Fan Favourite Store arrived

Bad

KGB said:

 The word guarantee is the problem. There is no guarantee when doing random number samples. The rate is 15%, but there is no guarantee you are going to hit it exactly when doing random number sequences no matter how many draws you make and despite the fact the number will converge toward 15% over a very very large amount (millions).

KGB

I will say it in another words: in this game the RNG algorithm takes the pre-rolled cards and it shuffles it.
There is a guaranteed 15% rate. However you never will know how the algorithm works and the order in wich those pulls are coming. Neither you will know in wich intervals the 15% ends and starts the next.
However there is a guaranteed rate in which those 5*s comes.
I deduced that in many hoards I did and in the fact that I can champ regularly almost all 5*s(except in some few cases that rng gives more covers on other characters)
Because that is not how true stadistics works and 15 is a really low amount.
If it wasn't guaranteed it would be impossible to pull 5*s so often.
You talked about each player's different rate pulls. Is that so? Pulling in the thread SC is in LL I unhoarded a good amount of pulls and it was a low rate.
However I didn't write that I pulled several times later and I got a streak of 8 5*s almost in a row.
Now when do you start counting? Do you will say 100 pulls before it was a 9% rate and another 30 pulls later it was a 60%?
No. All was the same algorithm giving you your guaranteed rate.

entrailbucket

It's a 15% chance of a 5* for each pull. 

This stuff about a "guaranteed" rate doesn't make any sense.  No one has explained what the difference is between that and a purely random chance.

Bad

The  game itself is explaining to you in each store when it says 1:7. It is not the same 1:7 as 15%.
However of course you are free to think what you like.

KGB

Bad said:

I will say it in another words: in this game the RNG algorithm takes the pre-rolled cards and it shuffles it.
There is a guaranteed 15% rate. However you never will know how the algorithm works and the order in wich those pulls are coming. Neither you will know in wich intervals the 15% ends and starts the next.

What you wrote here says they take some number of draws (say 10000) and then they make 1500 of those 5* and the rest 4*. Then they shuffle those 10000 draws randomly and let you draw from those 10000 with the idea being you are guaranteed a 15% draw rate if you draw 10000 times.

That's absolutely *NOT* what they are doing. There is no evidence they are pre-rolling a certain number of draws and making *exactly* 15% of those draws 5* tokens.

They are simply doing what entrailbucket suggests above, they are giving 15% chance per draw. Yes, they are storing a LARGE number of pre-draws on the server but the idea that they fixed 15% of them has never even been hinted at and doesn't make sense. If they wanted to go the way you suggest, they just need 100 pre draw and make 15 of them 5* and then every 100 draws, reshuffle the 100 numbers. That gives you a fixed 15% rate every 100 tokens you open (which I am sure just about everyone would love because it means no long streaks).

KGB

HoundofShadow

7 pulls 1 Strange. 

Bad

It is not possible to champ all 5*s regularly with a simple 15% rate.
The Law forces gacha games transparency to state the odds and this one says 1:7, not 15%.
I regularly play and grind another gacha game and having the same amount of pulls monthly. That one works with a 8% and each month the amount of 5*s I get is dramatically different: that is true stadistic.
In here the goal is always fulfilled: that is not purely luck.

entrailbucket

1/7 is about 15% (14.29%)

No one who's tracked extensively over a long period of time has ever noticed any manipulation of the rate in this game.

As humans, we are terrible at randomness and probability (this is why casinos can make money), so without objective tracking, subjective impressions are just not useful.

It's 15%.  That's it.

LavaManLee

Bad said:

It is not possible to champ all 5*s regularly with a simple 15% rate.
The Law forces gacha games transparency to state the odds and this one says 1:7, not 15%.
I regularly play and grind another gacha game and having the same amount of pulls monthly. That one works with a 8% and each month the amount of 5*s I get is dramatically different: that is true stadistic.
In here the goal is always fulfilled: that is not purely luck.

I honestly am not sure what you are getting at here.  Every single chance is 15% of a 5*.  That's it.  There is no "guarantee" other than a guarantee that it is a 15% chance.

As others have said, tossing a coin is a "guarantee" that it is a 50% chance of heads.  It does not "guarantee" you will ever, ever get a head.  It is simply a "guarantee" that you have a 50% chance of getting a heads.

