LI experiments #2: Food lucky blocks

[Roxrock]

Food lucky blocks drop a number of different items that can be consumed in some way to restore hunger bars and give saturation points. Randomization is used to determine the outcome of opening food lucky blocks. Different drops are believed to have different drop rates based on a system of weights. However, the exact values of these weights is not known by the general public and is probably kept secret by the developers. My goal was to use experimentation and sampling to speculate what these weights might be and what they might be effected by. By sampling the outcomes of food lucky blocks, we can create a probability distribution for the various distinct outcomes.
The following data sample was taken from December 24 to December 28, 2023. All trials were conducted in lucky islands solos on bedrock addition in North American servers. Voting options were not held constant. A total of 473 drops were recorded.

Spoiler: Data table

Drop	% prob	number of occurrences
5 bread	10.78224101	51
12 carrots	10.35940803	49
2 raw mutton	9.090909091	43
12 cookies	7.399577167	35
2 cooked salmon	7.399577167	35
2 cooked beef	7.399577167	35
6 chorus fruit	7.188160677	34
5 apples	6.553911205	31
5 baked potatoes	5.919661734	28
20 glow berries	5.708245243	27
5 golden apples	4.651162791	22
3 honey bottles	4.22832981	20
Cake	3.594080338	17
5 rotten flesh	2.114164905	10
Pumpkin pie	1.902748414	9
8 dried kelp	1.691331924	8
20 cooked chicken	1.057082452	5
10 raw rabbit	0.8456659619	4
Pufferfish	0.8456659619	4
3 spider eyes	0.422832981	2
5 golden carrots	0.422832981	2
Poison potato	0.2114164905	1
Enchanted golden apple	0.2114164905	1

According to this site, these are the hunger bars and saturation points each time a single item is consumed.

Spoiler: Data table

Drop	Hunger bars per item	Saturation points per item
5 bread	2.5	6
12 carrots	1.5	3.6
2 raw mutton	1	1.2
12 cookies	1	0.4
2 cooked salmon	3	9.6
2 cooked beef	4	12.8
6 chorus fruit	2	2.4
5 apples	2	2.4
5 baked potatoes	2.5	6
20 glow berries	2	0.4
5 golden apples	4	9.6
3 honey bottles	3	1.2
Cake	1	0.4
5 rotten flesh	2	0.8
Pumpkin pie	4	4.8
8 dried kelp	0.5	0.6
20 cooked chicken	3	7.2
10 raw rabbit	1.5	1.8
Pufferfish	0.5	0.2
3 spider eyes	1	3.2
5 golden carrots	3	14.4
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Here are the total hunger bars and saturation points for each drop (multiplied by quantity of items, or number of slices)

Spoiler: Data table

Drop	total bars	total sat
5 bread	12.5	30
12 carrots	18	43.2
2 raw mutton	2	2.4
12 cookies	12	4.8
2 cooked salmon	6	19.2
2 cooked beef	8	25.6
6 chorus fruit	12	14.4
5 apples	10	12
5 baked potatoes	12.5	30
20 glow berries	40	8
5 golden apples	20	48
3 honey bottles	9	3.6
Cake	7	2.8
5 rotten flesh	10	4
Pumpkin pie	4	4.8
8 dried kelp	4	4.8
20 cooked chicken	60	144
10 raw rabbit	15	18
Pufferfish	0.5	0.2
3 spider eyes	3	9.6
5 golden carrots	15	72
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Finally, here are the effect values for each drop. A value of zero means the item does nothing special when consumed. A value of 1 means it does something not very significn’t, and a value of 2 means it does something incredibly important.

Spoiler: Data table

Drop	Effect
5 bread	0
12 carrots	0
2 raw mutton	0
12 cookies	0
2 cooked salmon	0
2 cooked beef	0
6 chorus fruit	1
5 apples	0
5 baked potatoes	0
20 glow berries	0
5 golden apples	0
3 honey bottles	1
Cake	0
5 rotten flesh	2
Pumpkin pie	0
8 dried kelp	1
20 cooked chicken	0
10 raw rabbit	0
Pufferfish	2
3 spider eyes	2
5 golden carrots	0
Poison potato	2
Enchanted golden apple	2

Spoiler: Pie Chart

Spoiler: 2 quantitative variable plots

The two next graphs below only include drops with an effect value of 0

Spoiler: Statistics for hunger bars per item

Spoiler: Statistics for Saturation points per item

Spoiler: Statistics for Quantity of items

Spoiler: Statistics for Total hunger bars for drop

Spoiler: Statistics for Total saturation points for drop

Spoiler: Statistics for Effect value of drop

Spoiler: Statistics for % drop rate

Conclusions: There is very strong evidence supporting a weighted lucky block probability distribution. There is a weak negative correlation between hunger points or saturation points per item and quantity of items. Both of these correlations are stronger when drops with an effect value of 0 are isolated. There was a meaningful negative correlation between % drop rate and effect value. The distributions of total hunger bars and total saturation points per drop were both skewed right.

