Derakor's Drop Table

[Daldaen]

I know it's early but I feel like Derakor's drop table is wrong. The loot distribution is supposed to be:

Chestplate (55%) - Common
Boots (24%) - Uncommon
Ear (8%) - Rare

Magelo (http://eq.magelo.com/npc/16718/Derakor-the-Vindicator)

The numbers in parentheses above are the Magelo drop percentages.

Allakhazam (http://everquest.allakhazam.com/db/npc.html?id=5509&p=3#comments)

September 2001:

4) His drops, in order of most to least common, are the Chestplate of Vindication, Boots of the Vindicator and the Living Thunder Earring.



Thus far on blue I think we have see 1 Chest, 1 Boot, 3 Earrings. Also on Beta we saw 2 Earrings, 1 Chest, 1 Boot. I know - Small sample size but I wanted this mentioned so it can be looked at in the code to verify the drop rate is set to be Chest -> Boots -> Ear in common -> rare order.

[Colgate]

red has been 2 chestplates, 1 boots, 1 earring

confirmed more classic everquest experience than blue

[Alunova]

Drop rates are as intended. Small sample size.

[Daldaen]

So 4 Ears, 1 Boots, 1 BP so far on blue.

If the spread was 60% BP/30% Boots/10% Ear which is what the Magelo data suggests, we are looking like that's some very horrible luck.

[heartbrand]

Your sample size is WAY too low

[Daldaen]

The reason I posted was this is like getting 4 Hands of the Reaper off 6 Embalming Fluids. It seems off but it is possible. Alunova confirmed though so it appears fine.

[ddxdy]

Your sample size is WAY too low

Really, because binomial distribution says there is a 0.0002958% chance of getting 5 earrings out of 9 drops.

(9 C 5 )(.08)^5 * (1-.08)^4

You need a big sample size to determine drop percentages.

You don't need a big sample size to determine the rarity of an event occurring.

[Samoht]

Really, because binomial distribution says there is a 0.0002958% chance of getting 5 earrings out of 9 drops.

(9 C 5 )(.08)^5 * (1-.08)^4

You need a big sample size to determine drop percentages.

You don't need a big sample size to determine the rarity of an event occurring.

No, it's a steady 8% chance across the board. The likelihood might be less than 1%, tho.

[ddxdy]

No, it's a steady 8% chance across the board. The likelihood might be less than 1%, tho.


In probability theory and statistics, the binomial distribution with parameters n and p is the discrete probability distribution of the number of successes in a sequence of n independent yes/no experiments, each of which yields success with probability p.


In this case success is an earring dropping.

The formula is: https://upload.wikimedia.org/math/1/d/5/1d53e4a0d17e17adfeb7f53ecca9df5d.png

Probability or p of a "yes" is 8% or .08.


https://upload.wikimedia.org/math/1/d/5/1d53e4a0d17e17adfeb7f53ecca9df5d.png = (9 C 5 )(.08)^5 * (1-.08)^(9-5) = 0.0002958% chance of getting a "yes" 5 out of 9 times with a probability of 8%.


It's fully possible that it was just "bad luck," but chances are there is something else at play. For comparison, the odds of getting struck by lightning in your life are .00033%. You literally have a better chance of getting struck by lightning than getting 5 earrings out of 9 kills @ 8% drop rate.

[Samoht]

Chance of earring dropping does not decrease after the first earring drop, just like it does not increase if the earring does not drop. Flat 8% each time.

Chance of earring dropping 5 out of 9 times is 8%.

8% chance it drops again next time, too.

[ddxdy]

Chance of earring dropping does not decrease after the first earring drop

No, it sure doesn't, and that's exactly why the probability is so low. If you flip a coin, 50% chance it lands on heads. If you flip a coin again, 50% chance it lands on heads. If you flip 10 coins, there is not a 50% chance that all 10 flips will be heads. Hence binomial distribution theorem.

[Samoht]

No, it sure doesn't, and that's exactly why the probability is so low. If you flip a coin, 50% chance it lands on heads. If you flip a coin again, 50% chance it lands on heads. If you flip 10 coins, there is not a 50% chance that all 10 flips will be heads. Hence binomial distribution theorem.

