There's something wrong with probability！！！

[圣子到]

Not long ago, your company made the news of probability fraud, and now the new game, 20% of the drawings, I believe that everyone must brush at least 10-20 times to get, do you think that we do not understand math?

The mathematical expectation is 5 drops, when you drop at least 10 more times, then the drop rate is undoubtedly 10% or less!!！

[Mr.Mike]

yeah but you gotta understand, the drop rate is instance based, so the percent doesn't increase the more you do missions, IE, an item with a 10% drop rate will remain at that rate no matter how many times you do the mission, so doing it 10 times doesn't guarantee the drop. I feel your pain though man, did a mission over 70 times today to get a module that had a 10% drop rate from the mission before it even dropped.

[p00se2]

20 % chance

is 80% chance of not getting the item - every time

[TripleKillGG]

Yeah just go farm, log off close the client, go run it again and you can get it multiple people talking about it, its how a lot are abusing their ♥♥♥♥♥♥♥ system they got fined on a long time ago.

[Kyutaru]

Originally posted by ZY-china:

Not long ago, your company made the news of probability fraud, and now the new game, 20% of the drawings, I believe that everyone must brush at least 10-20 times to get, do you think that we do not understand math?

The mathematical expectation is 5 drops, when you drop at least 10 more times, then the drop rate is undoubtedly 10% or less!!！

Actually, the mathematical expectation after 5 runs is that there's a 32% chance you DIDN'T get the item yet. You seem to be adding percentages together. That's not how probability works.

Since you have a less than 70% chance after 5 runs, what are you chances of having obtained it after 10 runs? It's less than 90%. Meaning still a 10% chance you still don't have it.

10% of the playerbase is going to need more than 10 runs to get these items on average.

[Maya-Neko]

Originally posted by ZY-china:

Not long ago, your company made the news of probability fraud, and now the new game, 20% of the drawings, I believe that everyone must brush at least 10-20 times to get, do you think that we do not understand math?

The mathematical expectation is 5 drops, when you drop at least 10 more times, then the drop rate is undoubtedly 10% or less!!！

Yeah, i (and Nexon probably as well) actually think, that you don't understand the math behind, because your explanation is extremely lackluster and hints towards you expecting it to be additive and that's simply not how probability works.

You can't use one single drop to determine, if something is adding up or not. if you try 10 times, then you've still a 10 % chance that you still don't have the item afterwards. Even with 20 tries, 1 out of 100 people still don't get the item (considered the amount of people who're currently playing, that's still a huge amount of people and you will eventually be one of them).

To truly determine if a drop rate is rigged or not, you should rather get at least 100 drops to get an estimation of how close it is to the stated drop rate. Make it like 1k or even 10k drops and you're probably getting within like less than 1 % near the actual drop rate.

[Kyutaru]

Originally posted by SomeDude:

yeah but you gotta understand, the drop rate is instance based, so the percent doesn't increase the more you do missions, IE, an item with a 10% drop rate will remain at that rate no matter how many times you do the mission, so doing it 10 times doesn't guarantee the drop. I feel your pain though man, did a mission over 70 times today to get a module that had a 10% drop rate from the mission before it even dropped.

He's not measuring individual instances, he's measuring sets. Dice probability uses the same logic. Rolling an individual dice gives you 1/6 odds for any particular number. But rolling 2 dice and trying to get a 12 or a 2 results in a 1/36 chance. This is because the odds of each die getting a specific combination (two 6s or two 1s in this case) are 1/6 multiplied by 1/6. This is because for all possible dice combinations, any 6 results on 1 die are paired with any 6 results on the 2nd die. i.e. 6x6 = 36. You can test it yourself by checking for every possible combination of die results. You'll find that 12 and 2 only have 1 possible combination each out of 36 possibilities.

Similar to how if you flip three coins, there are 8 possible combinations of coin flip results, 7 of which contain at least one heads (only all Tails doesn't) so you have a 7/8 chance to flip at least one Heads, meaning an 87.5% after three flips of getting Heads (not 50%).

Individual runs don't change, and each coin flip will still be 50/50, but the odds change the more iterations you're checking against because the resulting pool gets larger and more complicated.

The odds for rolling Failure are 80%. After 5 sets, that's 80%*80%*80%*80%*80% that you still didn't get the item (aka, 32%). After 10 sets, it's about 10%, meaning a 90% chance that you rolled the item in those 10 attempts.