I hope you guys enjoyed learning more details about the loot table for food lucky blocks, and I hope the statistics aren’t too overwhelming. If you would like to give any recommendations or feedback on future experiments and research, please leave a reply below. This honestly drove me insane trying to complete this, so I’m hesitant to attempt more ambitious research projects in the future, but we shall see.

 

[Desiderata]

I'm at a loss of words, but still, great research project/analysis regardless

I remember struggling at the subject when I was your age lmaooo

lowkey looking forward to more posts like these in the future

 

[Laura]

Food lucky blocks drop a number of different items that can be consumed in some way to restore hunger bars and give saturation points. Randomization is used to determine the outcome of opening food lucky blocks. Different drops are believed to have different drop rates based on a system of weights. However, the exact values of these weights is not known by the general public and is probably kept secret by the developers. My goal was to use experimentation and sampling to speculate what these weights might be and what they might be effected by. By sampling the outcomes of food lucky blocks, we can create a probability distribution for the various distinct outcomes.
The following data sample was taken from December 24 to December 28, 2023. All trials were conducted in lucky islands solos on bedrock addition in North American servers. Voting options were not held constant. A total of 473 drops were recorded.

Spoiler: Data table

Drop	% prob	number of occurrences
5 bread	10.78224101	51
12 carrots	10.35940803	49
2 raw mutton	9.090909091	43
12 cookies	7.399577167	35
2 cooked salmon	7.399577167	35
2 cooked beef	7.399577167	35
6 chorus fruit	7.188160677	34
5 apples	6.553911205	31
5 baked potatoes	5.919661734	28
20 glow berries	5.708245243	27
5 golden apples	4.651162791	22
3 honey bottles	4.22832981	20
Cake	3.594080338	17
5 rotten flesh	2.114164905	10
Pumpkin pie	1.902748414	9
8 dried kelp	1.691331924	8
20 cooked chicken	1.057082452	5
10 raw rabbit	0.8456659619	4
Pufferfish	0.8456659619	4
3 spider eyes	0.422832981	2
5 golden carrots	0.422832981	2
Poison potato	0.2114164905	1
Enchanted golden apple	0.2114164905	1

According to this site, these are the hunger bars and saturation points each time a single item is consumed.

Spoiler: Data table

Drop	Hunger bars per item	Saturation points per item
5 bread	2.5	6
12 carrots	1.5	3.6
2 raw mutton	1	1.2
12 cookies	1	0.4
2 cooked salmon	3	9.6
2 cooked beef	4	12.8
6 chorus fruit	2	2.4
5 apples	2	2.4
5 baked potatoes	2.5	6
20 glow berries	2	0.4
5 golden apples	4	9.6
3 honey bottles	3	1.2
Cake	1	0.4
5 rotten flesh	2	0.8
Pumpkin pie	4	4.8
8 dried kelp	0.5	0.6
20 cooked chicken	3	7.2
10 raw rabbit	1.5	1.8
Pufferfish	0.5	0.2
3 spider eyes	1	3.2
5 golden carrots	3	14.4
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Here are the total hunger bars and saturation points for each drop (multiplied by quantity of items, or number of slices)

Spoiler: Data table

Drop	total bars	total sat
5 bread	12.5	30
12 carrots	18	43.2
2 raw mutton	2	2.4
12 cookies	12	4.8
2 cooked salmon	6	19.2
2 cooked beef	8	25.6
6 chorus fruit	12	14.4
5 apples	10	12
5 baked potatoes	12.5	30
20 glow berries	40	8
5 golden apples	20	48
3 honey bottles	9	3.6
Cake	7	2.8
5 rotten flesh	10	4
Pumpkin pie	4	4.8
8 dried kelp	4	4.8
20 cooked chicken	60	144
10 raw rabbit	15	18
Pufferfish	0.5	0.2
3 spider eyes	3	9.6
5 golden carrots	15	72
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Finally, here are the effect values for each drop. A value of zero means the item does nothing special when consumed. A value of 1 means it does something not very significn’t, and a value of 2 means it does something incredibly important.

