Real Legendary drop rates and other statistics for 9785 chars [Spreadsheet]

[lehadnk]

Btw. I think the amount of criteria 33096 (mythic 2+ completed in time) includes all higher levels too, so no need to add 5+ and 10+ numbers into it.

Confirmed that it's not. I've been doing some 10+ runs yesterday and have the same 168 2+ runs in my 33096 criteria: http://www.wowprogress.com/character...нмеркар

I would say the number 1 cause of wipes for me is healers running low on mana, unable to continue healing at full potential.

Wait. What is mana?

[Chemii]

Literally have no idea why you wouldn't make them all equal drop rate, never mind the soft cap debacle. Maybe its just inflated because of shared class loot though. Even so why do any of those exist.

What a joke.

[Deleted]

800 mythic+ dungeons completed? Character and realm name?

It's all in the signature ...

[impending doom]

The problem with this logic is that it assumes that the % of people who have a legendary goes up linearly with total amount of KP. In other words, you're assuming that if the average KP is 100 at 25%, then it's 200 at 50% and 400 at 75%. This is only true if you're equally likely to get a legendary at each KP level, which probably isn't true (note that this isn't true if you just roll a % on each KP).

How this effects the estimate depends on the exact system that Blizzard is using. For the bad luck system, it probably overestimates the numbers at high and low percentages, and underestimates around the middle. Kinda depends on the exact numbers though.

I did a little more analysis, and there's definitely something weird going on with the data. I graphed the amount of legendaries against KP for both the 2nd and 3rd legendaries, and both actually had the relative number of legendaries go down with more KP, which doesn't make any sense. Could be something weird on Blizzard's end, could be bad data, could be something in between.

(Poorly formatted, sketchy) analysis can be found at google docs /spreadsheets/d/1cGGnkokih6TYC-fP7WT7pPnGkD0glnEZX77irOC6KAg/edit?usp=sharing

Yes as I mentioned in my last post, the BLP effect makes the calculations more complex, and my formula doesn't take that into account yet. However also the BLP table I made suggests that there's a significant jump in KP needed after each additional legendary.

First to clarify, the G column (KP 1to2) in my data contains KP from everyone who got at least 1 legendary. It starts counting KP after 1st legendary is received, and stops when 2nd legendary is looted OR until the data ends.

I'm not sure what kind of graph you were looking for, but in your graph "% second leg vs. KP" you compare those who got the legendary within the first 10 KP, only to people who have collected max 10 KP, but that excludes everyone who collected more than 10 KP so I don't think this % can give meaningful info.

But if you edit the formulas on that page to something like this, it should give a "drop chance per KP" curve, similar to the BLP stats I made before:

=countIFS(ALL!$G:$G,CONCATENATE("<=",$A12),ALL!$G:$G,CONCATENATE(">",$A2),ALL!$C:$C,"0",AL L!$D:$D,">=2")/20
=countIFS(ALL!$G:$G,CONCATENATE(">=",$A12),ALL!$C:$C,"0",ALL!$D:$D,"1")

Ahh.. you are trying to calculate the underlying chance of a legendary drop. As opposed to a statistical analysis.

If that is the case that you can't hand wave your KP values. You need to show much more data as to why they are also reflective of the underlying drop rates. And since your data on specific drop rates don't match your KP values (Mythic boss is 6 KP with a 2.7% drop rate for 1+ leg and Norm boss is 3 KP with a 0.94%), I have to say I still think your conclusion is a misrepresentation of the data. I mean you admit yourself, in section 3, that the drop rates for M+ chests could be off by as much as 50%. If that is the case, add your data is based heavily on M+ drops as you said here:

Then all your data could vary by as much as 50%? And you still are claiming a 2x factor in effort to get the next legendary? Don't you mean anywhere from 50% more to 100% more?

Yes the goal is to find out how much effort is needed get the next legendary. Calculating just the statistical mean of the winners doesn't really help those who haven't won the legendary yet, as it doesn't represent the true probability of a drop.

That 2.7% vs 0.94% you took are drop rates just from 1 week, I took my KP weights from several weeks of data to get better sample size. But yeah they're all just estimations, especially the M+, and that's why I said in the beginning that "The data below gives only a rough estimation and there might be still errors too", maybe I should emphasize that better on next update.

