Strangest RNG

mattress

Any mathematicians? What are the odds of pulling a 5* Angela outta duals 3 times in a row? (About a week apart each). Not a complaint post, just honestly curious about statistics

Uvogin

Assuming 60 champs in the dual crystal pool,
P(Angela,Angela,Angela) is 1 in 21600
but this is misleading as this will the probability of pulling any particular tuple of 3 champs say (venom,groot,ironman).

what you're probably looking for is the odds of pulling the same champ 3 times, that is 1 in 3600.

Yodabolt21

I would say that’s good luck! She is awesome and really benefits from a high sig!

Sean_Who

Let's say there's 70 characters in each. I don't know the exact number but I believe each class has like 30-40?

1 in (70*70*70) or 1:343000 or 0.00029%

Stats was never my strong point but I think that might be close.

mattress

Ebony_Naw said:

Meet me halfway and tell me how many champs are in the dual crystal lol

My last 7-8 five*s have all been duals lookin for tech( only got sentinal,sentinel,, redskins a

Yodabolt21 said:

I would say that’s good luck! She is awesome and really benefits from a high sig!

Not unhappy necessarily but I'm struggling to pull techs and my last 7-8-9 crystals have all been duals with all cosmic.lol. I'm definitely not sour on pulling her like that but the odds seem so unlikely.lol

Thanks_D19

I think the odds are around 1 in 216k to pull Angela specifically 3 times in a row but the odds of pulling the same champ 3 times in a row is more around 1 in 3.6k. The math might be a bit off tho cus idk if I counted the number of champs in the dual class crystal right but it is something around that.

Btw if anyone is wonder how the math works the chance of pulling Angela is 1 in 60 so cube that and bam. And for the other calculation the chance of pulling any champ is 100% but the odds of pulling that same champ two more time is a 1 in 60 chance each so yeah

laserjohn26

I got an A in Stat so pretty simple. You want to know the odds of pulling Angela 3 times in a row. First you must list all the options.

1. Pull Angela 3 times in row
2. Don't pull Angela 3 times in a row

Simple. Only two options so it is 50/50

yuw

I know someone who pulled 6 star rhino three times in a row

mattress

yuw said:

I know someone who pulled 6 star rhino three times in a row

That's fn brutal.lol.hope the odds evened out for him since. I'd quit where I am as a lower-mid level player in that position 😒

Furious_Fighter1

laserjohn26 said:

I got an A in Stat so pretty simple. You want to know the odds of pulling Angela 3 times in a row. First you must list all the options.

1. Pull Angela 3 times in row
2. Don't pull Angela 3 times in a row

Simple. Only two options so it is 50/50

If only my A level Maths Statistics was this easy

Reference

I pull 6* Agent Venom in 3 consecutive pulls last month.

DNA3000

_Sham_ said:

@DNA3000 someone needs you here

@Uvogin and @Thanks_D19 are basically correct except for a typo (216,000 instead of 21,600 in @Uvogin's post). @Sean_Who also using identical calculation but with slightly different assumptions (70 instead of 60).

@laserjohn26 almost correct. You can do it that way, you just have to complete the list. Angela Angela Angela is three Angelas in a row. Angela Angela Blackbolt is not. Angela Angela Black Panther is not. Angela Angela Captain Marvel is not. Angela Angela Colossus is not. Angela Angela Darkhawk is not. The rest is left as an exercise for the reader.

The complex part is why people are mentioning the odds for three of any champ in a row. This one even a lot of math students get wrong. Sometimes, in subtle situations, professional statisticians get this wrong: it is a specific instance of one of the most common type of statistical error. The question is: what are you calculating the odds of, precisely?

If someone says they are specifically interested in the sequence Angela Angela Angela we can calculate the odds of that as being one in 216,000 (assuming 60 options in the crystal). The problem is that the odds of getting Corvus, Kamala, Thor in that order is also one in 216,000. They are both equally unlikely. If you're amazed that the odds against Angela Angela Angela are 216,000 to one, you should be equally amazed that the odds against Corvus, Kamala, Thor is also 216,000 to one.

In this situation, it is very likely that people would find Angela Angela Angela just as amazing as Thor Thor Thor, but not as amazing as Cull, Nova, Gamora. That's because we're focused on the triple, not on Angela herself. So since what we're keying off is the triple and not Angela, the triple Angela is just one of the 60ish possible ways to generate a triple and get noticed. Therefore, Angela Angela Angela is really just a kind of triple, and the odds of seeing a triple at all are one in 3600. AAA just happens to be one of them.

I can change the situation to make Angela Angela Angela far more unlikely. Suppose I have three dual class crystals ready to open, and I specifically *call out* pulling Angela three times. Now, the odds against that happening are one in 216000. That's because if I were to pull Thor three times, I would still have called it wrong. The odds against me if I attempt to call the crystals before opening are much higher than if I let them open and then after the fact call out what happened.

