Is this lich ephemera normal?

[Gundamu]

A friend of mine encountered a fire lich with a (fire?) ephemera. Usually, the ephemera should look like a regular fire (yellows/oranges) - to my knowledge.

I don't know much about the lich system. Is this intended?

Edited March 27, 2020 by Phosis

[Gundamu]

I deeply apologize for this post, and don't know how to delete it. A 1 second image search showed results (this ephemera is in fact normal). Sorry to waste anyone's time.

[(PSN)AbBaNdOn_]

Punch your friend in the face for me.  I've popped 30+ liches with fire and no ephemera yet =/  

[Zammara]

It's not a waste of time, don't worry! I also got puzzled when I saw ephemeras having different colors despite their elements. In the end, liches are just like us, they appreciate some glamour from time to time as well! (heck, there's even the quirk "Vain" out there!)

[Sunai_Moonswing]

First time for everything! No problem! 

1 hour ago, (PS4)AbBaNdOn_ said:

Punch your friend in the face for me.  I've popped 30+ liches with fire and no ephemera yet =/  

For me, it's ice, and every time I got tired of ice and switch to another element, yep. Ephemera. 

[ThorienKELL]

On 2020-03-28 at 12:15 AM, (PS4)AbBaNdOn_ said:

Punch your friend in the face for me.  I've popped 30+ liches with fire and no ephemera yet =/  

Liches have 20% chance of ephemera. Which means on average 1 lich in 5 will have it.
Which means that only 3% of ppl won't get it after 30 liches. If I calculated probability right. 
In other words, RNG Jesus hates you : ) Oh just to add insult to injury, I've got ephemera on my FIRST lich, back when chance was only 10%

[Zakkhar]

On 2020-03-29 at 11:34 PM, ThorienKELL said:

Liches have 20% chance of ephemera. Which means on average 1 lich in 5 will have it.
Which means that only 3% of ppl won't get it after 30 liches. If I calculated probability right. 

You calculated it wrong. Statistics is a science of big numbers, you are attempting to apply it to small numbers.

Note that you started your math from approximation (average), but you ended it on certainty. That is not how statistics works.

Plus your math is wrong anyway. This is not a variation or a combination, those are independent rolls. 

Chance of NOT getting and ephemera in a lich is 80%, so probability is 0.8. In next one it is again 0.8, so it adds up to 0.8*0.8.

Probability of NOT getting at least one Ephemera over 30 liches is 0.8 to the power of 30, which is 0.00123794003.

But that doesnt mean than only 0.12% of people will not get it after 30 liches. You can appoximate that many results in big player base, but it is not a fact.

Edited March 31, 2020 by Zakkhar

[ThorienKELL]

4 hours ago, Zakkhar said:

You calculated it wrong. Statistics is a science of big numbers, you are attempting to apply it to small numbers.

Note that you started your math from approximation (average), but you ended it on certainty. That is not how statistics works.

Plus your math is wrong anyway. This is not a variation or a combination, those are independent rolls. 

Chance of NOT getting and ephemera in a lich is 80%, so probability is 0.8. In next one it is again 0.8, so it adds up to 0.8*0.8.

Probability of NOT getting at least one Ephemera over 30 liches is 0.8 to the power of 30, which is 0.00123794003.

But that doesnt mean than only 0.12% of people will not get it after 30 liches. You can appoximate that many results in big player base, but it is not a fact.

Oh ty. I was in doubt, see my method, out of from top of my head was:
"you have 100% chance to get it in 5 liches ... so in next 5 you have 50% chance of not getting any ... in next 5 25% ..." 
I treated whole event like coin toss. Which is wrong.

Yea I understand how probability works, it's not that I attempt to apply it to small numbers, we can still extrapolate the chance, chance does not guarantee anything. It just approximately measures how unlucky you are : ) 

 

[Zakkhar]

5 hours ago, ThorienKELL said:

Oh ty. I was in doubt, see my method, out of from top of my head was:
"you have 100% chance to get it in 5 liches ... so in next 5 you have 50% chance of not getting any ... in next 5 25% ..." 
I treated whole event like coin toss. Which is wrong.

That is not a coin toss. I used a coin toss.

I do not understand where you get the following assumption from:

"you have 100% chance to get it in 5 liches"

You dont, with 20% chance of obtaining one, 5 tries gives you 67%, if you were looking for "nearly guaranteed" number of liches that would be, and that is around 30.

And that one:

"so in next 5 you have 50% chance of not getting any"

Why? How do your previous liches having or not having ephempera affect the chance of your next liches having or not having ephemera? It is independent coin toss. The chance stays the same.

