DLC % logic?

[Dry]

IMO the DLC trophy rarities should be based off the total game owners, not DLC owners (tracked users that have earned at least one trophy from a DLC trophy pack) and not by geometric mean either (however the fuck that works, I haven’t done any actual math since finishing school. Please help me...). I think that with the amount of easy game made available in the past five years (thanks Mayo) and all the accounts that focus purely on these games I’d prefer we use the normal calculations for DLC trophy rarities, therefore decreasing the value of the Ultra Rare rarity if it means those accounts will display a lover average rarity because trust me - they’re gonna need it.

 

By the way, there should really be a poll for this discussion. I can’t keep having these same forum threads being created once a year for me to waste my time writing the same shit on a different day without any conclusion to the topic at hand.

[DrBloodmoney]

2 minutes ago, Slim Vanilla said:

IMO the DLC trophy rarities should be based off the total game owners, not DLC owners (tracked users that have earned at least one trophy from a DLC trophy pack) and not by geometric mean either (however the fuck that works, I haven’t done any actual math since finishing school. Please help me...). I think that with the amount of easy game made available in the past five years (thanks Mayo) and all the accounts that focus purely on these games I’d prefer we use the normal calculations for DLC trophy rarities, therefore decreasing the value of the Ultra Rare rarity if it means those accounts will display a lover average rarity because trust me - they’re gonna need it.

 

By the way, there should really be a poll for this discussion. I can’t keep having these same forum threads being created once a year for me to waste my time writing the same shit on a different day without any conclusion to the topic at hand.

 

 

The fact of the matter is - even 7 years ago, prior to the most recent change - basing the rarity of the DLC off the base game owners was the favoured majority opinion on the site, and anecdotally, the sentiment seems to be even more in favour now than then.

 

In 2015 I started a poll asking this very question, (base game vs. earning 1 trophy - the 'geometric mean' approach was but a rancid apple in the site's eye at that point ?,) but because the results were skewing 2:1 in favour of using the base game data, one of the Mods who disagreed closed the thread...

 

...as a result, I don't imagine it is likely to get changed, but I'll be damned if I'm going to stop talking statistical sense just because some people refuse to put their hands in their pockets or learn how percentages work ?

 

Edited May 19, 2021 by DrBloodmoney

[TurtlePM]

4 hours ago, mekktor said:

 

Also not a good reason. Every game on PSN is (or was) available for everyone and people chose not to pay/start them. Does this mean every player should be counted in the rarity calculation for every game? Clearly we only count people who actually do pay and start the base game, and the only reason that isn't true for DLC is because that information isn't available.

 

 

Like I said, considering psnp % rate counts DLC trophies you don't own, yes,  it should count as well, it showed be calculated based on the total number of players. 

 

Although let me put this straight: I don't really care if the trophies are UR like a lot of people are complaining or implying. It's just dumb to watch a game like Gems of War with free DLC to everyone where a "Reach level 100" has a rarity way lower than "Reach level 250".

[You]

I don't think the DLC numbers should be manipulated like because in other calculations they are included. It also makes for odd statistics. The two biggest being

 

- DLC with one trophy is always 100% rarity

- missing DLC trophies are still counted towards your overall trophy percentage

[MrHealthy]

The big problem that I have with the argument that DLC rarity percentage should count all base game owners to be more accurate is that by default this sites base game rarity percentage are not accurate to begin with. By default it shows you the rarity of the trophy based on the profiles that have been updated on this site. This is widely inaccurate as many profiles are not on this site to begin with. As well even if the profile is on this site if it hasn't been updated in years it could have a trophy that it didn't before its last update. The result is more often then not percents on here that are far higher then they are according to PSN's own calculation.

 

There is an option to switch the default percentage view to PSN's which accurately displays both base game and DLC rarity.**

 

Sadly that option only reflects on the games main page. So the solution here should be that the PSN percentage option should be reflected everywhere if you have that option set. On the games page, on individual trophy pages, in your trophy log and on your rarest trophies box. I don't know how easy that would be to implement, but it would allow for people who want PSN accurate stats to have them while keeping PSNP's method of calculation for people who prefer that.

 

** It can sometimes be off by 0.1%, I assume that has to do with how often this site grabs the stats from PSN.

[Bumperklever]

1 hour ago, DrBloodmoney said:

 

 

The fact of the matter is - even 7 years ago, prior to the most recent change - basing the rarity of the DLC off the base game owners was the favoured majority opinion on the site, and anecdotally, the sentiment seems to be even more in favour now than then.

 

In 2015 I started a poll asking this very question, (base game vs. earning 1 trophy - the 'geometric mean' approach was but a rancid apple in the site's eye at that point ?,) but because the results were skewing 2:1 in favour of using the base game data, one of the Mods who disagreed closed the thread...

 

...as a result, I don't imagine it is likely to get changed, but I'll be damned if I'm going to stop talking statistical sense just because some people refuse to put their hands in their pockets or learn how percentages work ?