Spoiler: Data table

Drop	Effect
5 bread	0
12 carrots	0
2 raw mutton	0
12 cookies	0
2 cooked salmon	0
2 cooked beef	0
6 chorus fruit	1
5 apples	0
5 baked potatoes	0
20 glow berries	0
5 golden apples	0
3 honey bottles	1
Cake	0
5 rotten flesh	2
Pumpkin pie	0
8 dried kelp	1
20 cooked chicken	0
10 raw rabbit	0
Pufferfish	2
3 spider eyes	2
5 golden carrots	0
Poison potato	2
Enchanted golden apple	2

Spoiler: Pie Chart

View attachment 226103

Spoiler: 2 quantitative variable plots

The two next graphs below only include drops with an effect value of 0
View attachment 226109View attachment 226110View attachment 226111View attachment 226112View attachment 226113View attachment 226114View attachment 226115View attachment 226116View attachment 226117View attachment 226118View attachment 226119

Spoiler: Statistics for hunger bars per item

View attachment 226121View attachment 226122View attachment 226123

Spoiler: Statistics for Saturation points per item

View attachment 226145View attachment 226146View attachment 226147

Spoiler: Statistics for Quantity of items

View attachment 226148View attachment 226149View attachment 226150

Spoiler: Statistics for Total hunger bars for drop

View attachment 226151View attachment 226152View attachment 226153

Spoiler: Statistics for Total saturation points for drop

View attachment 226154View attachment 226155View attachment 226156

Spoiler: Statistics for Effect value of drop

View attachment 226157View attachment 226158

Spoiler: Statistics for % drop rate

View attachment 226120

Conclusions: There is very strong evidence supporting a weighted lucky block probability distribution. There is a weak negative correlation between hunger points or saturation points per item and quantity of items. Both of these correlations are stronger when drops with an effect value of 0 are isolated. There was a meaningful negative correlation between % drop rate and effect value. The distributions of total hunger bars and total saturation points per drop were both skewed right.

I hope you guys enjoyed learning more details about the loot table for food lucky blocks, and I hope the statistics aren’t too overwhelming. If you would like to give any recommendations or feedback on future experiments and research, please leave a reply below. This honestly drove me insane trying to complete this, so I’m hesitant to attempt more ambitious research projects in the future, but we shall see.

My guy. You can just ask for the drop rates you know?
I respect the R&D, but I can literally just give you the weights if you want. XD
It wouldn't break any NDAs or anything like that, it's obvious that some items have a higher weight than others, that's no secret.
But if you're enjoying doing this then by all means, carry on, it's an interesting read.

 

[Casualpoalrbear]

I don't play lucky blocks so while I don't understand any of this I need to give a clap of hands for this.

 

[Mr Jii Gamer]

Food lucky blocks drop a number of different items that can be consumed in some way to restore hunger bars and give saturation points. Randomization is used to determine the outcome of opening food lucky blocks. Different drops are believed to have different drop rates based on a system of weights. However, the exact values of these weights is not known by the general public and is probably kept secret by the developers. My goal was to use experimentation and sampling to speculate what these weights might be and what they might be effected by. By sampling the outcomes of food lucky blocks, we can create a probability distribution for the various distinct outcomes.
The following data sample was taken from December 24 to December 28, 2023. All trials were conducted in lucky islands solos on bedrock addition in North American servers. Voting options were not held constant. A total of 473 drops were recorded.

Spoiler: Data table

Drop	% prob	number of occurrences
5 bread	10.78224101	51
12 carrots	10.35940803	49
2 raw mutton	9.090909091	43
12 cookies	7.399577167	35
2 cooked salmon	7.399577167	35
2 cooked beef	7.399577167	35
6 chorus fruit	7.188160677	34
5 apples	6.553911205	31
5 baked potatoes	5.919661734	28
20 glow berries	5.708245243	27
5 golden apples	4.651162791	22
3 honey bottles	4.22832981	20
Cake	3.594080338	17
5 rotten flesh	2.114164905	10
Pumpkin pie	1.902748414	9
8 dried kelp	1.691331924	8
20 cooked chicken	1.057082452	5
10 raw rabbit	0.8456659619	4
Pufferfish	0.8456659619	4
3 spider eyes	0.422832981	2
5 golden carrots	0.422832981	2
Poison potato	0.2114164905	1
Enchanted golden apple	0.2114164905	1

According to this site, these are the hunger bars and saturation points each time a single item is consumed.

Spoiler: Data table

Drop	Hunger bars per item	Saturation points per item
5 bread	2.5	6
12 carrots	1.5	3.6
2 raw mutton	1	1.2
12 cookies	1	0.4
2 cooked salmon	3	9.6
2 cooked beef	4	12.8
6 chorus fruit	2	2.4
5 apples	2	2.4
5 baked potatoes	2.5	6
20 glow berries	2	0.4
5 golden apples	4	9.6
3 honey bottles	3	1.2
Cake	1	0.4
5 rotten flesh	2	0.8
Pumpkin pie	4	4.8
8 dried kelp	0.5	0.6
20 cooked chicken	3	7.2
10 raw rabbit	1.5	1.8
Pufferfish	0.5	0.2
3 spider eyes	1	3.2
5 golden carrots	3	14.4
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Here are the total hunger bars and saturation points for each drop (multiplied by quantity of items, or number of slices)

