Avatar increases the probability of loot 20. Loot generation system. Item Set - Workhorse Drop
This guide describes the game mechanism responsible for generating loot, as well as what affects it and why random always remains random.
Where's my loot, dude?
There are several ways to get loot - killing enemies and searching for chests, and the probability of dropping varies from place to place.
Types of enemies.
Regular.
These are the simplest mobs, they are found throughout the game. There is no prefix in the names. (bandit, marauder, etc.)
Crazy, super crazy and plump. All the same ordinary mobs, only more tenacious and strong, they can be larger, they are found quite often. The corresponding prefix is added to the names. (chubby skag, super crazy goliath, etc.)
Raid bosses and unique mobs.
Incredibly complex and rare Pandora creatures are found in strictly designated places. The names of these mobs, as well as their appearance, are unique. (Hunter Hellquist, Terramorph Invincible, etc.)
To have a sufficient understanding of how a drop works, it is necessary to go through the entire process of generating it, along the way explaining how the game makes its choice. Let's use the example of the murder of an ordinary bandit to see how this mechanism works.
Definition Sheet - Basics
When you kill an enemy, the game looks at its properties, which determines what kind of loot can fall from it. What kind of loot drops is determined by two types of data. The first type is called a "Definition Sheet", which contains a set of item names. The definition sheet provides designers with the easiest way to define what an enemy should drop in the larger sense of the word. Instead of attaching a bunch of similar definitions of certain items to a mob (enemy), of which there can be 5, 10, 20, etc., you can use the definition sheet to immediately assign many of them at once. By customizing this very sheet, you can change the drop of items for a huge number of enemies. This saves developers time. Below is a list of “StandardEnemyGunsAndGear”, which is used mainly by enemies in the “ordinary” category. 
As you noticed, everything that a mob can throw out is displayed here: weapons and equipment, skins for vehicles, eridium, etc. “DropODDS” or numbers after item set definitions indicate the drop chance of a particular item.
When a bandit dies, the game rolls the dice for each of the set of items, determining whether or not the item will drop from that set. For example, the “DropODDS” value for the “weapons and equipment” set is 0.085 in single player. When you kill an enemy, the game selects a random number ranging from 0 to 1. If the resulting number is less than 0.085, then the deceased bandit will throw away your weapon or equipment. If the number is greater than 0.085, then the drop will not occur in this set. The value of 0.085 works out so that for every 11.76 bandits killed, there is 1 weapon drop.
At the same time, drops are calculated for other categories of items. That is, for one kill, the game looks for several such numbers for each set: weapons, equipment, grenades, first aid kits, ammunition, money, etc.
The further into the forest, the more loot.
Have you noticed that the loading screen often shows a hint that the more players on the team, the better your loot. The number of players directly affects the amount of loot you receive when playing Borderlands. In co-op there are some values that change depending on the number of players. Above in the picture you can see an example of this, where it says “AmmoDrops_PerPlayer*0.2” after the entry for the ammunition set.
So how can this affect the behavior of the drop? Believe it or not, there is such a thing as “too much loot”! During the late stages of development, the design team spent most of their workday playing co-op. Usually it was not possible to find this amount of time for this, but in the end it was simply necessary in order to understand how the game would behave in terms of balance in a co-op of 4 people. The most interesting thing that the developers got in the end was the unanimous realization that the game produces an excessive amount of loot. It became tiresome having to collect it, sort it and choose the best from it. (author's note - I said somewhere that this is true. More players = more enemies = more items falling from them = more likely that there will be something worthwhile among them.)
The excess loot was due to the fact that the game used the same drop values, but multiplied by the fact that the number of enemies spawning increased in proportion to the number of players on the team. To solve the current situation, the drop values had to be slightly reduced as the number of players increased. The decrease was still smaller in relation to the increase in the number of mobs. Despite the decrease in the drop chance of each individual enemy, the chance in general increased, since there were more enemies with each player on the team. (author's note - This implies that the number of players in relation to the farming of a particular unique mob is a factor that does not have any influence on this.)
The item set is the workhorse of the drop.
So what is an item set? This is a list of items that the game understands and generates gear from it whenever it is found. A set can refer to others, creating cascading lists until an item or weapon is created.