It really doesn't matter what your other gacha games do or don't do.

And, it absolutely IS possible to champ all 5*s regularly with a simple 15% rate.  That is the whole reason of hoarding and trying to make sure you get enough pulls to overcome the 15% chance.  Is it foolproof?  No.  Does it work more often than not?  Yes.

Again, I don't think you quite understand the difference between "guarantee" and the difference between the 15% pull rate.  It is not a "guarantee" that you get anything but it is a "guarantee" that each pull has a 1 in 15 chance of being a 5*.

dianetics

entrailbucket said:

It's a 15% chance of a 5* for each pull. 

This stuff about a "guaranteed" rate doesn't make any sense.  No one has explained what the difference is between that and a purely random chance.

I said this before, but there is no difference if you don't know your seed.
If there was a way to find your specific seed then you could potentially figure out which variable affects where you fall in the algorithm, and then exploit that.
It is a perception of randomness, but it is the result of a mathematical equation.
Obviously, it is on the dev's to balance the algorithm in a manner where the 1:7 stands over rational number of pulls...say 100-200.
I don't know who would rely on the guaranteed idea since it is chance based. The only guarantee is a 4* or better.

Bad

entrailbucket said:

As humans, we are terrible at randomness and probability (this is why casinos can make money), so without objective tracking, subjective impressions are just not useful.

It's 15%.  That's it.

Subjective? I didn't stopped on giving reasons. 
There is no rating manipulation: the game gives what it says.
However players didn't got it.
Another fact(all of you are free to believe it or not): I've often read that it is required 320 pulls for to champ all 3 LLs. It is a good assumption based on stadistics. 
However in 2 occasions I champed all 3 with 180 pulls.
That cannot be explained with a 15% rate except being hyper lucky.
However it refutes my theory on the algorithm and it explains the ~1-:7.
Think about it

entrailbucket

320 pulls aren't *required* because the 15% chance is random.  I'm guessing that 320 is a recommendation based on someone working out the odds.  One player might do it in 320 pulls, another in 201, another in 180, or 150.  Another player might pull 50 straight 5*. 

It's random.

LavaManLee

Bad said:

entrailbucket said:

As humans, we are terrible at randomness and probability (this is why casinos can make money), so without objective tracking, subjective impressions are just not useful.

It's 15%.  That's it.

Subjective? I didn't stopped on giving reasons. 
There is no rating manipulation: the game gives what it says.
However players didn't got it.
Another fact(all of you are free to believe it or not): I've often read that it is required 320 pulls for to champ all 3 LLs. It is a good assumption based on stadistics. 
However in 2 occasions I champed all 3 with 180 pulls.
That cannot be explained with a 15% rate except being hyper lucky.
However it refutes my theory on the algorithm and it explains the ~1-:7.
Think about it

Bad, it does not refute your theory at all.  It absolutely can be explained with a 15% rate.  I am getting you do not work with statistics for a living and, if so, I don't know any other way to convince you.  Every pull is a 1:7 chance.  Every pull.  Doesn't matter if it's done ahead of time or at the time, but it is a 1:7 pull.

In theory, every single pull you make for the rest of your life COULD be a 5*.  Or, every single pull you make for the rest of your life COULD be a 4*.  It is completely random and completely 1:7.

Put it this way.  Roll a die.  In theory, if you had infinite rolls, you could roll a 6 for infinite times.  But you probably won't.  But you could.  This is the same.

Your logic above absolutely CAN be explained.

Bad

2 times? No. One time could be but 2 times not.
With a 15% you can champ all 3 or not because as you say you can pull 4*s until you die.
However : A/to champ all 3 LL 2 times and saving 100 pulls before the goal if it was a 15%
B/ to champ regularly almost all the characters and some of them with many covers beyond lev 450 before they exit LL
That is not stadistics you can say what you want.

LavaManLee said:

Bad, it does not refute your theory at all.  It absolutely can be explained with a 15% rate.  I am getting you do not work with statistics for a living and, if so, I don't know any other way to convince you.  Every pull is a 1:7 chance.  Every pull.  Doesn't matter if it's done ahead of time or at the time, but it is a 1:7 pull.



Your logic above absolutely CAN be explained.

If you know so much about stadistics you could calculate the probability of champing 3 characters, more than 40 5* covers on a 15% rate, in 200 pulls, and do it 2 times.
Hint: it's not a 15%.

