Mathematical analysis of the new loadout crafting system

(Ardez1) #21

I noticed my contradiction but decided not to change it. Neither system is worse without weapon kits. The new system is better with them.

But when we talk about RNG, we have to talk about expected out of each individual roll. It does not matter at all how many rolls you take. RNG is the potential output of a single roll. If you go beyond that you are getting into statistics, not odds. Neither will be exact, as with 9 attempts you may get duplicates even though there are 9 possible loadout results(excluding skins). So you are absolutely required to look at the one time rolls when determining RNG.

Also, when did the cases get changed? Are you talking about the Classy Case Event which ends soon? That was the increased gold drop. If you are talking about something else I may have missed it >.<.

Let me know.

@ArsonistCow

(Runeforce) #22

Drop chances got altered in the latest update, hence the salt about ‘devaluation.’

http://i.imgur.com/js0uGjH.jpg

(Ardez1) #23

Thanks @Runeforce ! I must have missed that.

(The_N00Ba) #24

I don’t think you were the only one Ardez. lol :slight_smile:

(Xenithos) #25

[quote=“Runeforce;195201”]Drop chances got altered in the latest update, hence the salt about ‘devaluation.’

http://i.imgur.com/js0uGjH.jpg[/quote]

W-WAIT! DID MULTIPLE PEOPLE NOT READ THE UPDATE!?! I just figured you guys added that to your equations! :open_mouth:

(Runeforce) #26

As usual. We’ll rather base our gut oppinion on dumb Youtubers! :slight_smile:

(ArsonistCow) #27

That’s conjecture. As it stands right now it’s worse, but SD has stated that there probably will be more ways of obtaining weapon kits, so we simply don’t know if it will be worse or better in the future.

No, that’s false. If we get more tries at something then we have a higher chance of succeeding.
It’s pretty simple probability math:
If something has a 10% chance of success and we only get one try then that’s obviously a total 10% chance of success. But if we get 10 tries we have a total (1-0.9^10)*100 = 65% chance of success.
More tries does equal less RNG.

(The_N00Ba) #28

I believe Ardez’s point is that “RNG is the potential output of a single roll.” All the chances of getting something during 1 roll. It is not just the chance of succeeding. It is all chances. it is the roller of the die who decides the chance of success, which mean chances of success is relative to the roller. once success is defined by the roller the chance of success can be determined and chance of success can be increased the more rolls you make. The RNG, however, remains the same. Those values are static and predetermined. No matter how many times you roll the the dice the RNG stays the same. The chances stay the same. For example rolling an actual 6 sided die 1000 times will never change the fact that there is only 1 one, 1 two, 1 three, 1 four, 1 five, and 1 six. You can roll the die 1 million times and still the numbers will be the same. Rolling more will increase the chance of your defined success but chances of occurrences provided by the RNG are never affected.

(Ardez1) #29

@ArsonistCow You are talking about statistics, not probability. We have to be clear what we are talking about when we say RNG. When I say RNG I am talking about the probability of one particular result happening. When you say RNG you are talking about taking a collection of results and counting your successes verse your failures, which is statistics.

In my experience RNG, when related to gaming, is only ever about the individual chance. Statistics don’t come into it for me.

So, speaking on probability, the chances are identical when not using weapon kits.

(ArsonistCow) #30

Okay, this is simply a case of differing definitions then. I’m using “RNG” as a shorthand for “randomness” (I believe most people in the community do the same), while you use it to refer to “odds per roll”.
Neither of those fit the actual definition of RNG, for the record. So we could say that we’re both wrong/right, however you wanna look at it.

But you should also consider what information is useful to the people reading this. Do people care about individual odds or do they care about total chance of success? I’m willing to bet my left ear that the chance of success is the more useful information here.

(Ardez1) #31

[quote=“ArsonistCow;195802”]Okay, this is simply a case of differing definitions then. I’m using “RNG” as a shorthand for “randomness” (I believe most people in the community do the same), while you use it to refer to “odds per roll”.
Neither of those fit the actual definition of RNG, for the record. So we could say that we’re both wrong/right, however you wanna look at it.

But you should also consider what information is useful to the people reading this. Do people care about individual odds or do they care about total chance of success? I’m willing to bet my left ear that the chance of success is the more useful information here.[/quote]

hahaha, we are already getting this confused again.

Sure, statistics are useful. But the ‘chance of success’ does NOT increase with more attempts. The ‘chance of success’ does not change. Just because you flip a coin twice doesn’t mean you are going to get heads and tails. Sure, when taking a pool of data you are bound to get a good mix of heads and tails, but the individual flip is always 50/50. The individual roll for crafting is always 11%(if you don’t care about the skin).