Let's assume that the definition list "StandardGunsAndGear" selects which item to drop from the first set in the list - Pool_GunsAndGear. It looks like this: 
This pool is located at the top of the chain and links to additional pools of the 5 main item types. Note that it also refers to the "_All" pool for these items, meaning that there is a chance that any item of that type will drop. It is important to know that when selecting from a definition sheet, a check occurs where it is decided whether the item will drop or not. Each pool can only issue 1 item from its list.
How much does it weigh in grams?
The game now produces a different random number to determine which set will be drawn. From the above set you will receive only one item of this type and this probability is shown as a percentage next to the corresponding set. The picture above shows that of all the sets of items, the most likely is a set of weapons, and the least likely is a set of relics. Probability is calculated a little differently. Here it is determined on the basis of the so-called “weight” numbers. Scales are pre-determined numbers by the developers, which again allow you to change the game balance and the odds of hits in one place, rather than changing many numbers in other places. In the example above, the probability value of “common” is equal to 48.13% probability and corresponds to 100 units of weight, and “unusual” is correspondingly 10. Some weight values are then multiplied by given arbitrary values so that the relative probabilities correspond to certain ideas about the relative rarity of objects . The probability values at the top of the picture are distributed as follows:
45 = 100*0.45
Now imagine that the weights and how the game chooses from sets of items are presented in the form of a drawing, a lottery. This is a type of lottery where you have to put a certain number of tickets into a jar. The more tickets you place in a particular jar, the greater your chances of winning that item. Our members are Weapons, Shields, Grenades, Classes and Relics. The weights are the number of tickets from each set of things that our participant will put in the jar to win. That is, whoever has more tickets has more chances - fair. Weapons have 100 tickets, shields hold 45, etc. The game then calls the ticket number and announces the winner. 
Ordinary means ordinary
So let's continue to track the path of the drop from the moment our bandit was killed. The game chose a drop of some item from the Pool_GunsAndGear set, we got a “Guns” ticket in the lottery. Next we find ourselves in the Pool_Weapons_All set, which looks like this:
As you can see, this set reveals another list of sets. In this case, he describes all levels of rarity of items from ordinary to legendary. Unfortunately, the editor is lying to us here, telling us that the degree of rarity for ordinary weapons is zero. This is wrong. The weight of an ordinary set is calculated by a formula that the editor is currently unable to calculate since the game is not running. IN real game the numbers look like this:
Rarity ProbabilityOdds
Regular 89.92% 9 out of 10
Extraordinary8.99% 1 out of 10
Rare0.89% 1 out of 100
Very rare0.09% 1 in 1000
Very rare (E-tech)0.09%1 out of 1000
Legendary0.009%1 out of 10000
So what is the lie of this formula, which writes 0 opposite ordinary objects? The relic of the Vault Seeker is to blame for everything, or rather, that’s how it works. It increases the chance of dropping items of a rank higher than ordinary only due to the fact that it reduces the chance of dropping items of ordinary rank. In an analogy with the lottery, this relic takes away some of the ordinary tickets, so their chance of winning decreases.
Come on, the gun! Give me the gun!
Even though the Vault Hunter Relic "steals" the tickets, let's assume that the Common_All set was the winner. The next step in our process looks like this: 
Now the game chooses what type of weapon you will receive. There are no predefined weights at this stage, just manually entered numbers. As you can see, pistols turn out to be the most common, and rocket launchers, on the contrary, are rare.
Hooray! Gun!
Let's assume that we did get a gun. Now we find ourselves in a set describing the final result. 
In this set we have items that represent each weapon manufacturer, and, as you may have already noticed, for each manufacturer the probability is the same. At this point, the game rolls the dice for the last time to select a specific weapon and we finally get our Tediorov pistol from the bandit.
Game level especially for beginners
Another interesting thing in the information above is something that is in the "Min Game Stage Requirement" field. This field describes a piece of information called the "game stage". This parameter tracks information about the current state of the player and brings the game world to match it, telling other subsystems of the game what enemies, boxes and items should appear.The game level determines why you are now fighting with an enemy of level 5, and not 15, and it also determines why a pistol of level 8, and not 43, fell from our bandit. the game maintains balance in relation to the growth of your level in the game and is an integral part game world, how it functions.