 

Edited April 1, 2020 by Zakkhar

[ThorienKELL]

7 hours ago, Zakkhar said:

That is not a coin toss. I used a coin toss.

I do not understand where you get the following assumption from:

"you have 100% chance to get it in 5 liches"

You dont, with 20% chance of obtaining one, 5 tries gives you 67%, if you were looking for "nearly guaranteed" number of liches that would be, and that is around 30.

And that one:

"so in next 5 you have 50% chance of not getting any"

Why? How do your previous liches having or not having ephempera affect the chance of your next liches having or not having ephemera? It is independent coin toss. The chance stays the same.

 

Curious. 
If ephemera has 20% drop chance, and on average drops once in 5 liches, is it wrong to say that you will get one in 5 liches on average? I mean clearly it is but I'm tying to track back steps of mistake. 

Oh I know ... I treated chances as additive and they are multiplicative. Right?

Oh, previous liches affect your chances if you calculate the whole event:

If you toss coin chances are 50/50 always, yes.
If you toss coin 10 times, chances to get heads 10 times in a row diminishes. 
So if you've got tails 9 times, chance to get either remains the same on next toss: 50/50.
But your chances of getting 1 side 10 times in a row is still low.

Edited April 1, 2020 by ThorienKELL

[Zakkhar]

On 2020-04-01 at 9:28 AM, ThorienKELL said:

If ephemera has 20% drop chance, and on average drops once in 5 liches, is it wrong to say that you will get one in 5 liches on average?

Yes, because it is plain untrue. Especially if you then take that approximation and try to build your model on it.

i am gonna break the math for you (like they do for kids learning this topic):

1st lich has 20% chance of having ephemera. If we had only 1 try that would be the chance. To move to the second one we have to assume he did not have it, that is 80% chance of happening.

2nd lich has 20% to have it. If it has it we end our model here. To move the the 3rd one we have to assume he didint have it, that is 80% chance.

3rd has 20% to have it. If it has it we end our model here. To move the the 4th one we have to assume he didint have it, that is 80% chance.

4th one has 20% to have it.  If it has it we end our model here. To move the the 5rd one we have to assume he didint have it, that is 80% chance.

5th one has 20% to have it.

Chance of obtaining ephemera on 1st try is 0.2.

Chance of obtaining an ephmera on 2nd try provided that 1st try was a fail is 0.8*0.2.

Chance of getting it on 1st OR 2nd try is 0.2+0.8*0.2

Chance of obtaining an emphemera on 1st, 2nd or 3rd try is 0.2+0.8*0.2+0.8*0.8*0.2

Chance of obtaining an emphemera on 1st, 2nd , 3rd or 4th try is 0.2+0.8*0.2+0.8*0.8*0.2+0.8*0.8*0.8*0.2

Chance of obtaining an ephemera on any of the 5 tries is 0.2+0.8*0.2+0.8*0.8*0.2+0.8*0.8*0.8*0.2+0.8*0.8*0.8*0.8*0.2=0.67

You may see why it is wrong to say that one gets 1 on 5 liches. Chance is only 67%, not 100%, not even 99%.

According to this you can say that 2 of 3 people will get at least one in 5 liches.

 

On 2020-04-01 at 9:28 AM, ThorienKELL said:

Oh, previous liches affect your chances if you calculate the whole event:

If you toss coin chances are 50/50 always, yes.
If you toss coin 10 times, chances to get heads 10 times in a row diminishes. 

It doesnt. It is the same as if you tossed the coin 100 times and picked any 10 rolls. In a row has no effect on coin toss.

It is always 50/50. It would change if you had a finite pool. Lets say had 10 M&M and 1 was poisonous, because as you draw the M&M the total amount of non-poisonous M&M would change so it would change the odds of drawing the poisonous one.

Also note that we do not talk about getting head or tails 100% of the time. We talk about getting at least one success, when drawing from inifnite pool (at given amount of tries).

 It's a different formula (much simpler one). Without factorial. 

It's really really basic math. Not the time or place to teach you.

Edited April 2, 2020 by Zakkhar

[ThorienKELL]

13 hours ago, Zakkhar said:

Yes, because it is plain untrue. Especially if you then take that approximation and try to build your model on it.

i am gonna break the math for you (like they do for kids learning this topic):

 

 

It doesnt. It is the same as if you tossed the coin 100 times and picked any 10 rolls. In a row has no effect on coin toss.

It is always 50/50. It would change if you had a finite pool.

Yes, yes, I understand. Thank you. No need to go that elaborate.

About coin toss: you fail to understand that one coin toss may be "an event" and 10 coin toss may be "an event" too. Single event. 
Probability changes with event complexity, although basic /per part probability remains the same.