 

We should make a new poll honestly with the options, base owners, dlc owners and the calculation method . see how it would go in 2021. 

 

I'm in for the game owner is based % as how it is on most other platforms as well. The others just seem to me like alternative metrics. 

 

btw what is the argument for platinum hunters against the idea of base game owners? why do they care, they only hunt platinum right?

Edited May 19, 2021 by Bumperklever

[DrBloodmoney]

8 minutes ago, Bumperklever said:

We should make a new poll honestly with the options, base owners, dlc owners and the calculation method . we will see how it pens out. 

 

I'm all for a new poll - you know where my vote would go ?

 

Quote

btw what is the argument for platinum hunters against the idea of base game owners? why do they care, they only hunt platinum right?

 

As far as I can tell, the only real arguments is "I don't want to feel like my not finishing DLC is diminishing my Platinums, and other people getting more Ultra Rare trophies does that" - which I can personally sympathise with, though I find it a strange argument, given that losing an S-Rank and a 100% seems much more of an obvious and visible diminishment of a Soft-Platinum than someone else's rarity percentage is. 

 

In the end though, maths is maths - it shouldn't be manipulated just to placate feelings.

 

I mean, if I lost a testicle, it might make me feel better if I came up with some protracted formula to 'prove' that 1 = 2 ...

...but that wouldn't mean that I could use that formula when doing my taxes.

Edited May 19, 2021 by DrBloodmoney

[Cleggworth]

35 minutes ago, MrHealthy said:

As well even if the profile is on this site if it hasn't been updated in years.

 

This is my other issue with the site at the moment. Again a consistency thing....

 

Auto updating was off for a long time. Then when it came back you had to manually input an ID to add it back to the list to be updated. There are thousands of accounts that have not been updated since, possibly tens and hundreds of thousands of accounts. Now I understand the argument for not updating them, the stats are more reflective of the actual trophy hunting community etc. I'm leaning towards agreeing with that as its makes sense to me.

 

The problem is again that consistency. I don't remember specifically when it happened but for arguments sake I'll say it was Jan '18. Every trophy list before this point has thousands of dormant accounts contributing to that games stats. After this point they aren't there. You cannot use the stats to compare two games either side of this point because the data is different. 

 

Either update all profiles on the leaderboard or remove those that haven't been updated in X amount of time. 

 

Use the base game owners for DLC trophy rarity like everything else or change it all to DLC owners. I don't know how you do that with the system as it is now ?‍♂️

 

[zid2016]

If you are arguing to use the base game owners then use the psn rarity. That reflects the whole player base instead of a small sample size,that is the psnp members.

[DrBloodmoney]

2 minutes ago, zid2016 said:

If you are arguing to use the base game owners then use the psn rarity. That reflects the whole player base instead of a small sample size,that is the psnp members.

no, that’s not true.

 

PSN includes everyone, not just trophy hunters - and it is therefore less useful.

 

Using base game owners on PSN values is like measuring the popularity of a particular style of bridal dress by measuring how many people wear it in a country - across men, women and children of all ages.

 

Using base game owners from PSNprofiles members is like asking only brides and potential brides.

 

Both are true, but one is of significantly greater value.

[EDGES]

On 5/18/2021 at 1:56 PM, Arcesius said:

Because the rarities are not computed based on the number of base game owners. 

 

Instead, the number of DLC owners is manipulated. It is computed as the geometric mean between the number of base game owners, and the number of players that have achieved at least a trophy in the DLC. 

 

It often results in NG+ trophies being less rare than NG trophies, for example. 

 

It is a manipulated, misleading, manufactured rarity, but it is what it is. 

that's crazy, addon trophies should be just like main trophies

[zid2016]

7 minutes ago, DrBloodmoney said:

no, that’s not true.

 

PSN includes everyone, not just trophy hunters - and it is therefore less useful.

 

Using base game owners on PSN values is like measuring the popularity of a particular style of bridal dress by measuring how many people wear it in a country - across men, women and children of all ages.

 

Using base game owners from PSNprofiles members is like asking only brides and potential brides.

 

Both are true, but one is of significantly greater value.

Then, wouldn't it be more objective to base dlc rarity on this site based on the players who have played the dlc instead of base game owners.

[DrBloodmoney]

4 minutes ago, zid2016 said:

Then, wouldn't it be more objective to base dlc rarity on this site based on the players who have played the dlc instead of base game owners.

no, because there is no way to measure who has played dlc except by seeing who has earned a dlc trophy. That means that for some dlc with a single trophy or two or three, the rarities are ridiculous - a dlc with a NG+ And nothing else in it would always be 100%, since the only way to measure who has played it is by who has already earned it.

[MrHealthy]

10 minutes ago, DrBloodmoney said:

no, that’s not true.

 

PSN includes everyone, not just trophy hunters - and it is therefore less useful.

 

Using base game owners on PSN values is like measuring the popularity of a particular style of bridal dress by measuring how many people wear it in a country - across men, women and children of all ages.

