don't think anything is broken with it - I have had huge runs of success at 60%.

But either way, the main point is the gem upgrade system kind of sucks. I would much rather go to an experience system where you get experience and your gem levels.
24 recordings is too small. It's a bad run but over a large number of times you should see it balance out to near 60%.
Basic Math:

(16 * 100) / 24 = 67%

But I know that this is irrelevant, as far I know, the run is different from each other, basically you have 60% success in that roll, not on all the runs.

You're only having bad luck.
The next run was 1 from 3...after that, I packed it in for the night.

I can guarantee that after all of the greater rift runs that I've done, I'm nowhere near 60%...not even close. I might occasionally get 2 from 3 @ 60%, but that might be 2 greater rifts in a row at most, and after that it's all bad, typically 1 in 3 for an extended run with the odd 0 from 3 thrown in...1 in 3 seems to be the predominant result from my experience.

Given that there is no mechanism to even this out (it's a fresh roll every time with no recording of prior rolls taken into account), if One is unlucky, One gets screwed...and this is the problem I have with Blizzard's RNG...

person a) is lucky and tends to have good RNG rolls with legendary gem upgrades, typically matching what Blizzard publishes...

person b) is unlucky and gets nowhere near the RNG rolls that Blizzard publishes...

Both persons play hard and put hours into the game...One is rewarded due to luck, the other not. Where is the reward for effort? That's what the game should be rewarding, not "luck".
05/06/2016 01:03 PMPosted by MeWhenum
I've never been one to complain too much about the RNG, and I usually dismiss claims of Urshi being bugged as people just getting a bad couple runs and observation bias. However, over the last couple days I have gotten ATROCIOUS rates on upgrades from her, like 10+ 60% fails in a row, so I decided to do a simple Chi squares test to see if her listed rates are legitimate or not.

I recorded the successes or failures of 100 60% upgrade attempts. They were not consecutive, but they shouldn't need to be if Urshi is fair. Result; 41 successes and 59 failures at 60% success chance. This already looks extremely rigged, but let's give her the benefit of the doubt and do the test out.

Degrees of freedom = 1 (two categories, pass or fail)

Expected (fail) = 40

Expected (pass) = 60

Chi square statistic calculated from this (look up the test if you want to confirm) = ((41-60)^2 / 60) + ((59-40)^2 /40) = 15.04

Calculated chi value for 99% confidence, DF = 1 is 6.64.

15.04 is WAY more than 6.64.

Thus, I can state with 99% mathematical certainty that Urshi DOES NOT give successes at the listed rate. Other people should do this test to confirm. This needs to be investigated because failing a gem upgrade that's 1-5 levels above your gem level is already annoying, but her being rigged makes it about 100x worse.

Edit; to clarify for those who are unfamiliar with statistical tests, I did NOT pull the number 99% out of thin air. The chi value I used for comparison to the calculated value is the defined value for 99% confidence. Based on the fact that Urshi failed by quite a bit (15.04 >> 6.64), the actual probability of this being random is significantly LESS than 1% ( >99% confidence).

Edit 2; The people saying more data would be better have a valid point, so please, anybody who has the time, record your experiences and report them here so that we may have a more accurate measurement.

Edit 3; Since apparently the size of this thread makes people unwilling to read the rest of it, I should make it exceedingly clear that this does not "prove" anything. It is merely a signal that we ought to be suspicious of Urshi. This is the point of asking for more people to supply data.


Thread:
http://us.battle.net/d3/en/forum/topic/20743535499?page=1#new-post
Got 10 from 10 @ 60% this morning on the bounce.
Does that mean 60% chance is wrong? No.
06/14/2016 12:09 AMPosted by PotusPanda
Can someone explain how I'm failing 16 out of 24 times at 60%??!?!?!?


Sure.

You have a 60% chance to succeed, every time you roll. After you roll, your chances are still 60%. They do not increase.

Technically if it's not 100%, it's 50%. Either it'll go, or it won't. I've seen 5+ rolls in a row fail @ 93%.
08/23/2016 08:50 PMPosted by AnimalMan
Thread:
http://us.battle.net/d3/en/forum/topic/20743535499?page=1#new-post

Citing something you don't understand from someone who doesn't know what he is doing is never a good idea.
08/24/2016 05:38 AMPosted by McLight
08/23/2016 08:50 PMPosted by AnimalMan
Thread:
http://us.battle.net/d3/en/forum/topic/20743535499?page=1#new-post

Citing something you don't understand from someone who doesn't know what he is doing is never a good idea.


MeWhenum's forum guide to looking like an idiot!

1) Someone says something that you don't understand, you disagree with, or doesn't support your narrative.

2) Disregard their evidence without providing any counter-evidence or arguments, or pointing out any flaws in their method.

3) Call them a troll/hope that nobody notices the content of the other poster. (Not yet completed.)

4) Profit!
06/14/2016 10:33 AMPosted by adm0ni
Don't listen to the clowns telling you to take a statistics course. It is they that need the refresher.

The best bet you can do is maintain a log of your pass/fails and submit them to the bug forum. I would suggest you don't submit anything with less than 30 attempts. More would be better.


LMFAO 30 attempts.

More like, don't submit anything with less than a hundred thousand attempts.

You're a retard.
06/16/2016 11:39 AMPosted by LocknLoad
Gem upgrade system is sh!t.
If you do the rift, it should be a 100% upgrade.
Blizzard can't do proper coding, just crappy RNG.
Wonder how people would react to Paragon Points being
awarded at 60% chance?
This forums would erupt into anarchy.


good gawd man, what's the criteria for gem upgrades in your opinion?

If it worked this way, people would have gem level 3,000 Bane of the Trapped. and one-shot everything in a level 200 Greater Rift. solo.

Join the Conversation

Return to Forum