Also your data on player activity needs to be much more accurate. For example how are there players that received a legendary after 1 KP worth of effort if you only award KP at a min 2 value? And why is it that so many of your bottom players have ~200 KP without 1 legendary? Also the post says 9785 characters, but you data table only has 6748 players worth of KP values, none with less than 200 KP. Where are the 3000+ players with less than 200 KP? And finally how accurate were you in starting the KP count after a player got a legendary? These things become really important if you are trying to calculate the underlying chance.

Actually I start counting the KP from 1 after legendary, so those people got next legendary (almost) right after last one. When I started this thread I had data from 9785 chars, but people stop playing or change servers, so for KP stats I cut out everyone who had less than 200 KP, to exclude inactive players. Then the data is sorted by KP. I parsed all data using the timestamp of each entry, which has 1 sec accuracy.

imgurDOTcom/a/xiVvm - The histogram I mentioned earlier.
Though you could have shown an example of this if in your "Bad Luck" protection tables in Section 2, you had done the percentages based on the total population instead of the remaining populations. Which is why the total win percent is higher than 100. It should look more like this:

The purpose of these tables is to show things like "drop chance per 100 KP" so we can compare how much BLP and number of legendaries affect the drop rate. But I will try to further analyse the accuracy of these numbers and improve the presentation.

Confirmed that it's not. I've been doing some 10+ runs yesterday and have the same 168 2+ runs in my 33096 criteria: http://www.wowprogress.com/character...нмеркар

It might be some issue with updating, or you didn't beat the timer on 10+ and those don't count, did it increase the 10+ counter? Your stats say 168/129/17 for 2+/5+/10+ but in your statistics it says Challenge mode dungeons completed: 290, this tracks mythic+ completions too(on top of old challenge mode) and 168+129+17 would be 314. Btw. world top M+ farmer has numbers 2086/1671/57 and Challenge mode dungeons completed: 2,113.

[Deleted]

Literally have no idea why you wouldn't make them all equal drop rate, never mind the soft cap debacle. Maybe its just inflated because of shared class loot though. Even so why do any of those exist.

What a joke.

To keep you playing until the next expansion, it was implemented after the first legendary "bug", when they realized that everyone would have their legendaries rather quick. They will be lowering the diminish return as the expansion go to alleviate the player stress induced by the system, when the next expansion is 6-8 away they will increase the droprate and ease the game extensively for alts so you keep subbed those months too while we are on a content draught.

Truth is, it not implemented as an honest game system for the players enjoyment, but a sub retention mechanism in the long run.

[Leetbeartank]

Thank you for doing the work to shut up everyone who thinks bad luck protection doesn't exist.

He just proved that it's fucking garbage and bull shit. Bad luck protection isn't enough. If you get 1, 2, or 3 (like me) useless legendaries and then you're completely and utterly fucked until they fix something.

[Deleted]

Not a single person... Not even a single person that I've talked with since the beginning of this expansion said that they enjoy the current legendary system. What is the main reason that we all hate this legendary system? This is probably going to be a long post, but please read it and share your opinions.

1- RNG Nature : Legendaries can be grouped into 3 : Performance enhancing, utility, defensive. Performance enhancing legendaries are obviously the ones that everybody is looking for. So let's talk about these performance enhancing legendaries for DPS classes a bit.

For most of the DPS classes, their BiS legendary is rougly a 10-15% DPS increase if not more. Wanting the best is in the human nature, and legendary items are no exception to this. We, as DPS players, want the best possible legendary for our class so that we can perform better and achieve better results. Some of you may ask : "Why are you obsessed with performing better and achieving better?". Here is my answer to this, because it is what makes this game fun for me. Competing with my guildies, competing with players from all other parts of the world for achieving the best rankings, percentiles is really fun for me. No, this doesn't mean we are DPS whores, this just means that we are players with a competitive mindset and that's what we enjoy doing in this game. Some of you enjoy fishing in game, some of you enjoy levelling alts all the time, some of you are just playing the game for friends and you name it.

Now the first problem starts at this point : You have absolutely no control over getting that BiS legendary for your class. It is completely random and the only thing you can do about it, is just hope that you will get lucky soon and obtain it. Such legendaries that grant you an immense power can not be based on pure luck. If you were lucky enough to get it, fine enjoy the top parses and amazing dps. However if you are one of the unlucky players that has yet to lay hands on your BiS legendary, you start to get more and more frustrated as time passes.