Notice though that if I specifically predict Cull, Nova, and Gamora before opening, the odds against being right are still the same: 216000 to one. If we try to predict the crystals, all sequences are equally unlikely. Angela Angela Angela is not special. It is just as hard to predict and then pull any other sequence. It is only when we open the crystals first, then go back and go "aha!" that the odds change, because now we're exploiting the fact that some sequences look more interesting to human eyes. That's the danger of analyzing probability *after* something has happened. It is very difficult to remove human bias.

In fact, if you open three crystals and Angela shows up three times in a row, the odds of that happening are exactly 100%. That's because it already happened. It is guaranteed to happen since it happened. What people are usually asking is: what were the odds of that happening before it happened. But once it happened, you have to eliminate a subtle source of bias. Had the crystals opened Cull, Nova, Gamora, we wouldn't be here asking what the odds of that happening were. We're only here because the opening "looked interesting." And that in a sense loads the dice in a certain direction, something you have to account for when doing proper calculations.

The odds of pulling Angela three times in a row is one in 216000 (assuming 60 options). But the odds of us all being here discussing it is closer to one in 3600, because most crystal openings are not interesting enough to start forum threads about.

I'll end on this note:

https://www.youtube.com/watch?v=rqHRQdmjdrg

Uvogin

DNA3000 said:

_Sham_ said:

@DNA3000 someone needs you here

@Uvogin and @Thanks_D19 are basically correct except for a typo (216,000 instead of 21,600 in @Uvogin's post). @Sean_Who also using identical calculation but with slightly different assumptions (70 instead of 60).

@laserjohn26 almost correct. You can do it that way, you just have to complete the list. Angela Angela Angela is three Angelas in a row. Angela Angela Blackbolt is not. Angela Angela Black Panther is not. Angela Angela Captain Marvel is not. Angela Angela Colossus is not. Angela Angela Darkhawk is not. The rest is left as an exercise for the reader.

The complex part is why people are mentioning the odds for three of any champ in a row. This one even a lot of math students get wrong. Sometimes, in subtle situations, professional statisticians get this wrong: it is a specific instance of one of the most common type of statistical error. The question is: what are you calculating the odds of, precisely?

If someone says they are specifically interested in the sequence Angela Angela Angela we can calculate the odds of that as being one in 216,000 (assuming 60 options in the crystal). The problem is that the odds of getting Corvus, Kamala, Thor in that order is also one in 216,000. They are both equally unlikely. If you're amazed that the odds against Angela Angela Angela are 216,000 to one, you should be equally amazed that the odds against Corvus, Kamala, Thor is also 216,000 to one.

In this situation, it is very likely that people would find Angela Angela Angela just as amazing as Thor Thor Thor, but not as amazing as Cull, Nova, Gamora. That's because we're focused on the triple, not on Angela herself. So since what we're keying off is the triple and not Angela, the triple Angela is just one of the 60ish possible ways to generate a triple and get noticed. Therefore, Angela Angela Angela is really just a kind of triple, and the odds of seeing a triple at all are one in 3600. AAA just happens to be one of them.

I can change the situation to make Angela Angela Angela far more unlikely. Suppose I have three dual class crystals ready to open, and I specifically *call out* pulling Angela three times. Now, the odds against that happening are one in 216000. That's because if I were to pull Thor three times, I would still have called it wrong. The odds against me if I attempt to call the crystals before opening are much higher than if I let them open and then after the fact call out what happened.

Notice though that if I specifically predict Cull, Nova, and Gamora before opening, the odds against being right are still the same: 216000 to one. If we try to predict the crystals, all sequences are equally unlikely. Angela Angela Angela is not special. It is just as hard to predict and then pull any other sequence. It is only when we open the crystals first, then go back and go "aha!" that the odds change, because now we're exploiting the fact that some sequences look more interesting to human eyes. That's the danger of analyzing probability *after* something has happened. It is very difficult to remove human bias.

In fact, if you open three crystals and Angela shows up three times in a row, the odds of that happening are exactly 100%. That's because it already happened. It is guaranteed to happen since it happened. What people are usually asking is: what were the odds of that happening before it happened. But once it happened, you have to eliminate a subtle source of bias. Had the crystals opened Cull, Nova, Gamora, we wouldn't be here asking what the odds of that happening were. We're only here because the opening "looked interesting." And that in a sense loads the dice in a certain direction, something you have to account for when doing proper calculations.

The odds of pulling Angela three times in a row is one in 216000 (assuming 60 options). But the odds of us all being here discussing it is closer to one in 3600, because most crystal openings are not interesting enough to start forum threads about.

I'll end on this note:

https://www.youtube.com/watch?v=rqHRQdmjdrg

Aah typos, my greatest nemesis...we meet again