Plentiful Harvest Passive does not equal 50%... TESTED

Rittings

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

VaranisArano

This matches pretty well with all the data we're seeing on the other thread: https://forums.elderscrollsonline.com/en/discussion/564705/plentiful-harvest-cp-star-underperforming#latest

My own data from Pure Water nodes which only drop Lorkhan's Tears is out of 272 nodes:
97 doubled with 4-6 Lorkhans Tears (36%)
175 undoubled with 2-3 Lorkhans Tears (64%)
If this were a coin toss, no one would hesitate to say something was messed up with those results.

It's pretty disappointing. If I'm going to be spending - what is it, 50 points? - to get a 50% bonus, I'd like to actually get 50% of my items doubled. If they aren't going to fix it, the cost should be lowered.

Rittings

Ahh I never saw that thread... I even searched for "Plentiful Harvest" but got no hits... so thanks for sharing that too.

Yeah, I think this is something where the numbers are incorrect, especially seeing how this is the same results as force lock...

SirAndy

People have been reporting issues with ZOS's *weighted* rng all the way back since early beta.

They just can't seem to get their code right ...

Ackwalan

It feels as if ZOS uses hidden calculations. Cooldowns or anti-farming rules that affect everything in game from harvesting to loot drops.

Scardan

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

tmbrinks

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

That's not how it's worded. Nor is it how it works. You do indeed get double drops (more truthfully, double "pulls"), not at 50% increase.

It's a 50% chance of getting a double drop. Which means you would expect the nodes to double 50% of the time.

Rittings

Scardan wrote: »

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

Theory of Probability works as thus:

The probability of the outcome based on the amount of possible outcomes.

---

So, let's say there is are 2 possible outcomes:- double yield, or single yield, and the chance of either is even (because it's at 50% chance of double yield) then theory of probability would state that there is a 1 in 2 chance either way - meaning an even chance of probability.

It's a coin flip, in effect. I want you to find a coin now... flip it twice and the likelihood of 2 heads is as great as 2 tails. The likelihood of either of those over one of each is also the same. However, we find that over the course of many flips (and the larger the pool of data the closer to even we become) is that since there is a 1 in 2 chance, you would end up with an even split of both (or very close to it).

Thus, my 1,000 node pulls with a 50% chance of double yield only coming back with 31.3% is quite unlikely. It's even more unlikely when so many are reporting the same findings... thus, it can be concluded that the 50% information given is likely to be incorrect.

tmbrinks

Rittings wrote: »

Scardan wrote: »

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

Theory of Probability works as thus:

The probability of the outcome based on the amount of possible outcomes.

---

So, let's say there is are 2 possible outcomes:- double yield, or single yield, and the chance of either is even (because it's at 50% chance of double yield) then theory of probability would state that there is a 1 in 2 chance either way - meaning an even chance of probability.

It's a coin flip, in effect. I want you to find a coin now... flip it twice and the likelihood of 2 heads is as great as 2 tails. The likelihood of either of those over one of each is also the same. However, we find that over the course of many flips (and the larger the pool of data the closer to even we become) is that since there is a 1 in 2 chance, you would end up with an even split of both (or very close to it).

Thus, my 1,000 node pulls with a 50% chance of double yield only coming back with 31.3% is quite unlikely. It's even more unlikely when so many are reporting the same findings... thus, it can be concluded that the 50% information given is likely to be incorrect.

The chances are 0 for all intents and purposes. Meaning we can definitively say that it is not 50%

ThorianB

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

You are right. It is 50% per roll. If you do two rolls, you don't have a 100% chance of getting a double node. You have a 50% chance twice. But the average over time should be ~ 50% especially when there is only double or not double as an option.

If we have 10 players pick 1000 nodes and compare their results, they should mostly be in the range of 40-60%. However, everyone is reporting well below 50%. Some should be close to 50% and some should be over 50%.

The same is true with the furnishing print perk. I average a double print 3-5% of the time and others have similar results. Most players should be getting a double print 7%-13% of the time with some outliers but again many people are well below the stated amount. This would indicate either a lot of people have really bad luck that pay attention to what they are getting as loot or that the RNG is not true to the description.

tmbrinks

ThorianB wrote: »

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

You are right. It is 50% per roll. If you do two rolls, you don't have a 100% chance of getting a double node. You have a 50% chance twice. But the average over time should be ~ 50% especially when there is only double or not double as an option.