2- The Uncertainty : We can say that it has been 1,5 raid tiers since the Legion was launched. So 1,5 raid tiers have already passed and some of us are still playing without our BiS legendaries. At this point, not knowing when you will get your BiS legendary is one of the biggest sources of frustration among players. For how much longer are we supposed to wait? 1 more raid tiers? 2 more raid tiers? 6 months? 10 months? Am I just supposed to be okay playing without my bis legendary, for lets say 2 more raid tiers?

Don't say that it is not possible, because you all know that it is completely probable that you will not get your BiS legendary for a very very long time. I'm talking about 9-10 months and possibly even more here. How?

Currently we have 8 obtainable legendaries for all specs in the game. In 7.1.5 all classes are getting 2 more legendaries to their loot table, so after 7.1.5 the number of legendaries you can "obtain" will be 10.

For most classes there are 2 legendaries that are the absolute BiS. For arms warriors, for example, the execute ring and the gloves are by far the best ones. So the 2nd big problem starts right here. As we all know, the amount of effort that is required for your next legendary increases parabolically. Which means that each time you get a legendary, you have to do more "legendary eligible content" for your next legendary drop. So the problem is, if you are a player who was unlucky to get mediocre/bad legendaries, bad news, you gotta work much harder for a "chance" to obtain your bis legendary next time you get one.

So yes, it is completely probable that you will get ALL OTHER 8 LEGENDARIES except those 2 BiS legendaires for a veryyyyyyy very long time.

Let me give you an example, currently there are very few players with 5 legendaries. Even people with over 1000 m+ dungeons completed along with constant mythic raid farming are currently having 4 because you simply need A LOT of effort for your 5th one. Imagine yourself in a situation where you have 4 legendaries, 2 mediocre and 2 bad ones. Can you imagine how hard you need to work for your 5th legendary? And after putting in all that hard work, imagine getting another mediocre/bad legendary but not the bis one, yet again. This can keep going for a VERY long time. For how long is a player supposed to just wait until getting his BiS legendary? A year? Until the end of the expansion where it doesn't even matter anymore?

This uncertainty is by no means "a fun experience that you will share with your friends" but rather a disappointing/annoying one that will make you quit the game eventually. The Nighthold will become available in about 1,5 months from now and as we all know, a raid tier lasts for 2-3 months for your average guild. There will be many players who will spend another raid tier without their BiS legendary. For how long are they supposed to wait? Maybe forever because it is completely probable that you will not get your BiS legendary for the entirety of this expansion. Is this supposed to be fair and fun? I don't think so.

3- Bad Legendaries : Speaking from a DPS player point of view, nobody wants prydaz or sephuz or norgannon's foresight because they are simply not a DPS increase. You know that there is something terribly wrong going on with the "legendary" system when people straight out laugh at the player who gets prydaz/sephuz or any other crap legendary. You are supposed to be happy when you get a "legendary" item, not frustrated and disappointed.

This is how it exactly feels when a dps player gets one of the useless legendaries instead of the BiS one. So if the whole point of the current legendary system is to make you feel "unique, excited and happy when you get one", then there is something terribly wrong with it. In the current system, even when players loot their BiS legendaries, they feel happy of course but they also feel "relieved" because they could've received Prydaz, The Destroyer of the Dreams. You got the point.

A havoc demon hunter doesn't want the blur chest, a hunter doesn't want "stand still"legs, a mage doesn't want the blink chest, a rogue doesn't want the feint legs. Lets all be honest, there are legendaries that nobody wants, other than prydaz and sephuz. So what was even the point to make a "legendary" like that? They do not make people happy, they make people annoyed and frustrated.

4- Imbalance Between Legendaries As I mentioned above we can categorize legendaries as performance enhancing, utility, defensive. The problem is, the performance enchancing legendaries are very imbalanced and as a result of this there are only 2 legendaries for each spec that players really pursue. For each spec, there are 2 legendaries that are best for pretty much any situations, and then there are 3-4 other performance enhancing legendaries that are extremely situational.

Lets take havoc demon hunter legendaries for example. For havoc, among the performance enhancing legendaries, the ring is by far the best one, being roughly a 10% dps increase. The belt comes right after that. These 2 items will almost always be a DPS gain for you, meaning that regarless of what the situation is, AoE/Cleave/Single target, these legendaries will provide a significant DPS boost. On the other hand the head, bracers and the boots for havoc are extremely situational legendaries and will only provide a bonus IF their "niche" is present.