In the case of our example, in the “Min Game Stage Requirement” field in the picture above, “Gamestage_02” is written. This suggests that this set of items works if the game level for our enemy is at least level 2.
Limiting loot
We use the minimum game level field mainly to limit the drop of specific items during playthrough. You may be wondering why we actually limit the drop so much. Wouldn't it be better if all items had a chance to drop at any time? In the matter of receiving gaming experience, especially in the later stages of the game, this expression will sound quite correct, it is acceptable to find all the items. But the early stage of the game is a completely different case. In it, the player just enters the world of Borderlands. When you are literally bombarded with all types of different objects, it can be overwhelming and prevent you from fully understanding each of them. Banning the drop of specific items at one level or another of the game experience gives a chance for others to drop, which allows each new item to be studied at its appointed time before encountering a new item to study.
The second important thing about banning things is that bans are an important aspect of finding something new and interesting early in the game. Imagine that within 10 minutes you fill out all your available slots. It's not such a pleasant moment to find them, because as soon as you find one, you immediately come across the next one. You need time to study a thing and just use it before finding the next one.
Below is a table that displays at what level certain items are introduced into the game. In this case it is weapons. 
As you can see, we do not give drop chances for any guns until you reach the second level. This is to ensure that you don't have any other weapons until you receive the pistol from Claptrap. The next type is automatic rifles, because they perform well in most cases and are the main representative of automatic weapons in a shooter. We then give the player some time for the next missions to introduce shotguns and sniper rifles, which will begin to fall out by that time. Submachine guns (SMGs) are next because they are more utilitarian and we want to give players time to compare them to automatic rifles and make their choice. And finally, at level 10, the last type of weapon is introduced into the game - rocket launchers. They're being introduced now because they're powerful and really cool to find in the game, and their limited ammo makes them a weapon you'll only use on special occasions.
Random Random
I'm sure some of you have experienced times in the game where there were times when loot would just fall out and times where it was missing. In other words, when it was bad and when it wasn’t. At these moments, it seems that the game intentionally somehow distorts the drops, either for the better or for the worse, according to some desire. Let me assure you: this is not true.
Random number generators that are used in programs like Borderlands are technically pseudo-random and do not generate real random due to the existence of various limiting factors such as hardware, operating system etc. There is also a certain limited number of attempts for the gamer, which he cannot overcome in order to get a result that reflects reality.
In Borderlands, all you need to know is that when the game rolls imaginary dice to determine what you should drop from a killed mob or a chest you open, it knows absolutely nothing, neither about your previous attempts, nor about the results that you received in the course of their implementation. Each number has the same chance of being drawn every time an event occurs.
This means that every time any item drops, there is an equal chance for each number to drop. There is no trick with the game reassigning weight for one or another group of items based on how long it has been since the legendary dropped for you. This fact explains the appearance of such moments when you are “lucky”, so to speak, and when you are “unlucky”, supposedly because the game itself changes something.
It is important to remember that the probability is more accurately calculated the more attempts you have. Using our good old bandit to get a legendary from him from the probability table, it becomes clear that our deal is 10,000 to 1. This does not mean that you have to kill him 9,999 times to no avail and only the next time you are guaranteed to get an orange. You may have no luck at all during 19,998 attempts, and then you will immediately get two, one after another. Or it may even happen that you finish off the poor guy 50,000 times without getting a single orange item from him, but then in a short period of time 5 lucky numbers will appear. If such strange cases occur, then your results will somehow be true in a ratio of 10,000 to 1, but you will still feel like the game is cheating behind your back. This is not science fiction, this is how mathematics works.
Fire incident
There are a few small exceptions to the randomization rules just described. They are in the minority and relate to very specific things. Modclasses are always overweighted in relation to game classes, but in the game it is still possible to get a modclass for a character who is not currently in the game. This exception was implemented because it really sucked to play solo and only have a 25% chance of getting a modclass for your character.
You probably noticed such strange entries in the very first picture:
These records refer to ammunition kits in the cases of “need” and “urgent”. The game checks the selected items before dropping the item to see if you need it. The rules are such that the drop is based on knowledge of the game about how much ammo or health you currently have. It adapts to your needs. If you're running low on ammo, you'll likely receive a pack of ammo marked "needed" but not "urgent." If there are no cartridges at all, then the game will throw you two packs at once with one and the other mark. It’s exactly the same with health. If you don't have enough of it, the game will give it to you.