If we have 10 players pick 1000 nodes and compare their results, they should mostly be in the range of 40-60%. However, everyone is reporting well below 50%. Some should be close to 50% and some should be over 50%.

The same is true with the furnishing print perk. I average a double print 3-5% of the time and others have similar results. Most players should be getting a double print 7%-13% of the time with some outliers but again many people are well below the stated amount. This would indicate either a lot of people have really bad luck that pay attention to what they are getting as loot or that the RNG is not true to the description.

Agree, but the range is actually even smaller than that.

80% of the people should be between 48% and 52% with 1000 nodes and a 50% chance. and 94% between 47% and 53%

Once you get to larger sample sizes (especially for things with around 50/50 chances) the range of acceptable values get very narrow quite quickly.

(What I mean by acceptable are results that don't make us think that it's not a 50% chance).

The 99.7% confidence range (3 standard deviations in either direction) would be between 45% and 55%. So seeing numbers in the 30% range is definitely "not acceptable"

Rittings

Okay... 170 surveys completed = 1,020 nodes

Double Pulls: 301 .... 29.5%
Single Pulls: 719

It's the worse than my previous pool... but my previous pool did atleast include non-survey nodes.

Here is something extra that seems quite damning too...

If you toss a coin 6 times, the chance of it being heads 6 times is 1/64 (which is pretty much what we should be seeing with the survey nodes... only a 1/64 chance of them all being single or double).

However, 16 of my surveys were ALL single nodes... and ZERO were all double. In fact, I'm yet to see a full double survey and I've done a ton of them. I mean... from 1/64 chance, my findings are closer to 1/10 chance of pulling a full single node survey... and 1/Infinity of a full double.

I think we need some answers on this one.

BergisMacBride

Running my numbers on surveys and regular farming for a week, but something looks off, particularly for BS, Cl, WW and JC. Alchemy is closer but still only 42% doubles at the moment. Most others are running at around 33% chance of doubles. Still better than before but atm nowhere near 50%.

redspecter23

This matches up with some of the data I've been collecting. I did a small sample size of 300 nodes last week and saw about 1/3rd double drops as many others are stating here. I think this is enough info for ZOS to at least look at the code on their end. Players don't like feeling cheated and this is looking more and more like a 30% chance and not a 50% chance.

Similarly to what is mentioned in the post above, while I was doing my surveys (not included in my count here), I noticed zero of them were 6/6 doubled while a very large amount were 6/6 single. This was over approximately 350 surveys.

Scardan

tmbrinks wrote: »

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

That's not how it's worded. Nor is it how it works. You do indeed get double drops (more truthfully, double "pulls"), not at 50% increase.

It's a 50% chance of getting a double drop. Which means you would expect the nodes to double 50% of the time.

No, I wouldn't. It is 50% chance which does not mean I get double drop every first or second try. Probability. When you flip a coin, it has a 50 percent chance to lye with digit up. How often does this happen?

EDIT: Plentiful Harvest is worded in a way that shows: 50% chance when farming a node, which I understand that with a 50% chance of getting double resources, you will get double resources with a frequency of "from and to" times.

VaranisArano

Scardan wrote: »

tmbrinks wrote: »

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

That's not how it's worded. Nor is it how it works. You do indeed get double drops (more truthfully, double "pulls"), not at 50% increase.

It's a 50% chance of getting a double drop. Which means you would expect the nodes to double 50% of the time.

No, I wouldn't. It is 50% CHANCE which does not mean I get double drop every first or second try. Probability. When you flip a coin, it has a 50 percent chance to lye with digit up. How often does this happen?

As was explained, Plentiful Harvest works by doing a second pull of the RNG for a node. This is why nodes have the range that they do.

Consider the "Pure Water" nodes, which at maxed Alchemy will only contain Lorkhan's Tears. A single RNG pull of the node will yield 2 to 3 Tears. A double RNG pull will yield 4 to 6 tears. i.e. any 4 to 6 tears node is a doubled node. (2+2, 2+3, 3+2, 3+3 are all the result of the double RNG pull).

The more we flip your coin, we should see our actual results approach 50/50.