For example if you have the head and bracer legendary for havoc DH, you will have 0 dps gain from those for single target because they simply have no effect on 1 target. In order for them to be effective there has to be mass aoe or constant cleave all the time so that you can feel the "legendary" effect.

5- Carrot on A Stick and Long Term Profits : Yes, at the end of the day Blizzard is a company which's main goal is to make sustainable profit. Since WoW has a subscription based model, the most effective way to make profits out of WoW is to maintain a high number of subscribers for as long as possible. Obviously, in order to keep players playing, Blizzard has to find a way that will keep players playing, they need a system that will be the "carrot" on that stick.

So at this point, it makes perfect sense from a business point of view to implement such legendary system. The player plays the game and tries to hit the jackpot on the slot machine. Failed? No problem, keep trying. Keep trying and trying, MAYBE in 5 months you will get that amazing legendary, but for that you need to keep playing. You need to keep that subscription active, you need to keep paying us money.

If this is actually their logic behind implementing this abysmal legendary system, I'm sorry for their managers but they are short sighted and they can't understand what players feel and want. Not even the lucky players who has 2 BiS legendaries are happy with this system. Right now, the unlucky players dont feel like playing more and trying harder for their next legendary, they feel burned out and demotivated. Just yesterday one of my guildies decided to step out of raiding because he basically got the shitty feint legs for his rogue, and his first legendary was prydaz. No blizzard, this player doesn't feel like playing more and keeping his subscription, this player feels like quitting because he is tired of this stupid legendary system. He is tired of having absolute no control over legendaries.

I really really really hope that blizzard will understand this legendary system is ruining the expansion for most players before it's too late.


Here are my suggestions to fix this legendary system and implement a better one

Option 1 : Let us "target" our next legendary drop. I am fine with farming hundreds of m+ dungeons and other "legendary eligible content" as long as I KNOW what will my next legendary be. I am sorry for slot machine addicts at this point, but the possibility of getting prydaz, sephuz or any other useless legendary for my 4th one is by no means exciting for me. It's stressful.

Option 2 : Introduce a legendary token system. This is pretty much the same thing with the first option, just in a different form. Basically you get a "legendary token" and with that token you get to choose a legendary. Too boring? Okay, make a special and unique questline before acquiring that legendary that we want, or anything else. You are getting paid for it, it is your JOB to think of something unique

Option 3 : Introduce a legendary questline on each new raid tier, which requires you to kill the end boss of that tier in mythic difficulty. After you kill the boss on mythic difficulty, you choose a legendary you want. This system would motivate players to get that last mythic kill so that they can get their bis legendary, which will be ready for next raid tier.

Another alternative could be : You get a legendary questline, you have to collect 20 "legendary tokens" from bosses on The Nighthold. These tokens only drop from heroic and mythic difficulty (shared) and has a 33% drop rate for each boss. By this way the gap between lucky and unlucky players will be 1-3 weeks which is acceptable. When you collect 20 "legendary tokens" you get to choose a legendary.

In my opinion, a system like this would make everyone happy. You can basically choose a legendary at the beginning of each raid tier. This system would help fix the inequality between players who has BiS and who doesnt so that they would start each content on equal or at least very close conditions. Considering that an average raid tier lasts 5-8 months in WoW, I believe it is perfectly fair and logical to give players a guaranteed legendary every 5-8 months. They can keep the RNG part as long as they will give us guaranteed legendaries like this, every raid tier.

Option 4 : If you really really really want to keep the drops on RNG basis, at least give us a certain way to get a guaranteed but random legendary drop. A very basic example could be : Complete 100 level 5+ m+ dungeons (level 5 and above so that it won't be super easy) to get a legendary drop, but you cant control which legendary will drop. Too easy? Make it 150, but just give us a certain way for it.

Option 5 : The legendary trader. You trade 2 of your current legendaries for 1 that you want. This is just the basic idea, it is their job to present this in a cool and logical way.

Another form of this : Destroy your current legendaries in obliterum forge. Each time you destroy a legendary you get 100 legendary dust. Create a "legendary token" out of 200 legendary dust and get a new legendary with that token.

Option 6 : Remove the diminishing return on the bad luck protection. As you all know, it gets harder and harder for you to obtain your next legendary. If you need to kill 10 mobs to obtain from 0 to 1 legendary, you have to kill 20 mobs from 1 to 2 legendary and so on. The problem is, if you are one of the unlucky players who are stuck with 2 mediocre + 2 bad legendaries, you have to put SO MUCH effort to the game for a CHANCE to get your BiS legendary on your 5th one. This can be a very frustrating and long process, talking about 5 months+ here.