Hello again everyone! As you already understood, I will talk about the mathematical aspect of the esteemed mmorpg. Today my task is to give an idea of how such a very important aspect of the game as item drops is subject to the laws of mathematics. I hope many will learn something new from the article, or simply enjoy reading things they already know about. I'm not going to talk about anything incredibly complicated, so relax and let's get started!
A little about wowhead.com
As the day begins, I would like to say a few words about the well-known site. I think that almost everyone knows how it works, but, nevertheless, I will explain. Players download the app Wowhead Client, it installs an addon that collects information in the game. This way you help the site replenish its database. An excellent scheme, proven over the years. Let's look at a few examples. The Ashes of Al'ar are the dream of any player who is partial to vehicles. If you go to the corresponding page on Wowhead, then in the "mined" section we will see the following:
In the quantity section it is written (at the time of writing) 369 out of 17477. This means that using the application, 17477 boss kills were uploaded to the site, of which the mount dropped 369 times. You can immediately draw conclusions about the number of players helping the site. However, the site calculates the result of dividing 369 by 17477 and gives a rounded drop probability of 2%. We’ll talk about probabilities a little later, now let’s focus on something else. Let's look at another example - Reins of the Invincible. We'll see:
It’s written just above that the mount only drops at 25xm, but that’s not the point. Today we see a drop chance of as much as 5%. This is no longer plausible. It is now August 2012. When I went to Wowhead and looked at the chance of an invincible drop half a year ago, it said 15%. What's the secret? It's quite simple. Before the Cataclysm, the mount dropped with a 100% chance, just like the Purebred Firehawk and the Life-Binder's Handmaiden today with their respective hard mods. A cataclysm occurred, the drop was reduced to 1%, but the information remained in the database. By rough I guess it's possible assume, that before the drop reduction, the mount dropped 163 out of 163 times, and after 35 out of 3512 times, that is, approximately 1 case out of 100.
With this example, I want to show that when using Wowhead data, you should always think about what factors may influence a given situation. Similar patterns can be seen in the cases of Millazor's Smoking Egg, Essence of the Firelord and many others.
So, let's finally figure out what probability is, essentially, and how to calculate how many days you will need to farm this or that item and much more.
Materiel
I'll start with a little copy-paste from Wikipedia.
Probability (probability measure) is a numerical measure of the possibility of the occurrence of some event.
From a practical point of view, the probability of an event is the ratio of the number of those observations in which the event in question occurred to the total number of observations. This interpretation is acceptable in the case of a sufficiently large number of observations or experiments. For example, if about half of the people you meet on the street are women, then you can say that the probability that the person you meet on the street will be a woman is 1/2. In other words, the probability of an event can be assessed by the frequency of its occurrence in a long series of independent repetitions of a random experiment.
The most important thing to understand is that the results of events of this kind, such as meeting a person on the street, rolling a die, or dropping an item in WoW, are pseudo-random, that is, without delving into the philosophy of determinism, we practically cannot make assumptions about the result. relying on some sound judgment. And this means that probability in fact, is not preliminary data that allows one to make assumptions, but, on the contrary, a numerical expression of the result from many experiments. This does not mean that it is not worth taking, the 2% drop chance of Ashes of Al'ar does give an opportunity to guess how often you will see the mount in loot, but still this does not even mean that you will see it exactly one in 50 cd with almost a yearly interval. In general, this whole topic has more to do with philosophy, and mathematics is simply a tool for analyzing real data. So let's apply it. To do this, we will define several terms that are most likely familiar to many in the form of the concepts themselves and their applications , but not names.
Probability multiplication rule
If events A and B are independent, that is, the probability of one event does not depend in any way on the other, then the probability that both will happen is equal to the product of their probabilities.
For example, if the probability of getting heads on a coin is 1/2, and the probability of getting one on a regular die is 1/6, then by tossing a coin and a die at the same time, we get the probability that heads and one will appear 1/2 * 1/6 = 1/ 12. The rule is essentially trivial, but it underlies the calculations in which most people make mistakes.