As I pointed out, my actual results from Pure Water nodes which only drop Lorkhan's Tears is out of 272 nodes:
97 doubled with 4-6 Lorkhans Tears (36%)
175 undoubled with 2-3 Lorkhans Tears (64%)

If this were a coin toss, no one would hesitate to say something was messed up with those results.

Scardan

VaranisArano wrote: »

Scardan wrote: »

tmbrinks wrote: »

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

That's not how it's worded. Nor is it how it works. You do indeed get double drops (more truthfully, double "pulls"), not at 50% increase.

It's a 50% chance of getting a double drop. Which means you would expect the nodes to double 50% of the time.

No, I wouldn't. It is 50% CHANCE which does not mean I get double drop every first or second try. Probability. When you flip a coin, it has a 50 percent chance to lye with digit up. How often does this happen?

As was explained, Plentiful Harvest works by doing a second pull of the RNG for a node. This is why nodes have the range that they do.

Consider the "Pure Water" nodes, which at maxed Alchemy will only contain Lorkhan's Tears. A single RNG pull of the node will yield 2 to 3 Tears. A double RNG pull will yield 4 to 6 tears. i.e. any 4 to 6 tears node is a doubled node. (2+2, 2+3, 3+2, 3+3 are all the result of the double RNG pull).

The more we flip your coin, we should see our actual results approach 50/50.

As I pointed out, my actual results from Pure Water nodes which only drop Lorkhan's Tears is out of 272 nodes:
97 doubled with 4-6 Lorkhans Tears (36%)
175 undoubled with 2-3 Lorkhans Tears (64%)

If this were a coin toss, no one would hesitate to say something was messed up with those results.

No. First 10 times I flipped a coin - 8 times with number up. Second try - 4 times with number up. Now toss a coin. You get different results then me. It will not approach 50/50.

What you found out there is the frequency of double drops, which has a 50% chance of giving a double drop. And if you continue the experiment, you will find that your percentages have changed again. You have not figured out the likelihood of the event itself, you are observing its consequences.

VaranisArano

Scardan wrote: »

VaranisArano wrote: »

Scardan wrote: »

tmbrinks wrote: »

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

That's not how it's worded. Nor is it how it works. You do indeed get double drops (more truthfully, double "pulls"), not at 50% increase.

It's a 50% chance of getting a double drop. Which means you would expect the nodes to double 50% of the time.

No, I wouldn't. It is 50% CHANCE which does not mean I get double drop every first or second try. Probability. When you flip a coin, it has a 50 percent chance to lye with digit up. How often does this happen?

As was explained, Plentiful Harvest works by doing a second pull of the RNG for a node. This is why nodes have the range that they do.

Consider the "Pure Water" nodes, which at maxed Alchemy will only contain Lorkhan's Tears. A single RNG pull of the node will yield 2 to 3 Tears. A double RNG pull will yield 4 to 6 tears. i.e. any 4 to 6 tears node is a doubled node. (2+2, 2+3, 3+2, 3+3 are all the result of the double RNG pull).

The more we flip your coin, we should see our actual results approach 50/50.

As I pointed out, my actual results from Pure Water nodes which only drop Lorkhan's Tears is out of 272 nodes:
97 doubled with 4-6 Lorkhans Tears (36%)
175 undoubled with 2-3 Lorkhans Tears (64%)

If this were a coin toss, no one would hesitate to say something was messed up with those results.

No. First 10 times I flipped a coin - 8 times with number up. Second try - 4 times with number up. Now toss a coin. You get different results then me. It will not approach 50/50.

What you found out there is the frequency of double drops, which has a 50% chance of giving a double drop. And if you continue the experiment, you will find that your percentages have changed again. You have not figured out the likelihood of the event itself, you are observing its consequences.

Plentiful Harvest says it should be doing a double RNG pull 50% of the time, with the double RNG pull demonstrated by the observance of nodes with double RNG pull results: 4 to 6 Lorkhans' Tears.

Roughly 50% of nodes should have the doubled RNG pull, give or take some. In my data set, only 36% did.

If you want to counter that, I would like to see your stats on your Lorkhan's Tears RNG.

After all, when we average your stats and my stats, the results should approach 50%...assuming ZOS is correct.

