Beating up the golden number

Options
Bad
Bad Posts: 3,146 Chairperson of the Boards
in MPQ General Discussion
As many of you may know, in my (closed) thread the LL algorithm I developed a kind of a theory based on my experience pulling, which was fiercely debated.
However now I have a sweet new toy (oh yes we have it, my precious. Very nice, very nice).
I've searched for an online generator number with repetitions. Sadly I didn't thought about it before.
https://nosetup.org/php_on_line/numero_aleatorio  (there must be on English too out there).

So giving him the numbers 1 to 9(powers of 3 characters) and a number of tries (successful 5* covers) I can simulate a hoard.
On my thread I pulled and recorded a hoard which gave me 36 5* covers and I champed 2 characters.
So I tried it with my new friend, the online number generator with repetitions. On 10 attempts how many times I could champ 2 hypothetical characters with 36 successful covers? The answer is none.

Then I thought to attempt the golden number, 300 pulls, widely accepted, which will give 45 successful 5*s and will champ the 3 characters.
So I pulled the lever and my new good friend generated the numbers:
432683555944713896731899557111688915178737296
Pulling them into a result depicting 3 hypothetical characters would be:
1. 7
2. 2
3. 5
4. 3
5. 6
6. 4
7. 6
8. 6
9. 6
Only 1 character champed! The final one. And almost the first one.
It must be a flush. Second attempt:
267427796668377419952346283113174233441567947
And putting into the familiar graphic:
1. 5
2. 5
3. 6
4. 7
5. 2
6. 6
7. 8
8. 2
9. 4
Another time only one character champed! The first one and almost the third.
Let's speed things up. 4 attempts more:
A: 218279112712351874116946629792186764729849879
B: 751877328571558584977563764928255223387644137
C: 839975117567973577944236214774692356197653178
D:483257279891465917949374169941923814169283388
    A  B.  C   D
1. 8/ 3/ 5/ 6
2. 7/ 5/ 3/ 4
3. 1/ 5/ 5/ 5
4. 4/ 4/ 4/ 6
5. 1/ 8/ 5/ 2
6. 5/ 3/ 5/ 3
7. 7/ 9/ 10/ 4
8. 5/ 6/ 2/ 6
9. 7/ 2/ 6/ 9
Awesome! The only attempt champing 2 characters are the C and D!
But how is this possible? 
By the way it's simple. Like I said on my previous thread, all the numbers are in the bag and they could appear in a total random way. However appearing all the numbers 5 times is a really odd result.

Resuming: champing 1 character 100%, champing 2 characters 33.3%, champing 3 characters 0%. (You'll understand I won't try it more times. That was really fast but actually it takes a lot of time transcribing all the numbers, counting them many times, reread all the typos, and restarting again because of losing concentration).

So the golden number has been defeated 6 times!
Would that mean that the widely accepted golden number 300 pulls is false? No, not at all. Because when hoarding:
-For crystal, MK and Elektra a forum user ( I won't say the name, it could be a reson to close this thread :P ), got C 4/6/4 E 5/5/2 and MK 4/2/4, 34 5*s on 248 pulls (bad odds)
- For Gargantos,MT and KK a user got Gargantos: 12/4/7 Kamala: 6/4/6
Dr. Thor: 7/2/2, 50 5*s (golden number) on 255 pulls.
-On my LL tread AB, Elektra and Crystal was AB 3/6/5  E 5/4/4 C 3/3/3, 36 5*s on 246 pulls.
-For gargantos, crystal,MK was Crystal 6/2/3
MK 7/4/3 G 2/7/6, 252 pulls 41 5*s.
-For SC, Electro and Odin I could champ the first 2 in less than 230 pulls
- For sersi, Big Wheel and gamora I could champ the ladies on less than 250 pulls
-Another user previous 259 pulls with 39 5*s he champed sersi and big wheel and almost  ABrand.

Final conclusion: my new buddy seems to have, by far, a lot less luck than actually real pulls on game and amazingly with less covers.
Hey, that reminds me, there wasn't a guy out there with a crazy theory? Now what was his name? Bud, Ban, Ball, something like that...