Spoiler: Data table

Drop	total bars	total sat
5 bread	12.5	30
12 carrots	18	43.2
2 raw mutton	2	2.4
12 cookies	12	4.8
2 cooked salmon	6	19.2
2 cooked beef	8	25.6
6 chorus fruit	12	14.4
5 apples	10	12
5 baked potatoes	12.5	30
20 glow berries	40	8
5 golden apples	20	48
3 honey bottles	9	3.6
Cake	7	2.8
5 rotten flesh	10	4
Pumpkin pie	4	4.8
8 dried kelp	4	4.8
20 cooked chicken	60	144
10 raw rabbit	15	18
Pufferfish	0.5	0.2
3 spider eyes	3	9.6
5 golden carrots	15	72
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Finally, here are the effect values for each drop. A value of zero means the item does nothing special when consumed. A value of 1 means it does something not very significn’t, and a value of 2 means it does something incredibly important.

Spoiler: Data table

Drop	Effect
5 bread	0
12 carrots	0
2 raw mutton	0
12 cookies	0
2 cooked salmon	0
2 cooked beef	0
6 chorus fruit	1
5 apples	0
5 baked potatoes	0
20 glow berries	0
5 golden apples	0
3 honey bottles	1
Cake	0
5 rotten flesh	2
Pumpkin pie	0
8 dried kelp	1
20 cooked chicken	0
10 raw rabbit	0
Pufferfish	2
3 spider eyes	2
5 golden carrots	0
Poison potato	2
Enchanted golden apple	2

Spoiler: Pie Chart

View attachment 226103

Spoiler: 2 quantitative variable plots

The two next graphs below only include drops with an effect value of 0
View attachment 226109View attachment 226110View attachment 226111View attachment 226112View attachment 226113View attachment 226114View attachment 226115View attachment 226116View attachment 226117View attachment 226118View attachment 226119

Spoiler: Statistics for hunger bars per item

View attachment 226121View attachment 226122View attachment 226123

Spoiler: Statistics for Saturation points per item

View attachment 226145View attachment 226146View attachment 226147

Spoiler: Statistics for Quantity of items

View attachment 226148View attachment 226149View attachment 226150

Spoiler: Statistics for Total hunger bars for drop

View attachment 226151View attachment 226152View attachment 226153

Spoiler: Statistics for Total saturation points for drop

View attachment 226154View attachment 226155View attachment 226156

Spoiler: Statistics for Effect value of drop

View attachment 226157View attachment 226158

Spoiler: Statistics for % drop rate

View attachment 226120

Conclusions: There is very strong evidence supporting a weighted lucky block probability distribution. There is a weak negative correlation between hunger points or saturation points per item and quantity of items. Both of these correlations are stronger when drops with an effect value of 0 are isolated. There was a meaningful negative correlation between % drop rate and effect value. The distributions of total hunger bars and total saturation points per drop were both skewed right.

I hope you guys enjoyed learning more details about the loot table for food lucky blocks, and I hope the statistics aren’t too overwhelming. If you would like to give any recommendations or feedback on future experiments and research, please leave a reply below. This honestly drove me insane trying to complete this, so I’m hesitant to attempt more ambitious research projects in the future, but we shall see.

Can you give a short summarize of your speech in order to understand the topic step by step and thank you

 

[Roxrock]

My guy. You can just ask for the drop rates you know?
I respect the R&D, but I can literally just give you the weights if you want. XD
It wouldn't break any NDAs or anything like that, it's obvious that some items have a higher weight than others, that's no secret.
But if you're enjoying doing this then by all means, carry on, it's an interesting read.

PLEASE give me the drop rates if you can. Before doing this I asked around in the programming discord channel, but was mostly ignored. Doing so would help me so much with documenting the game and creating constructing feedback about it. To be completely honest, I have not enjoyed this process in the slightest. I’ve basically been opening a lucky block, screenshotting its drop, jumping into the void, and repeating, and creating those graphs and plots wasn’t easy either (Google sheets and statsmedic both kept crashing)

 

[Roxrock]

Can you give a short summarize of your speech in order to understand the topic step by step and thank you

Most importan thing to look at is the first data table. It shows the probability of each food lucky block drop, based off of the sample trials I conducted.

 

[Mr Jii Gamer]

Most importan thing to look at is the first data table. It shows the probability of each food lucky block drop, based off of the sample trials I conducted.