Probability deals in large numbers. And the fact that you think flipping a coin 20 times, out of which you got an average of the coin landing number up 60% of the time is a good example compared to the sample sizes we're looking at says a lot about your argument. The more you flipped the coin, the closer you got to 50%. If you continue flipping that coin, and continue adding to the data set, you should get closer to 50%

Here's my challenge: go track your own farming results. Pure Water nodes are the easiest to check because it's obvious when you get the double RNG pull. You think we're all missing something and that Plentiful Harvest is in fact proccing 50% of the time? Okay. Prove it.

Feric51

Scardan wrote: »

VaranisArano wrote: »

Scardan wrote: »

tmbrinks wrote: »

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

That's not how it's worded. Nor is it how it works. You do indeed get double drops (more truthfully, double "pulls"), not at 50% increase.

It's a 50% chance of getting a double drop. Which means you would expect the nodes to double 50% of the time.

No, I wouldn't. It is 50% CHANCE which does not mean I get double drop every first or second try. Probability. When you flip a coin, it has a 50 percent chance to lye with digit up. How often does this happen?

As was explained, Plentiful Harvest works by doing a second pull of the RNG for a node. This is why nodes have the range that they do.

Consider the "Pure Water" nodes, which at maxed Alchemy will only contain Lorkhan's Tears. A single RNG pull of the node will yield 2 to 3 Tears. A double RNG pull will yield 4 to 6 tears. i.e. any 4 to 6 tears node is a doubled node. (2+2, 2+3, 3+2, 3+3 are all the result of the double RNG pull).

The more we flip your coin, we should see our actual results approach 50/50.

As I pointed out, my actual results from Pure Water nodes which only drop Lorkhan's Tears is out of 272 nodes:
97 doubled with 4-6 Lorkhans Tears (36%)
175 undoubled with 2-3 Lorkhans Tears (64%)

If this were a coin toss, no one would hesitate to say something was messed up with those results.

No. First 10 times I flipped a coin - 8 times with number up. Second try - 4 times with number up. Now toss a coin. You get different results then me. It will not approach 50/50.

What you found out there is the frequency of double drops, which has a 50% chance of giving a double drop. And if you continue the experiment, you will find that your percentages have changed again. You have not figured out the likelihood of the event itself, you are observing its consequences.

Without going into a long diatribe about your understanding of statistics and probability (I’ll leave that to @tmbrinks). Your example, despite not having nearly enough tests to be statistically relevant, shows the exact reason why so many are complaining.

Your first test showed an 80% success rate, your second test showed a 40% based on a 50/50 chance. Yet, NOBODY who has been tracking their pulls have had anything above 50% which should realistically happen to at least some portion of the population if the odds were truly 50% for each pull. Everybody who has tracked their drops have been below, and often well below, the 50% threshold.

tmbrinks

Feric51 wrote: »

Scardan wrote: »

VaranisArano wrote: »

Scardan wrote: »

tmbrinks wrote: »

Scardan wrote: »

Rittings wrote: »

Some may remember me keeping logs on lockpicks and forcing locks - and finding that the percentage that the tooltip reads, does NOT reflect the actual outcome...

Bad RNG?? Or just simply bad coding? I'll let you decide...

Here's the original post on the force lock issue - for reference...

https://forums.elderscrollsonline.com/en/discussion/390876/force-lock-percentage-is-incorrect-not-working-as-intended#latest

Now, onto the new CP Plentiful Harvest "boost"... well, it's a boost, but it's not what we have spent our CP on that we are getting. Which is why I wish this to be moved to a simple 100% chance, that way we KNOW what we are getting... even if the CP cost is higher.

After so many of my guildies, and random people in zone are stating similar things, I decided to open up my notepad and start recording the raw data once again. It started with noticing I would go on a streak of 8-10 single harvest nodes, then get get 1 or 2 doubles... but never having a true double node streak like the single ones I was having. And at 50% I thought it should have evened itself out... it didn't.

The last 1,000 nodes I have farmed (not special nodes, not even runes since they can be a bit weird... but alchemy/woodwork/jewelry/blacksmith/clothing ones)... I have had 687 single harvests and 313 doubles. I do have the 50% plentiful harvest maxed out, and have it on my active bar (if I didn't I wouldn't have gotten any double nodes). My crafting is maxed - and I have everything set up perfectly.

So... 31.3% is better than we had, but again, it's not what the CP tooltip states.. not even close. I'll monitor the next 1,000 too - but from what others are telling me, this is about what we are going to get. Which is disappointing to say the least.