MadScientist

The probability to get 40 or more 5-Star covers from 200 LTs at a 15% rate is 3.355%. 

Note: you just have to sum a binomial distribution. In Excel just do 1-BINOM.DIST(39,200,15%,true)

LavaManLee

MadScientist said:

The probability to get 40 or more 5-Star covers from 200 LTs at a 15% rate is 3.355%. 

Note: you just have to sum a binomial distribution. In Excel just do 1-BINOM.DIST(39,200,15%,true)

Exactly.  Is it possible?  Yep.  Is it probable?  Nope.

LavaManLee

Bad said:

If you know so much about stadistics you could calculate the probability of champing 3 characters, more than 40 5* covers on a 15% rate, in 200 pulls, and do it 2 times.
Hint: it's not a 15%.

As stated in the post by MadScientist, it is 3.355%.  Not sure what your point is, though.

Bad

Wow. I must buy a colection of lotto tickets, because that probability is easier than to win a lot of types of lotto.
And I must be the only player experiencing this because the things that I'm saying seem to not ring any bell in no other players. 
Perhaps I must contact support because it seems something happens with my account and my 5* rate, too.
Unfortunately that it's the only experience I have playing this game day to day from almost 4 years so I cannot be convinced from my personal experiencie.
Perhaps I should try to hoard another 200 pulls and champ another 3 LLs. That way surely I would decrease that rating from 3.355% to perhaps a 1.xxxx.

LavaManLee

Honestly, I don't know what is so difficult to understand.  You did something that is possible but improbable.  Happens all the time  Every winner of the lottery.  There is nothing you did that is out of the bounds of possibility.  But you continue to act as if it can't happen.

Sekilicious

LavaManLee said:

Bad said:

If you know so much about stadistics you could calculate the probability of champing 3 characters, more than 40 5* covers on a 15% rate, in 200 pulls, and do it 2 times.
Hint: it's not a 15%.

As stated in the post by MadScientist, it is 3.355%.  Not sure what your point is, though.

Because I cannot stop from adding my understanding, 3% means that if 100 people do something or you do it 100 times (in this case try to pull 40 5* covers in less than 200 pulls at a 15% rate) you should expect that, on average, 3 of the 100 to succeed and 97 to fail. To put it another way, if the lottery odds are 1 in 17 million then, again on average, you should only expect someone to win if 17 million people play or you played 17 million times. Frankly, pulling 40 at less than 200 twice isn’t even that big of a deal considering the number of people that play. It happens all the time, just as the opposite (whatever the equivalent of not pulling 40 in 320 is) does.  Just like the lottery (I do not expect myself to win, but I expect someone to). The only time you should ever expect something not to ever happen is if there is a 0% chance. It is difficult to intuitively understand this because experience generally works with very small samples. 

entrailbucket

dianetics said:

entrailbucket said:

It's a 15% chance of a 5* for each pull. 

This stuff about a "guaranteed" rate doesn't make any sense.  No one has explained what the difference is between that and a purely random chance.

I said this before, but there is no difference if you don't know your seed.
If there was a way to find your specific seed then you could potentially figure out which variable affects where you fall in the algorithm, and then exploit that.
It is a perception of randomness, but it is the result of a mathematical equation.
Obviously, it is on the dev's to balance the algorithm in a manner where the 1:7 stands over rational number of pulls...say 100-200.
I don't know who would rely on the guaranteed idea since it is chance based. The only guarantee is a 4* or better.

I meant to address this the other day and forgot. 

It is highly, highly unlikely that they're using a seeded pseudorandom algorithm to determine these pulls.  That kind of random number generator fell out of fashion decades ago (I remember the rand() function on my graphing calculator was quite predictable!).

These days, random number generators have gotten much better, to the point that even the worst ones are basically indistinguishable from true randomness.  The most basic pseudorandom algorithms currently in use are essentially "seeded" with something like electronic noise generated by some hardware component in the server. 

The more advanced ones use a natural, randomly occurring phenomenon, like cosmic background radiation or radioactive decay.  These algorithms are actually truly random by the mathematical definition, and they're accessible to everyone now, via the magic of the internet!

Unless they've somehow chosen to use an algorithm from the 60s or 70s, there is no practical way to reverse-engineer it.