Wow this work should take from you a long time good job man

 

[Laura]

PLEASE give me the drop rates if you can. Before doing this I asked around in the programming discord channel, but was mostly ignored. Doing so would help me so much with documenting the game and creating constructing feedback about it. To be completely honest, I have not enjoyed this process in the slightest. I’ve basically been opening a lucky block, screenshotting its drop, jumping into the void, and repeating, and creating those graphs and plots wasn’t easy either (Google sheets and statsmedic both kept crashing)

Oki, I'll see what info I can put together, what I can share, and I'll make a mini Behind the Cube style post here depending on what info I can provide.
The mechanics aren't actually as simple as just picking something from a table of chances, there's more that goes into the selection, but your data is interesting to see how the system translates into actual results.

Honestly I thought you were doing this for some sort of school project or something. XD

 

[MajorMythG]

Food lucky blocks drop a number of different items that can be consumed in some way to restore hunger bars and give saturation points. Randomization is used to determine the outcome of opening food lucky blocks. Different drops are believed to have different drop rates based on a system of weights. However, the exact values of these weights is not known by the general public and is probably kept secret by the developers. My goal was to use experimentation and sampling to speculate what these weights might be and what they might be effected by. By sampling the outcomes of food lucky blocks, we can create a probability distribution for the various distinct outcomes.
The following data sample was taken from December 24 to December 28, 2023. All trials were conducted in lucky islands solos on bedrock addition in North American servers. Voting options were not held constant. A total of 473 drops were recorded.

Spoiler: Data table

Drop	% prob	number of occurrences
5 bread	10.78224101	51
12 carrots	10.35940803	49
2 raw mutton	9.090909091	43
12 cookies	7.399577167	35
2 cooked salmon	7.399577167	35
2 cooked beef	7.399577167	35
6 chorus fruit	7.188160677	34
5 apples	6.553911205	31
5 baked potatoes	5.919661734	28
20 glow berries	5.708245243	27
5 golden apples	4.651162791	22
3 honey bottles	4.22832981	20
Cake	3.594080338	17
5 rotten flesh	2.114164905	10
Pumpkin pie	1.902748414	9
8 dried kelp	1.691331924	8
20 cooked chicken	1.057082452	5
10 raw rabbit	0.8456659619	4
Pufferfish	0.8456659619	4
3 spider eyes	0.422832981	2
5 golden carrots	0.422832981	2
Poison potato	0.2114164905	1
Enchanted golden apple	0.2114164905	1

According to this site, these are the hunger bars and saturation points each time a single item is consumed.

Spoiler: Data table

Drop	Hunger bars per item	Saturation points per item
5 bread	2.5	6
12 carrots	1.5	3.6
2 raw mutton	1	1.2
12 cookies	1	0.4
2 cooked salmon	3	9.6
2 cooked beef	4	12.8
6 chorus fruit	2	2.4
5 apples	2	2.4
5 baked potatoes	2.5	6
20 glow berries	2	0.4
5 golden apples	4	9.6
3 honey bottles	3	1.2
Cake	1	0.4
5 rotten flesh	2	0.8
Pumpkin pie	4	4.8
8 dried kelp	0.5	0.6
20 cooked chicken	3	7.2
10 raw rabbit	1.5	1.8
Pufferfish	0.5	0.2
3 spider eyes	1	3.2
5 golden carrots	3	14.4
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Here are the total hunger bars and saturation points for each drop (multiplied by quantity of items, or number of slices)

Spoiler: Data table

Drop	total bars	total sat
5 bread	12.5	30
12 carrots	18	43.2
2 raw mutton	2	2.4
12 cookies	12	4.8
2 cooked salmon	6	19.2
2 cooked beef	8	25.6
6 chorus fruit	12	14.4
5 apples	10	12
5 baked potatoes	12.5	30
20 glow berries	40	8
5 golden apples	20	48
3 honey bottles	9	3.6
Cake	7	2.8
5 rotten flesh	10	4
Pumpkin pie	4	4.8
8 dried kelp	4	4.8
20 cooked chicken	60	144
10 raw rabbit	15	18
Pufferfish	0.5	0.2
3 spider eyes	3	9.6
5 golden carrots	15	72
Poison potato	1	1.2
Enchanted golden apple	2	9.6

Finally, here are the effect values for each drop. A value of zero means the item does nothing special when consumed. A value of 1 means it does something not very significn’t, and a value of 2 means it does something incredibly important.