Comments

  • KGB
    KGB Posts: 3,066 Chairperson of the Boards
    Options
    I don't really understand the point of this topic.

    - got C 4/6/4 E 5/5/2 and MK 4/2/4, 34 5*s on 248 pulls (bad odds)

    This failed to champ all 3

    - Gargantos: 12/4/7 Kamala: 6/4/6 Dr. Thor: 7/2/2, 50 5*s (golden number) on 255 pulls.

    This also failed to champ all 3

    - AB 3/6/5  E 5/4/4 C 3/3/3, 36 5*s on 246 pulls.

    This also failed to champ all 3

    - Crystal 6/2/3 MK 7/4/3 G 2/7/6, 252 pulls 41 5*s.

    This also failed to champ all 3


    Anyway, the golden number also assumed you set your shard targets to the characters you were trying to cover. Doing that allows you to generate at least 1 more colorless cover for each character (you adjust as you pull since once you cover the 1st character of your 3, you don't need to target shard that one any more).

    Lastly the golden number is not a 100% guarantee. It's just a confidence number as in you are 80% confident you will cover all 3 with 300 pulls + shards. If you want to be 90% confident you need to increase the draws to about 400 and so on. These numbers are posted in older threads.

    KGB

  • Bad
    Bad Posts: 3,146 Chairperson of the Boards
    edited July 2022
    Options
    KGB said:
    I don't really understand the point of this topic.


    Anyway, the golden number also assumed you set your shard targets to the characters you were trying to cover. Doing that allows you to generate at least 1 more colorless cover for each character (you adjust as you pull since once you cover the 1st character of your 3, you don't need to target shard that one any more).


    KGB

    I think the point is clear. No hoard with 300 pulls champ the 3, but in game pulls are more frequent to champ 2 characters than my online rng and with less than 45 covers.
    Sorry, but the golden number was before shards and doesn't assumes you shardfavorite nothing. 
  • TheXMan
    TheXMan Posts: 162 Tile Toppler
    Options
    His theory is that there is some code helping distribute covers when you open a large horde.  Not sure at what point this supposedly kicks in.  No matter how many examples that are given showing bad spreads, he is convinced it is not completely random.
  • Bowgentle
    Bowgentle Posts: 7,926 Chairperson of the Boards
    Options
    TheXMan said:
    His theory is that there is some code helping distribute covers when you open a large horde.  Not sure at what point this supposedly kicks in.  No matter how many examples that are given showing bad spreads, he is convinced it is not completely random.
    He's also the only one subscribing to that, because he thinks his sample size of three proves his theory.
  • jsmjsmjsm00
    jsmjsmjsm00 Posts: 268 Mover and Shaker
    Options
    I really don't get this thread or topic either. I don't get the claims that doing this kind of math is something fancy either. This is basic statistics. I've made the same thing in Excel in 10 minutes before I broke my first hoard to procrastinate at work. 

    I can take literally 5 seconds spamming refresh on my Excel sheet and generate more samples than are presented in the parent post.

    It's great that people are interested in math, don't get me wrong, but please stop making LLs out to be some kind of conspiracy/magic/overly complex formula. 
  • Bad
    Bad Posts: 3,146 Chairperson of the Boards
    Options
    Well the other thread was like everything was perfectly normal and that's how combinatory works (champing 2 characters with 36 covers lol).
    This one the argument is I don't get the point of this thread. 
    I took one step forward clearly  :D:#
    No problem I will continue registering hoards until the evidence will be bigger with normal maths and hoard pullings.
  • TheEyeDoctorsWife
    TheEyeDoctorsWife Posts: 829 Critical Contributor
    Options
    I’ve made 183 pulls the last 3 months . The last 5 released 5* have 13,14,11,9,7 covers from oldest to newest. You can see it’s at least possible with all the new cover bonuses that an average of 70 pulls a month plus high PvE 10 placement is now all it takes to champ a 5 before it leaves . This is not based on any math equation, just personal play style which can vary substantially 

Categories