[Simulacrum111]

For the common person these explanations mean very little. Most are going to assume the most simplistic explanation.

I am in the camp that the drop rates are not accurate. They are not inaccurate because the stated rates are incorrect. The are not accurate because of what is not disclosed.

Is the rate fixed?
In what context(s) does it apply?
Does player activity affect the drop rate?
Is there a mercy rule?

Nexon has never claimed that the percentages given to players are the only factors to take into account and that the rate does not change for any reason they elect it should.

I guess I should say I don't feel the complete explanation of the drop systems has been or ever will be provided to players for a host of reasons.

[Maya-Neko]

Originally posted by Kyutaru:

He's not measuring individual instances, he's measuring sets. Dice probability uses the same logic. Rolling an individual dice gives you 1/6 odds for any particular number. But rolling 2 dice and trying to get a 12 or a 2 results in a 1/36 chance. This is because the odds of each die getting a specific combination (two 6s or two 1s in this case) are 1/6 multiplied by 1/6. This is because for all possible dice combinations, any 6 results on 1 die are paired with any 6 results on the 2nd die. i.e. 6x6 = 36. You can test it yourself by checking for every possible combination of die results. You'll find that 12 and 2 only have 1 possible combination each out of 36 possibilities.

Though, in case of the dice, we're not trying to look at a combined value, but rather look at every instance individually. That said, if we try to get at least one 1 rolled, then rolling 1 dive has a 1/6-chance. When we roll 2 dices, then we're already at a 11/36-chance to get at least one 1-roll. We don't search for a 2 or a 12, we explicitly just search for a 1, no matter how many dices we keep rolling.

And that's the problem some people seem to have with probability: They assume, that one run can influence the next roll. They expect a specific pattern and get confused, when the RNG doesn't follow it.

You can tell a person to toss a coin 1000 times and write down the sequence of what they've thrown and you can usually tell, if they've actually threwn them or just written down something they think is random, because when people write down random sequences, they're more likely to change the next result with a significantly higher likelyhood than the 50 % of a coin.

[[SHH] Unreal4Real]

Lots of scandals over this and your frustration is understandable. Nexon is famous for being sketchy with drop rates. I don't think TFD has issues though. You just gotta realized, most stuff has a 2%-5% drop rate. Every single time you attempt it that rate is the same, it doesn't increase the more you do it. So.... statistically.... you need to grind ALOT to get an item with a drop rate below 10%

[Min/Max]

I have already farmed Freyna, Sharen, Valby and Blair. I had better chance with the drops coming back in fresh rather than continues grind over and over.

[Maya-Neko]

Originally posted by Simulacrum111:

For the common person these explanations mean very little. Most are going to assume the most simplistic explanation.

I am in the camp that the drop rates are not accurate. They are not accurate because the stated rate is incorrect. The are not accurate because of what is not disclosed.

Is the rate fixed?
In what context(s) does it apply?
Does player activity affect the drop rate?
Is there a mercy rule?

Nexon has never claimed that the percentages given to players are they only factors to take into account and that the rate do not change for any reason.

I guess I should say I don't feel the complete explanation of the drop systems has been or ever will be provided to players for a host of reasons.

Why should you assume it not being fixed? Obviously there could always be a pity system in the background to make it actually more likely to get drops, but as of now, i've not seen anything that could prove a drop rate below 20 % other than people thinking, that doing something 10-30 times proves anything in this context

And since even with a drop rate of as high as 20 %, the grind still takes dozens, if not hundreds or even thousands of hours, depending on how much of a completionist any given player is. Okkham's razor makes us rather assume, that there's no reason for nexon to manipulate the rates even further, especially given that players might stop just blindly throwing accusations around and actually start to write down their drop rates and i'm starting to get baffled about the ignorance of the many people who find the time to discuss about this topic, but didn't have this very basic idea of how to actually prove it.

[Kyutaru]

Originally posted by Maya-Neko:

Though, in case of the dice, we're not trying to look at a combined value, but rather look at every instance individually. That said, if we try to get at least one 1 rolled, then rolling 1 dive has a 1/6-chance. When we roll 2 dices, then we're already at a 11/36-chance to get at least one 1-roll. We don't search for a 2 or a 12, we explicitly just search for a 1, no matter how many dices we keep rolling.