Spoiler: Data table

Drop	Effect
5 bread	0
12 carrots	0
2 raw mutton	0
12 cookies	0
2 cooked salmon	0
2 cooked beef	0
6 chorus fruit	1
5 apples	0
5 baked potatoes	0
20 glow berries	0
5 golden apples	0
3 honey bottles	1
Cake	0
5 rotten flesh	2
Pumpkin pie	0
8 dried kelp	1
20 cooked chicken	0
10 raw rabbit	0
Pufferfish	2
3 spider eyes	2
5 golden carrots	0
Poison potato	2
Enchanted golden apple	2

Spoiler: Pie Chart

View attachment 226103

Spoiler: 2 quantitative variable plots

The two next graphs below only include drops with an effect value of 0
View attachment 226109View attachment 226110View attachment 226111View attachment 226112View attachment 226113View attachment 226114View attachment 226115View attachment 226116View attachment 226117View attachment 226118View attachment 226119

Spoiler: Statistics for hunger bars per item

View attachment 226121View attachment 226122View attachment 226123

Spoiler: Statistics for Saturation points per item

View attachment 226145View attachment 226146View attachment 226147

Spoiler: Statistics for Quantity of items

View attachment 226148View attachment 226149View attachment 226150

Spoiler: Statistics for Total hunger bars for drop

View attachment 226151View attachment 226152View attachment 226153

Spoiler: Statistics for Total saturation points for drop

View attachment 226154View attachment 226155View attachment 226156

Spoiler: Statistics for Effect value of drop

View attachment 226157View attachment 226158

Spoiler: Statistics for % drop rate

View attachment 226120

Conclusions: There is very strong evidence supporting a weighted lucky block probability distribution. There is a weak negative correlation between hunger points or saturation points per item and quantity of items. Both of these correlations are stronger when drops with an effect value of 0 are isolated. There was a meaningful negative correlation between % drop rate and effect value. The distributions of total hunger bars and total saturation points per drop were both skewed right.

I hope you guys enjoyed learning more details about the loot table for food lucky blocks, and I hope the statistics aren’t too overwhelming. If you would like to give any recommendations or feedback on future experiments and research, please leave a reply below. This honestly drove me insane trying to complete this, so I’m hesitant to attempt more ambitious research projects in the future, but we shall see.

Respect. This is very interesting.

 

[Dualninja]

This is a HUGE data set, it's about 500 games worth of Food Lucky Blocks. Have you been using Replay Mod or something to track this? Or have you been manually writing down your drops from every game in real time? It's impressive either way; wait is Replay Mod even available on Bedrock?

 

[Roxrock]

This is a HUGE data set, it's about 500 games worth of Food Lucky Blocks. Have you been using Replay Mod or something to track this? Or have you been manually writing down your drops from every game in real time? It's impressive either way; wait is Replay Mod even available on Bedrock?

I’ve actually been opening the lucky block, picking up the drop, screenshotting, then jumping in the void. Then I go through my screenshots and count how many times each type of drop appears, deleting the screenshots I count. I do this for every different type of drop until I go through them all, so then I have a column with the description of each drop, and a column for the number of times each drop appears. The rest of the work consisted of adding columns for the hunger bars, effect value, saturation, etc., then throwing that data into statsmedic applets to create those plots and graphs (this was the part that drove me insane)
I was fortunate enough to a response when asking for help in the google forums, which ended up making that last bit so much easier.
Here is the spreadsheet I used, it’s turned into a complete and utter mess by the end of it.

 

[Laura]

Ok. Here is the list of the possible loots and their weights. However, as I said before, their weights aren't the only thing taken into consideration when picking a loot for a lucky block. I'll explain more on this below.
Loot Table

So... The different blocks have different loot tables they pull from.
Rainbow Blocks have the loot listed under the Rainbow section in the link above.
Food Blocks have all the loot listed under the Food section in the link above.
Blocks Blocks () have all the loot listed under the Blocks section in the link above.
Glitched Blocks have just the loot under the Glitched section.
Normal Blocks have nearly all the loot in the regular section.
Diamond Blocks have most of the very positive loot in the regular section, and sets all of their weights to 1.
(Note that some of this loot is Bedrock only, and some is Java only. Didn't have time to get that info.)

All loot is classified as either Good, Passive, or Dangerous. You can probably figure out which is which, depending on if they make the game easier for someone, harder, or don't really affect it much.

When a block is broken, its possible loot is divided into 3 lists, one list of Good loot, one of Passive, and one of Dangerous loot.
Each list is then sampled to get one loot from each pool, making 3 in total. One good, one bad, one ok.
This is where their weights are used. The higher their weight, the more likely they are to be selected amongst others of their danger category.
Sometimes, some blocks have no dangerous loot at all, so they only have one good and one ok, or just one good.

After the one of each loot has been selected, it picks one of them at random, favouring the positive one if things like Extra Lucky or Blessed are active.
Note: Some loot in there, like the normal Blocks event, have their own randomness when deciding which blocks to drop, once they've been selected by the loot picker. This one in particular picks a random number of either Stone, Sandstone, or Dirt blocks to drop if not OP, and if it's OP then it drops both Stone and Obsidian. There's stuff like this for most loot too.

As you can probably see, this makes calculating the exact likeliness of each possible loot outcome immensely difficult, as you have compounding chances based on mode, block, extra effects, and specific loot logic.