Negation
If the probability of an event is p, then the probability that it will not occur is 1 - p. An event with a probability of 1 is called reliable, and an event with a probability of 0 is called impossible.
Let's get down to business. Let's assume that the probability of the drop of the same Ashes of Al'ar is 2%, that is, 1/50 or 0.02. Suppose we kill the boss 2 times and want to know what the chances are that the mount will drop. In this case, the drop can be 2 times , then everything is simple, according to the multiplication rule, the chance of dropping 2 times is 1/50 * 1/50 = 1/2500. However, this is not very interesting and vital. It is worth calculating what the chances are that the mount will drop at least 1 time. I want to take it, add it up and, having received 4%, be happy, because this is not so far from the truth. However, if we kill the boss 50 times, it turns out that the drop probability will be 100%. It should already seem like something is wrong. After all, if you add a few more kills, then the drop chance will become more than 1, it just doesn’t happen that way. The fact is that with the addition of probabilities everything is much more complicated than with multiplication. And the situation in which you need to calculate the chance of “either, or” is more difficult than when you need to calculate the chance of “and”.
Let's wrap our problem up a bit. What are the chances that the mount will not fall after a kill? Obviously 0.98. What is the probability that after the next kill he will not fall either? Again 0.98. According to the multiplication rule, the probability that the mount will not fall out in 2 times is equal to the product of the probabilities, that is, 0.9604. This is calculated the same as the chance of dropping 2 mounts for 2 cd. 0.9604 is the probability that the mount will not fall out at all in 2 times, which means 1 - 0.9604 is the probability that it will fall out for 2 cd, that is, it will fall out at least 1 time. 1 - 0.9604 = 0.0396, that is, 3.96%. This is really close to 4% and for good reason. The difference between them is 0.0004, which is exactly the probability that the mount will appear both times. However, this is not the point. We just got a formula for calculating the probability of dropping items after several approaches. If the probability of an event is p, then the probability that it will occur at least 1 time in n attempts is calculated as follows:
For the case of 2 boss kills n=2 and before that we calculated 1 - (1-0.02)*(1-0.02) = 0.3906. Using the calculator, we will find the chance of a phoenix drop in 10 approaches - approximately 0.18.
Let's turn the problem around a little. Let us be given a probability P, and we want to find n, that is, find out how many times we need to kill the boss to get a certain drop probability. Simply rearranging the variables in the formula, we get:
According to the new formula, to get a 50% drop chance, you need to kill the boss about 34 times. It is worth noting that when substituting P=1 into the formula, the logarithm will not be defined. By the way, for a 99% chance you need to kill the boss 228 times.
Unfortunately, many, having learned about the formula, think that if for n approaches the probability was P, then after any number of attempts it remains so. For example, let the probability of an event be 80% in 10 times. Then, according to “logic”, after 9 failures, the chance of success on the 10th attempt will be 80%. It's not like that at all. The probability can be determined in advance, that is, we can say that if the probability of an event is 2%, then the probability in 10 attempts is 18%. However, if we interpret this probability as the probability that the event will occur in remaining attempts, then it will decrease during experiments.
Now let's look at another small aspect. According to the logic of ordinary addition, in 50 times we got a 100% phoenix drop. When calculated, the actual drop will be 63%. If we assume that the chance of dropping Reins of the Deathsteed is 1%, then for 100 kills of Baron Lord Auria Rivender, we will get 63% again. These are not coincidences at all. If the probability p = 1/n, then the formula will take the form
Those who have memories from school will recognize the “wonderful limit” in the last term. In simple terms, the larger n, the closer it will be purely to 1/e, where e is Euler's constant, more details. This number is approximately 2.718281828459045. This means that the larger n, the closer the result is to 1 - 1/e = 0.632102. Since the chance of item drops in WoW is usually quite small, n is large enough that we can say that the chance of dropping items in n times is 63%.
Now let's consider a situation in which you need to get several different items from 1 boss. A great example would be Warblade of Azzinoth and Warblade of Azzinoth. Judging by data from Wowhead, the drop is approximately 5.5%. For a more general example, consider that one has a 5% drop and the other has a 6%. After n kills of Illidan, the drop of blades will be 
To get a chance to collect legendary blades in n trips to the mighty half-elf, half-demon, just multiply the resulting numbers.