Probability does not work this way if I remember probability theory correctly. The chance of getting 50% boost (half of normal drop btw, double drop would be 100% more) is rolled on every action. It shouldn't be even.

That's not how it's worded. Nor is it how it works. You do indeed get double drops (more truthfully, double "pulls"), not at 50% increase.

It's a 50% chance of getting a double drop. Which means you would expect the nodes to double 50% of the time.

No, I wouldn't. It is 50% CHANCE which does not mean I get double drop every first or second try. Probability. When you flip a coin, it has a 50 percent chance to lye with digit up. How often does this happen?

As was explained, Plentiful Harvest works by doing a second pull of the RNG for a node. This is why nodes have the range that they do.

Consider the "Pure Water" nodes, which at maxed Alchemy will only contain Lorkhan's Tears. A single RNG pull of the node will yield 2 to 3 Tears. A double RNG pull will yield 4 to 6 tears. i.e. any 4 to 6 tears node is a doubled node. (2+2, 2+3, 3+2, 3+3 are all the result of the double RNG pull).

The more we flip your coin, we should see our actual results approach 50/50.

As I pointed out, my actual results from Pure Water nodes which only drop Lorkhan's Tears is out of 272 nodes:
97 doubled with 4-6 Lorkhans Tears (36%)
175 undoubled with 2-3 Lorkhans Tears (64%)

If this were a coin toss, no one would hesitate to say something was messed up with those results.

No. First 10 times I flipped a coin - 8 times with number up. Second try - 4 times with number up. Now toss a coin. You get different results then me. It will not approach 50/50.

What you found out there is the frequency of double drops, which has a 50% chance of giving a double drop. And if you continue the experiment, you will find that your percentages have changed again. You have not figured out the likelihood of the event itself, you are observing its consequences.

Without going into a long diatribe about your understanding of statistics and probability (I’ll leave that to @tmbrinks). Your example, despite not having nearly enough tests to be statistically relevant, shows the exact reason why so many are complaining.

Your first test showed an 80% success rate, your second test showed a 40% based on a 50/50 chance. Yet, NOBODY who has been tracking their pulls have had anything above 50% which should realistically happen to at least some portion of the population if the odds were truly 50% for each pull. Everybody who has tracked their drops have been below, and often well below, the 50% threshold.

I just got done teaching math all week... don't really want to do it anymore on my Friday night.

Suffice to say, you two explained it well... the other person is incorrect in their logic/interpretation.

I stated it all earlier. With a sample of 1000, the 95% confidence interval is doubling rates between 47% and 53%, and we have that level of data. Getting well outside of that proves the doubling rate is not 50%

Q.E.D.

Now I'm going to play the game.

BergisMacBride

Though I would add my latest results to this thread for posterity's sake. Completed a week of running daily surveys and harvesting nodes in between, documenting single and double pulls on all types regular\survey nodes with the exception of enchanting nodes.

Results by Node Type:

Alch - 377 nodes - 230(68%) single, 147(32%) double
BS - 118 nodes - 74(63%) single, 44(37%) double
CW - 153 nodes - 94(61%) single, 59(39%) double
WW - 149 nodes - 100(67%) single, 49(33%) double
JC - 107 nodes - 66(62%) single, 41(38%) double

Total 904 nodes, with 564 (62.4%) single pulls and 340 (37.6%) double pulls. This number included 65 surveys. The only other thing I tracked on the surveys was the number of 6:0 ratio pulls. There were four 6:0 single pull surveys and no 6:0 double pull surveys. I'd also note during my data collection, the patterns and ratios across all node types were pretty much set the first day and held pretty constant throughout, account for minor RNG variations.

I've seen enough to know the tooltip is not correct, either through a bug or intent. Granted, we are much better off than before with the redesigned perk by a factor of 3-4, but it is clear the actual number is no where near the 50% that's been stated by the devs and in the tooltip.

@ZOS_GinaBruno, can we get any response on this from the devs? We've provided pretty good testing and evidence here that something is amiss with this perk or its tooltip.

beadabow

For the first time since I read this post, I had a streak of double drops that was 11 (included two surveys: 1 of 5 double drops in a row immediately followed by a survey of 6 double drops). I know this is statistically an outlier, but wanted to state that a 6x double drop (every node in the survey) can happen. Prolly super rare though.