If you want any more info, I can probably explain a few more things.
Hope this helps clear things up!

 

[Tripleee]

Ok. Here is the list of the possible loots and their weights. However, as I said before, their weights aren't the only thing taken into consideration when picking a loot for a lucky block. I'll explain more on this below.
Loot Table

So... The different blocks have different loot tables they pull from.
Rainbow Blocks have the loot listed under the Rainbow section in the link above.
Food Blocks have all the loot listed under the Food section in the link above.
Blocks Blocks () have all the loot listed under the Blocks section in the link above.
Glitched Blocks have just the loot under the Glitched section.
Normal Blocks have nearly all the loot in the regular section.
Diamond Blocks have most of the very positive loot in the regular section, and sets all of their weights to 1.
(Note that some of this loot is Bedrock only, and some is Java only. Didn't have time to get that info.)

All loot is classified as either Good, Passive, or Dangerous. You can probably figure out which is which, depending on if they make the game easier for someone, harder, or don't really affect it much.

When a block is broken, its possible loot is divided into 3 lists, one list of Good loot, one of Passive, and one of Dangerous loot.
Each list is then sampled to get one loot from each pool, making 3 in total. One good, one bad, one ok.
This is where their weights are used. The higher their weight, the more likely they are to be selected amongst others of their danger category.
Sometimes, some blocks have no dangerous loot at all, so they only have one good and one ok, or just one good.

After the one of each loot has been selected, it picks one of them at random, favouring the positive one if things like Extra Lucky or Blessed are active.
Note: Some loot in there, like the normal Blocks event, have their own randomness when deciding which blocks to drop, once they've been selected by the loot picker. This one in particular picks a random number of either Stone, Sandstone, or Dirt blocks to drop if not OP, and if it's OP then it drops both Stone and Obsidian. There's stuff like this for most loot too.

As you can probably see, this makes calculating the exact likeliness of each possible loot outcome immensely difficult, as you have compounding chances based on mode, block, extra effects, and specific loot logic.

If you want any more info, I can probably explain a few more things.
Hope this helps clear things up!

Are glitched lucky blocks Java exclusive (never heard of one before)

 

[Roxrock]

Are glitched lucky blocks Java exclusive (never heard of one before)

No, they come out of normal lucky blocks rarely.

 

[Roxrock]

Ok. Here is the list of the possible loots and their weights. However, as I said before, their weights aren't the only thing taken into consideration when picking a loot for a lucky block. I'll explain more on this below.
Loot Table

So... The different blocks have different loot tables they pull from.
Rainbow Blocks have the loot listed under the Rainbow section in the link above.
Food Blocks have all the loot listed under the Food section in the link above.
Blocks Blocks () have all the loot listed under the Blocks section in the link above.
Glitched Blocks have just the loot under the Glitched section.
Normal Blocks have nearly all the loot in the regular section.
Diamond Blocks have most of the very positive loot in the regular section, and sets all of their weights to 1.
(Note that some of this loot is Bedrock only, and some is Java only. Didn't have time to get that info.)

All loot is classified as either Good, Passive, or Dangerous. You can probably figure out which is which, depending on if they make the game easier for someone, harder, or don't really affect it much.

When a block is broken, its possible loot is divided into 3 lists, one list of Good loot, one of Passive, and one of Dangerous loot.
Each list is then sampled to get one loot from each pool, making 3 in total. One good, one bad, one ok.
This is where their weights are used. The higher their weight, the more likely they are to be selected amongst others of their danger category.
Sometimes, some blocks have no dangerous loot at all, so they only have one good and one ok, or just one good.

After the one of each loot has been selected, it picks one of them at random, favouring the positive one if things like Extra Lucky or Blessed are active.
Note: Some loot in there, like the normal Blocks event, have their own randomness when deciding which blocks to drop, once they've been selected by the loot picker. This one in particular picks a random number of either Stone, Sandstone, or Dirt blocks to drop if not OP, and if it's OP then it drops both Stone and Obsidian. There's stuff like this for most loot too.

As you can probably see, this makes calculating the exact likeliness of each possible loot outcome immensely difficult, as you have compounding chances based on mode, block, extra effects, and specific loot logic.

If you want any more info, I can probably explain a few more things.
Hope this helps clear things up!

Ok, so based on the way you described the weight system, I tried my best to assign each drop as either good, bad, or neutral, and calculated each drop’s probability.
Modified loot table with probabilities
For now, I’ve left it at normal mode but shouldn’t be too tricky to calculate it for the other modes
However, in order for me to create an accurate probability distribution, I would really need to know for sure which drops are good, bad and neutral. If you want, you could make a copy of this spreadsheet and correct anywhere that I misinterpreted if a drop is good, passive, or dangerous, and the probability should automatically update(this could be a bit laggy)

I also left some notes about why I didn’t add certain drops (most of the drops I don’t know are probably Java only), drops I thought were missing from the sheet, and a couple of other small things.