For example, after 20 trips you will receive swords with a probability of 36%. This is all true if we consider the blade drops to be independent of each other. However, I have not seen information that there are known cases of both blades dropping at the same time, so if the drop of one blade blocks the other, the task becomes more difficult, although the result practically does not change.
Conclusion
Initially, I wanted to choose a broader topic for the article, talk about characteristics, dimming, etc., but, alas, I won’t talk much about this yet. I decided to make this article not very long, so that an interested reader could really figure it out without scrolling through walls of text about the waiting mat, scary formulas, etc. I hope you liked the article, see you next time!
Introduction
In general, I decided to open a few packages. And at the same time I decided to summarize some statistics.
There are statistics on the Internet from a year ago. But personally, from my own feelings, it flew better before. Legendaries fell more often and stuff like that.
I made a video for those who are interested in watching the opening of the packages. And separately text version for those who are only interested in numbers.
Actually the statistics themselves
Before we begin, let's turn to online sources. We have a wiki that links to a reddit topic.
Standard - 383
Non-standard - 331
Rare - 260
Special - 145
Legendary 42
Chances in percentage:
Standard (33%)
Non-standard (28.5%)
Rare (22.4%)
Special (12.5%)
Legendary (3.6%)
For those in the tank. Legendary - yellow, Special - purple, Rare - blue, Non-standard - white, Standard - gray.
Now let's see my 200 packages.
The packages were specially opened on the day of release new operation so that the data is as fresh as possible.
Number of dropped items:
Standard - 79
Non-standard - 49
Rare - 43
Special - 25 Including thin ice in the amount of 8 items.
Legendary 4
Chances in percentage:
Standard (39.5%)
Non-standard (24.5%)
Rare (21.5%)
Special (12.5%)
Thin Ice - (4%)
Legendary (2%)
As you can see, my statistics are a little sadder than the statistics made a year ago. But as I already said, it feels good items began to fall less frequently. I specifically highlighted thin ice separately, as it is often more desirable than the legendary one.
Well, I propose to combine all the statistics and get the average value.
Number of dropped items:
Standard - 383 + 79
Non-standard - 331 + 49
Rare - 260 + 43
Special - 145 + 25
Legendary 42 + 4
Chances in percentage:
Standard (33.9%)
Non-standard (27.9%)
Rare (22.3%)
Special (12.5%)
Legendary (3.4%)
And a few more interesting things.
I spent 720,000 tokens on packages.
I got 88,300 tokens back from replays
In general, from all this information the following can be distinguished:
The chance of a normal drop from a package is approximately 2 to 4 percent. If you get a legendary, a new lottery starts, since more than half of the skins are complete bullshit. There is almost no point in spending tokens on packages, since the returns are in the form of a heap good skins you probably won't get it.
In general, I fixed up a website here: D
The idea of the site is to create one large library of Rainbow 6 skins. And not just a list, but an overview of each skin, header, shape, icon. It took me a long time to fill the site, it’s quite difficult and monotonous work, but still. I now have over 500 articles with reviews, which are divided into categories, which should make it easier to find the skin you need.
There is still a lot of work left and now the whole difficulty mainly lies in the lack of any skins. You can help me with this.
I created a separate forum thread in the group:
There are all discussions of the site, ideas and other thoughts. Ibid. sample list missing skins.
If possible, all new skins will immediately appear on the site and probably with video reviews. Separate video reviews on the channel:
https://www.youtube.com/user/Bygagator
My other guides
Add this guide to your favorites and don't forget to check out my other guides.
The full list of my guides can be viewed on or through my Steam profile.
You can also subscribe to my channel on YouTube.
https://www.youtube.com/Bygagator
Or join my group on steam:
There I publish links to my guides and answer questions.
If you have anything to add - write
Found errors or inaccuracies - Write
If you don’t agree with my opinion - Write
YOU CAN LEAVE YOUR WISHES AND IDEAS FOR THE FOLLOWING GUIDES IN MY GROUP AT THE LINK:
DID YOU LIKE MY GUIDE? ADD TO YOUR FAVORITES, LIKE!
And a few more words.
Some people believe that the guide is copy-paste from some sites. I have almost all the information written by hand, not counting the description of the character and some small details. Please take this into account when accusing me of any plagiarism or copying. ;)