And yeah, the definition of RNG is Random Number Generator. Of course we are using it as slang for odds. All we know is that a random number that corresponds to a result will be pulled.

I personally find the individual chance to be the more important number because that is the constant. There is no such thing as ‘total chance of success’ because the ‘chance of success’ doesn’t ever change, even with more attempts. It will be 11% every single time, even if you try 1,000 times with either the old trade up/in system or the new crafting system.

So if you just say 11% chance to get the loadout you want, then try 1,000 times you will theoretically have 110 successes. That doesn’t mean they are guaranteed. You may get zero successes. You may get 300.

The only reason you get more noticeable success with the old system is because they FORCED you to roll more times. You didn’t have a choice. So yeah, if the new system required you to roll 100+ trade ups because even getting the rarity you want then you would probably have more of what you consider ‘success’.

But a success of a rarity you don’t want isn’t actually a ‘success’, otherwise you wouldn’t care about the rarity and just get a bronze.

For all practical purposes this thread is useless because the ‘chance of success’ has not changed in terms of normal crafting.

(ArsonistCow) #32

Wrong. It does. But only if you look at all the attempts as a collective. Each separate attempt only has X% chance, but when tried Y number of times you get (1-(1-X/100)^Y)*100 = Z% total chance of success if only one of the attempts need to be successful for it to count as a success.
Once again: probability math.

(jokingShale) #33

This assumes a person can’t want several different rarities at the same time, which is wrong. Personally when I wanted a specific loadout I was fine if I got it in any rarity higher than bronze, as long as it was one of the skins I liked (usually urban, crown, swiss, or either cobalt, but it might be a bit different depending on the merc in question).

So in the end, the “chance of walking away with a card I want to play”, if you want to call it that, was a LOT higher in the previous system.

(Ardez1) #34

Wrong. It does. But only if you look at all the attempts as a collective. Each separate attempt only has X% chance, but when tried Y number of times you get (1-(1-X/100)^Y)*100 = Z% total chance of success if only one of the attempts need to be successful for it to count as a success.
Once again: probability math.[/quote]

And that formula hasn’t changed at all when they went to the new system. Hence the lack of change in RNG(unless you use weapon kits).

What I am saying is that the ‘chance of success’ hasn’t changed one bit from the old system. The only substantial change is that the system requires fewer overall attempts to get the loadout at the rarity you want(rather then doing all previous tiers first). There is no ‘overall better rng’ with the new system because if you did that probability math for BOTH the new system and the old 30 attempts in either would equal the same odds of success in either. There is no change in the possibilities.

It is very important to note what is considered a ‘success’. In my mind, that will vary from person to person. I usually only want a particular loadout at a particular rarity, without caring about the skin. That makes my odds 1/9 at every rarity on every attempt. Sure, you can apply your probability math to imply at what point I ‘should’ get the loadout I want, but it isn’t a guarantee and more crafting doesn’t necessarily improve my personal ‘chance of success’ because the game doesn’t care what I rolled before. It just isn’t taken into account.

If I only cared about the loadout and not the rarity/skin I would grab a bronze for credits and be done with the whole thing.

My entire point all along is that the math hasn’t changed for the actual randomness of getting what you want. It just lets you skip what you didn’t want before. And of course you can argue “But what if I didn’t care what rarity it was as long as it was over bronze?” Ok. Then craft for lower rarities and you odds haven’t changed there either.

We’re beginning to go in circles though, so I’m going to stop here.

Enjoy.

(jokingShale) #35

The thing is that with the same amount of resources we used to have many rolls to get a card we like, as long as we weren’t incredibly fussy about exactly which rarity and skin it has to be. That’s gone now, there’s only one roll, and that’s a massive difference to the final result.

I don’t think anyone was arguing that a single roll has changed at all, because that much should be obvious to everyone here. So it’s reasonable to assume that everyone who’s complaining about the new system in this vein is talking about the difference between multiple rolls in the old system versus the strictly single roll in the new one.

(ArsonistCow) #36

You really don’t understand the math, do you?
Yep, the formula is exactly the same, but the variables change, hence the result also changes. As Y (number of attempts) increases so does Z (chance of success).

Cmon man, it’s getting more and more obvious that you’re lacking knowledge on probability math. Maybe you should consider the words of somebody that has a decent amount of knowledge on the topic instead of stubbornly sticking to your own incomplete understanding.