No, you're misunderstanding the point of referencing 12 or 2. It's because those are the rarest results. Each combination of dice results is a 1/36. Some numbers can be achieved from more than one of these results, like 7 which has the most possible results (which is why Craps is based around rolling 7). 12 and 2 are referenced here because they both only have a single possible result each, and so are true 1/36 odds to roll.

Rolling the same number twice is a measure of joint probability. The same can be applied to failure rolls in an RNG system. I *want* to roll Nothing twice in a row, thereby failing to roll the item I want. The chance of doing that is 80% each time, or 80%*80% for a joint probability attempt. At 5 attempts, it's 0.8^5 and at 10 attempts it's 0.8^10 -- again, this is measuring the chance you do not get any of the item. Getting ANY of the item is the inverse.

If you wanted to roll dice attempting solely to get a 1, much like rolling to get a 20%, you're looking at a 5/6 chance of failing to roll a 1 multiplied by a 5/6 chance of failing to roll a 1 yet again. That's a 25/36 probability, the inverse of which is the 11/36 chance you referenced to get a 1. Which again, is still not 1/6 (6/36). The odds increased because you rolled two dice instead of one.

Originally posted by Maya-Neko:

And that's the problem some people seem to have with probability: They assume, that one run can influence the next roll. They expect a specific pattern and get confused, when the RNG doesn't follow it.

No, the issue is you're not understanding what we're looking for or checking against and then make the assumption that we're wrong about it.

Originally posted by Maya-Neko:

You can tell a person to toss a coin 1000 times and write down the sequence of what they've thrown and you can usually tell, if they've actually threwn them or just written down something they think is random, because when people write down random sequences, they're more likely to change the next result with a significantly higher likelyhood than the 50 % of a coin.

The coin example isn't changing the results or doing anything randomly.

These are all eight possible combinations of flipping coins:

Heads-Heads-Heads
Heads-Heads-Tails
Heads-Tails-Heads
Tails-Heads-Heads
Tails-Tails-Heads
Tails-Heads-Tails
Heads-Tails-Tails
Tails-Tails-Tails

You can literally see there are 7 outcomes containing heads, resulting in 87.5% chance of flipping a Heads in three coin flips.

[Simulacrum111]

Another issue I feel is being missed is not so much that the drop rates are accurate or that people understand how to calculate statistics properly.

To be frank, who cares? Players do not feel rewarded for their time and the system does not appear to behave as they would expect. The actual math can be right in the game but the affect on players is negative.

I'll state it another way. No one would challenge that perfect rolls should fall into the 2% category of drops. However, basic abilities and access to game-play mechanics is not something I feel should be subject to this and this is where my own personal gripe comes from the drop rate.

I would have liked access to ice skills when fighting Pyro but since I could not get the materials to drop for Viessa after grinding some 7 hours straight for ONE of them I had no real choice but to cobble together some CHILL ATK on my weapons and hope for the best. I couldn't do what was obviously optimal because I didn't plan in advance or put in effort to prepare. Access was effectively denied due to the drop rate and it still to this moment.

RNG should not control whether I have access to the same basic gameplay mechanics as other players and if RNG is to play any role at all it should be small. It's akin to waiting for Warlocks or Hunters to drop in Destiny 2. When you say it out loud you have to chuckle.

[Shuyin178]

If you think it only takes 10-20 tries for a 20% chance you have clearly never played Pokemon before.
Focus Blast at 70% accuracy is called "Focus Miss" by the community for a reason, where you'll spend back to back to back turns missing the attack
Attract is 50-50 on if you can attack that turn, to which you'll spend 3-4 turns in a row not attacking most of the time
Confusion is 50-50 on if you attack the target or yourself, to which you'll spend 3-4 turns in a row hitting yourself
Paralysis is a 25% chance for you not to attack, meaning 75% chance you will attack, to which you'll not be able to attack several turns in a row

Trust me, you are not hitting a 20% chance at maximum 20 attempts as a game mechanic, you'll need to get lucky.

[Rogue]

Do you only have elemental school math knowledge? 5 attempts and you get 20% chance?This is not fraction or addition where you add five 20% and you get 100%...

Each attempt is 20% you get it or 80% you don't get it , is distinct and not influenced by previous or subsequent attempt.

Another condition people has is called gambler's fallacy. Basically the more you lose/don't get it, the more likely you get it in the next try. But no, the chance is still the same. In this example, still 20%, not magically becomes 100%.