I would like to see your feedback/suggestions about the options that I provided to fix the legendary system. I'd be happy to hear which one you like most/disliked and discuss.

[barackohmama]

Maybe a bit OT but what are those "other/untrackable" drops? I've heard that you can get legendaries from withered scenario, but I haven't seen any confirmation on it, or if there are any other sources.

[kryptik93]

Just in case if someone is curious how much of KP he has: docs.googleDOTcom/spreadsheets/d/1uRPy-cc98TLHscgqMyjAz7yD3ON_dhaPDIr_Nand07U/edit#gid=474315725
Why this information is useful? I don't know. It's not? But you could whine to your guildies about how unlucky you got with the legendary drop, at least!

P.S. Silly mmo-champion isn't allowing me to post links, so you have to deal with it being posted this retarded manner. Someone else could repost clean version of it if you guys would like.

Who's the owner of this and how long does it take to gain edit access?

[Shadocat]

While this spreadsheet is nice, people need to keep in mind that despite the bad luck protection(the spreadsheet shows its there), its pure RNG. Trying to find some sorts of patterns or whatever is completely pointless.

Well, you *wish* it is pure RNG but with all the bugs right now and the lack of effort to fix any of this mess I would no be to surprised if the random number generator is bugged as well, or more the pile of rules that is applied to an assumed role of the dice to determine if it was successfull or not.
There is just no way to tell from the outside though, too many variables we do not even know about.

That said, a chart like this is amusing at best.

[zoned]

So yes, it is completely probable that you will get ALL OTHER 8 LEGENDARIES except those 2 BiS legendaires for a veryyyyyyy very long time.

Let me give you an example, currently there are very few players with 5 legendaries. Even people with over 1000 m+ dungeons completed along with constant mythic raid farming are currently having 4 because you simply need A LOT of effort for your 5th one. Imagine yourself in a situation where you have 4 legendaries, 2 mediocre and 2 bad ones. Can you imagine how hard you need to work for your 5th legendary? And after putting in all that hard work, imagine getting another mediocre/bad legendary but not the bis one, yet again. This can keep going for a VERY long time. For how long is a player supposed to just wait until getting his BiS legendary? A year? Until the end of the expansion where it doesn't even matter anymore?

You just described my exact situation. I currently have 4 legs and last drop was 3 months ago! Since 4th leg drop I have completed over 900 M+ dungeons, every emissary, daily, weekly lockouts etc. Three of my legednaries are completely worthless and one is practically worthless. If my 5th legendary drop also happens to be bad i'm left with no choice but to quit. Just the thought of grinding for another 6 months makes me want to puke. NO THANKS! You can keep your legendaries Blizzard. You would actually have to be some sort of masochist to keep playing after 5 worthless leg drops.

[dcc626]

You just described my exact situation. I currently have 4 legs and last drop was 3 months ago! Since 4th leg drop I have completed over 900 M+ dungeons, every emissary, daily, weekly lockouts etc. Three of my legednaries are completely worthless and one is practically worthless. If my 5th legendary drop also happens be bad I''m left with no choice but to quit. Another 6 months of grinding for a chance at decent 6th leg? NO THANKS! You can keep your legendaries Blizzard. You would actually have to be some sort of masochist to keep playing after 5 worthless leg drops.

legion has been out for just over 3 months. are you saying you got 4 legendaries in the first 2 weeks of the xpac?

[zoned]

legion has been out for just over 3 months. are you saying you got 4 legendaries in the first 2 weeks of the xpac?

Legion has been out for 15 weeks. I got 4 legs in the first 5 weeks of intense grinding. So it's been almost 11 weeks since last leg drop or about 3 months.

Would it honestly make a difference if it was 11 weeks or 3 years? At the rate I grind the leg system is obviously still gated with diminishing returns.

[lehadnk]

Who's the owner of this and how long does it take to gain edit access?

File - Make a copy...

Your stats say 168/129/17 for 2+/5+/10+ but in your statistics it says Challenge mode dungeons completed: 290, this tracks mythic+ completions too(on top of old challenge mode) and 168+129+17 would be 314. Btw. world top M+ farmer has numbers 2086/1671/57 and Challenge mode dungeons completed: 2,113.