RNJesus does not abide by any rules written by mortal men.

[ddxdy]

RNJesus does not abide by any rules written by mortal men.

You should buy a lottery ticket. You either win or you don't. 1 out of 2 of those will happen, so you have a 50% chance of winning the lottery.

[Samoht]

You should buy a lottery ticket. You either win or you don't. 1 out of 2 of those will happen, so you have a 50% chance of winning the lottery.

Lottery is not Boolean. And there's more than 2 numbers involved.

And if my chances to win were 1 in 60 million, my chances would not go up or down if I played three consecutive drawings in a row without winning.

[ddxdy]

Lottery is not Boolean. And there's more than 2 numbers involved.

And if my chances to win were 1 in 60 million, my chances would not go up or down if I played three consecutive drawings in a row without winning.

But your chances of winning three times in a row are not 1 in 60 million.

[Samoht]

But your chances of winning three times in a row are not 1 in 60 million.

I'm pretty sure that at this point neither one of us knows what your point is.

[ddxdy]

I'm pretty sure that at this point neither one of us knows what your point is.

My point is exactly as I first stated. If the earring is an 8% drop rate chance, the chance of it dropping 5 out of 9 times is .00023%. NOT 8%.

Yes, one drop rate is 8%. But in a collection of drop rates, the chances of getting an 8% drop 5 out of 9 times is SIGNIFICANTLY lower. In fact, it's .00023%.

[Ciroco]

Chance of earring dropping 5 out of 9 times is 8%.

wat

[Samoht]

My point is exactly as I first stated. If the earring is an 8% drop rate chance, the chance of it dropping 5 out of 9 times is .00023%. NOT 8%.

Yes, one drop rate is 8%. But in a collection of drop rates, the chances of getting an 8% drop 5 out of 9 times is SIGNIFICANTLY lower. In fact, it's .00023%.

Which you claim is more common that lightning striking. Something with your napkin math is obviously very, very wrong.

Especially since staff has already confirmed the drop rate is correct.

The fact of the matter is, you simply cannot measure "collection of drop rates." Get a better sample size or get out of here with your anecdotal evidence.

[ddxdy]

Which you claim is more common that lightning striking. Something with your napkin math is obviously very, very wrong.

Especially since staff has already confirmed the drop rate is correct.

The fact of the matter is, you simply cannot measure "collection of drop rates." Get a better sample size or get out of here with your anecdotal evidence.

https://en.wikipedia.org/wiki/Binomial_distribution

[Samoht]

https://en.wikipedia.org/wiki/Binomial_distribution

Why do you keep linking things you clearly don't even comprehend?

[ddxdy]

Why do you keep linking things you clearly don't even comprehend?

Enlighten me how it doesn't apply then.

[Samoht]

Enlighten me how it doesn't apply then.

Sample size too small. Everybody has been saying it all along, you're just ignoring it.

[Daldaen]

It is rather unlucky to roll a 1/10 (10%) 4 out of 6 times though. Thats all.

[Samoht]

It is rather unlucky to roll a 1/10 (10%) 4 out of 6 times though. Thats all.

Unlucky, maybe, illogical? Hardly. Plus, you're not even considering the drop rate from red.

[Daldaen]

5/10 with 10% drop - Again unlucky.

[Raev]

It's painful to read some of the math illiteracy in this thread.

Also while dx's math is correct, the P99 drop rate is probably higher than 8%. If the drop rates were 60% / 25% / 15%, then the math is 126 * 0.15^5 * 0.85^4 = 0.5% Which is still quite unlikely and worth a bug report, but the reality is that rare events happen all the time.

[Samoht]

It's painful to read some of the math illiteracy in this thread.

Or in this one (http://www.project1999.com/forums/showthread.php?t=92423&highlight=charm). Whoever made that post can't even count to 11.

But of course, if anybody would come in here and support someone jumping to conclusions over 10 total kills, it would be someone just as simple minded.