Edit to add type of survey: Clothier

kringled_1

I've also had one 6 double node survey and a few 5s. My anecdotal observations, though, are in line with the original post. Most surveys are at 1-3 doubled nodes, with 0 not being super rare. 4 isn't thatuncommon either but 5 and especially 6 are. This is from a few hundred surveys since FoA went live.

Blacknight841

You all are missing the point here. If you are getting around 30% of nodes with bonus mats, and the algorithm is set to 50%, then that means someone eventually would get 70% of nodes with bonus materials. Since that is not the case, it is clearly not 50%.

VaranisArano

Blacknight841 wrote: »

You all are missing the point here. If you are getting around 30% of nodes with bonus mats, and the algorithm is set to 50%, then that means someone eventually would get 70% of nodes with bonus materials. Since that is not the case, it is clearly not 50%.

Well, yes. The more data we gather, our results should be approaching 50% we'd expect to see some under and some over, averaging out at close to 50% the more data we add.

On both threads, that's just not happening. We're all coming in under 50%, in the 30-40% range. That much has already been said here.

When i did my data on different nodes, I would get similar streaks of doubled nodes like some people observed in their surveys. So it's not that streaks can't happen. Rather, when we add all the results together, those streaks don't shift the numbers toward 50% as much as it might seem when I'm in the middle of a nice streak.

lillybit

Has anyone tested this again since this weeks patches?

I've only done a few surveys but getting 2-4 doubles with each which is what I'd expect for 50% - it's too small a sample size to say anything for sure but hoping it's fixed!

redspecter23

lillybit wrote: »

Has anyone tested this again since this weeks patches?

I've only done a few surveys but getting 2-4 doubles with each which is what I'd expect for 50% - it's too small a sample size to say anything for sure but hoping it's fixed!

Be careful what you wish for. When they finally "fix" the issue, which will require them acknowledging it first, they will probably just adjust the tooltip on the CP to read 30% instead of 50% and call it a day. They may even try to say that it was always intended to be 30% and the tooltip was just a typo.

Obsidian3

I'm running at 52% right now, but I am only at 173 nodes. Will update by the end of the weekend.

I've always had great RNG when it comes to plentiful harvest, back in 2017 I was running at 34% (Non-Surveys)

I usually average about 4-5 Atherial Dust a year

Now Perfect Roe and Nirncrux I cannot get to save my life, last time I did a Craglorn survey run, I did 33 surveys (Cloth, Wood, Metal, Jewelry combined) I got 1 potent and 2 fortified. I've been told from guildies that's beyond bad for Crux Survey drops.
<Shrug>

SirAndy

redspecter23 wrote: »

lillybit wrote: »

Has anyone tested this again since this weeks patches?
I've only done a few surveys but getting 2-4 doubles with each which is what I'd expect for 50% - it's too small a sample size to say anything for sure but hoping it's fixed!

Be careful what you wish for. When they finally "fix" the issue, which will require them acknowledging it first, they will probably just adjust the tooltip on the CP to read 30% instead of 50% and call it a day. They may even try to say that it was always intended to be 30% and the tooltip was just a typo.

It's only funny because it's true. ZOS has done this before.

adirondack

Lies, damn lies, and statistics.


Joking aside - all great info and hard work here folks. Please keep this rolling.

RedMuse

beadabow wrote: »

For the first time since I read this post, I had a streak of double drops that was 11 (included two surveys: 1 of 5 double drops in a row immediately followed by a survey of 6 double drops). I know this is statistically an outlier, but wanted to state that a 6x double drop (every node in the survey) can happen. Prolly super rare though.

Edit to add type of survey: Clothier

I've had streaks too. My longest was 15. Problem with that is that it is such a tiny sample size and every time I've had a steak of 5-7 I've tried to keep going for a while if I at all have the time to do it and it always evens out to somewhere 30%-34%.
Just like flipping a coin. Yes you can get 10 or 15 heads in a row but continuing to go you should, assuming a non-weighted coin, average out near 50%.

No one I think have argued that streaks didn't and doesn't happen. They do. It's the average we're arguing about. And have anyone with a decent sample size reported anything above 50%? If so, I have missed it. I had one single run of 77 nodes which ran to 37.6%, that's the highest I've been able to get. And I haven't read of anyone getting much higher than that, if at all. Most seem to land at about 33%.