I don’t think I have enough information to create a distribution for diamond lucky blocks, since they do contain some bad drops, like raw mutton, but don’t contain some good drops, like sharpness stone swords, triedents, etc. It shouldn’t be a huge pain for me to figure out its list of drops for bedrock, if every drop has equal probability, but I would still appreciate it if you want to do that for me anyway.

Ive heard from some people that crazy mode modifies the drop weights of certain drops. Is that true?

Here are a few other questions I have about how lucky islands works:

• How do events in crazy mode work (I know this is probably its own can of worms, the most important thing I would like to know is just how long each event lasts)
• What are the ranges for the number of arrows players receive for the various special bows
• What different mobs can and cannot come out of a mob wand
• when a player receives blocks from a lucky block, what is the range of the number of blocks they can receive
• For the “Food” drop from normal lucky blocks, what are the different possible outcomes
• What different drops can players receive in chests

I could probably go on and on about questions I have about the game and understand if you don’t want to just keep answering them. I already am super grateful the fact you created this spreadsheet and explained how lucky block drops work. If you want, I could try to reorganize the spreadsheet so that new drops can be added if you are interested in using it to calculate probabilities when adding new drops to the game.

Again, thank you so much for providing this for me, I really appreciate it, I can finally start putting together some coherent and up to date loot tables, and you have saved me untold hours of experimentation.

 

[Desiderata]

saved me untold hours of experimentation

damn, i genuinely wanted to see more of lucky islands experiments

 

[Roxrock]

damn, i genuinely wanted to see more of lucky islands experiments

I might make a posts reflecting on these experiments to see how accurate they were to finding the actual probabilities, but yeah, most of what I would end up calculating would have to do with how different things effect lucky block outcomes. I Might make 1 last experiment posts about maps. The silver lining is that now, I can continue working on other projects that use information from this spreadsheet, and I might be able to finally make progress towards completing my guide. Nonetheless, I’m glad you enjoyed this short series of experiments.

 

[Laura]

damn, i genuinely wanted to see more of lucky islands experiments

I didn't have it in my heart to continue his suffering.
Man's shaved literal years off of his life doing this.
God knows what his death stat is at after joining a game, breaking a block, then jumping in the void for all those games... I feel like resetting it out of sheer good will...

But I'm not going to.

 

[Laura]

Ok, so based on the way you described the weight system, I tried my best to assign each drop as either good, bad, or neutral, and calculated each drop’s probability.
Modified loot table with probabilities
For now, I’ve left it at normal mode but shouldn’t be too tricky to calculate it for the other modes
However, in order for me to create an accurate probability distribution, I would really need to know for sure which drops are good, bad and neutral. If you want, you could make a copy of this spreadsheet and correct anywhere that I misinterpreted if a drop is good, passive, or dangerous, and the probability should automatically update(this could be a bit laggy)

I also left some notes about why I didn’t add certain drops (most of the drops I don’t know are probably Java only), drops I thought were missing from the sheet, and a couple of other small things.

I don’t think I have enough information to create a distribution for diamond lucky blocks, since they do contain some bad drops, like raw mutton, but don’t contain some good drops, like sharpness stone swords, triedents, etc. It shouldn’t be a huge pain for me to figure out its list of drops for bedrock, if every drop has equal probability, but I would still appreciate it if you want to do that for me anyway.

Ive heard from some people that crazy mode modifies the drop weights of certain drops. Is that true?

Here are a few other questions I have about how lucky islands works:

• How do events in crazy mode work (I know this is probably its own can of worms, the most important thing I would like to know is just how long each event lasts)
• What are the ranges for the number of arrows players receive for the various special bows
• What different mobs can and cannot come out of a mob wand
• when a player receives blocks from a lucky block, what is the range of the number of blocks they can receive
• For the “Food” drop from normal lucky blocks, what are the different possible outcomes
• What different drops can players receive in chests

I could probably go on and on about questions I have about the game and understand if you don’t want to just keep answering them. I already am super grateful the fact you created this spreadsheet and explained how lucky block drops work. If you want, I could try to reorganize the spreadsheet so that new drops can be added if you are interested in using it to calculate probabilities when adding new drops to the game.

Again, thank you so much for providing this for me, I really appreciate it, I can finally start putting together some coherent and up to date loot tables, and you have saved me untold hours of experimentation.

I will see what I can do. I'm not sure I'm going to be able to give you the specifics on which loot results are in which block/category, I think that might be revealing too much of the game. I'll see though.

I'll have a look into the other questions too.