You can't use challenge mode statistics (as well as data in challenge mode records array) since it's totally bugged and doesn't keeps an information about all the mythic+ runs person did. So looking at achievement criteria is the only reliable source of such information, unfortunately (which is kinda sad, because looping through challenge mode records you could also take +2/3 chest runs into account, which would give us access to much more precise stats for mythic+ runs if this method were working).

[darkChozo]

I'm not sure what kind of graph you were looking for, but in your graph "% second leg vs. KP" you compare those who got the legendary within the first 10 KP, only to people who have collected max 10 KP, but that excludes everyone who collected more than 10 KP so I don't think this % can give meaningful info.

But if you edit the formulas on that page to something like this, it should give a "drop chance per KP" curve, similar to the BLP stats I made before:

=countIFS(ALL!$G:$G,CONCATENATE("<=",$A12),ALL!$G:$G,CONCATENATE(">",$A2),ALL!$C:$C,"0",AL L!$D:$D,">=2")/20
=countIFS(ALL!$G:$G,CONCATENATE(">=",$A12),ALL!$C:$C,"0",ALL!$D:$D,"1")

Okay, yeah, that was dumb. I was trying to get a picture of how many people had legionaries at particular KP thresholds, but I ended up figuring out that most of the people who are at <30 KP are probably still there because they already got their legendary (duh).

I still feel it's unlikely that there's any appreciable difference between the different legendaries. I've put together cumulative legendary % vs. KP charts for all the legendaries and they're all remarkably similar, leveling off roughly in the 600-800 KP range. The only major difference is the percentage at which they level off, which corresponds with the total population legendary % numbers.

In fact, if you discard the people who haven't gotten a legendary yet so the leveling off happens at ~100%, you get very similar curves, and I feel it's unlikely for this to be a coincidence:

imgur DOT com/6vFxCYu

I realize that discarding the non-winners introduces some pro-winner bias into the system, but you can actually track this in the data. You'd expect this bias to be larger the smaller your "sample size" was, and this is in fact the case. The 4th legendary (7.5%) is a bit above the 3rd, and the 3rd (17.6%) is a bit above the second (36%). For the 4th legendary to actually be 4x as unlikely as the second, this bias effect would have to completely cancel out the multiplier while also appearing to support the idea that all chances are the same; Occam's Razor says probably not.

(1st legendary also lines up nicely and has a very pretty curve but I still don't trust it)

data: docs DOT google DOT com/spreadsheets/d/1cGGnkokih6TYC-fP7WT7pPnGkD0glnEZX77irOC6KAg/edit?usp=sharing

[Deleted]

Wow... it's doubled EVERYTIME exponentially... I got Sephuz and Prydaz... good ones will takes doubled time... some people got their BiS ones first, and will play with it for the whole xpac, and when you get it last it takes doubled time + doubled + doubled...

[Deleted]

He just proved that it's fucking garbage and bull shit. Bad luck protection isn't enough. If you get 1, 2, or 3 (like me) useless legendaries and then you're completely and utterly fucked until they fix something.

Don't worry, next one will be when you change loot spec for that trinket!

[kryptik93]

7.1.5 would be the best time to re-introduce glyphs... keep legendary items higher ilvl the same way they are than you usual epics... but apply their passives into glyphs where players can choose 2. But one could only dream, right?

[Deleted]

Am I the only one that finds it weird that it's not an accecible % on the wow ui? it's just more cloak and dagger from the devs team about this awfull awfull system.

[impending doom]

I still feel it's unlikely that there's any appreciable difference between the different legendaries. I've put together cumulative legendary % vs. KP charts for all the legendaries and they're all remarkably similar, leveling off roughly in the 600-800 KP range. The only major difference is the percentage at which they level off, which corresponds with the total population legendary % numbers.

There are many ways to get the results biased in one way or another. For example in your chart "2nd leg cdf" 1285 people have looted 2nd legendary within 500 KP, now you divide that by total pop 3977, but that doesn't count in that many of those people never reached 500 KP at all. Also I'm not sure how well they can be compared if they're on different % scale. We have to find some method that is as unbiased as possible, and to be able to compare the effort needed at different levels.

I made couple new charts which might be more accurate than the ones in my november update. And these also are showing some kind of dimishing returns:



Full data: https://docs.google.com/spreadsheets...it?usp=sharing