 

Using base game owners from PSNprofiles members is like asking only brides and potential brides.

 

Both are true, but one is of significantly greater value.

There are millions of accounts on this site which are not 'trophy hunters'. Anyone can add any account to this site so accounts get looked up by friends, alts get added, etc.

 

Looking at the leaderboard from page 42,473 to the last page of 84334 accounts are all under level 100. The vast majority don't have a single platinum. At 50 accounts per page thats 2,093,050 accounts. That is just shy of 42% of all accounts tracked on this site. There are still hundreds of thousands of accounts without a single platinum tracked on this site that I didn't include in that calculation because they happened to be over level 100. I'm seeing pages and pages of 0 platinum accounts all the way from pages as early as 20,000. Using your analogy I'd wager that probably close to 70% of tracked accounts are men and children rather then brides.

 

There is no additional value here. It's just a smaller sample size, but just as random as looking at the full picture.

[Shikotei-kun]

Oooohh, math problems!

 

I'm surprised nobody has yet explained what the rarity of the DLC trophies are supposed to reflect, but I can take a gander.

 

Take a game with no DLC. PSNP can accurately determine the number of people who have played this game, regardless of having earned a trophy or not.

So of each earned trophy the rarity can accurately be determined:

nrEarnedTrophy / nrOwnGame * 100% = rarity

 

If that game gets DLC trophies added, PSN determines the rarity according to the same formula:

nrOwnGame is equal to the amount of players who own the (base)game.

 

What PSNP is trying to do, is set the rarity of the DLC trophies based on players who own the DLC (regardless of trophy earned).

The base game owners are to be ignored here, and use nrOwnDLC instead of nrOwnGame.

The problem lies in the inaccuracy of nrOwnDLC. Simply because PSN does not provide a means to determine this.

 

So PSNP has to calculate or make a guestimate to determine nrOwnDLC.

 

- Number of people who earned any trophy in the DLC = nrOwnDLC

Issue/case: DLC has only 1 trophy.

Result: Rarity = 100%

 

- Number of people who own the base game = nrOwnDLC

Issue: Not everyone owns the DLC in reality

Result: Heavily inflated rarity across the list

 

- Geometric mean to "guess" number of DLC owners higher than number of people who earned any trophy in the DLC = nrOwnDLC

Issue: Varying results that may be a gross misrepresentation of the actual number of DLC owners

Result: Inaccurate representation of rarity, but more accurate than previous two methods.

 

My two cents.

[DrBloodmoney]

7 minutes ago, MrHealthy said:

There are millions of accounts on this site which are not 'trophy hunters'. Anyone can add any account to this site so accounts get looked up by friends, alts get added, etc.

 

Looking at the leaderboard from page 42,473 to the last page of 84334 accounts are all under level 100. The vast majority don't have a single platinum. At 50 accounts per page thats 2,093,050 accounts. That is just shy of 42% of all accounts tracked on this site. There are still hundreds of thousands of accounts without a single platinum tracked on this site that I didn't include in that calculation because they happened to be over level 100. I'm seeing pages and pages of 0 platinum accounts all the way from pages as early as 20,000. Using your analogy I'd wager that probably close to 70% of tracked accounts are men and children rather then brides.

 

There is no additional value here. It's just a smaller sample size, but just as random as looking at the full picture.

 

Not every account on this site is a trophy hunter - that's true - but virtually all trophy hunter accounts are on this site.

 

Does using the user base of PSNP give a perfect, 'Trophy-Hunter-Only' sample set?

No.

 

Is it the best way to get as close to a that goal possible? 

Yes - given that PSNP is the most used trophy-hunting site, with the largest user base.

 

 

I really don't see how you can look at a situation where 2million 'non-trophy hunter' accounts are skewing statistics in one direction or another, and conclude that the solution is to add 96 million more.

 

Regardless - it is immaterial - the stats you are wanting to use are freely available - both on PSN and on PSNP, and the user options on the site allow you to choose between them. 

My argument is more about how to calculate the PSNP-specific ones - the PSN ones will always be available direct from Sony.

Edited May 19, 2021 by DrBloodmoney

[MrHealthy]

19 minutes ago, DrBloodmoney said:

I really don't see how you can look at a situation where 2million 'non-trophy hunter' accounts are skewing statistics in one direction or another, and conclude that the solution is to add 96 million more.

I never once said I wanted this. I am happy with the current system for both base game and DLC as I find it most useful.

 

I am simply arguing against the notion that including all owners in DLC rarity is more accurate. It's not if we are already fudging the numbers to begin with.

 

I am fine with fudging the numbers for more useful stats. To me personally, including all owners in DLC rarity isn't useful as all the new ultra rares result in less actual useful information to be gleamed from the stats. Because the fact is that I do use rarity here as an indicator of a grindy or difficult trophy. I understand thats not what rarity means, but I use rarity as a first glance into the experience of a particular list. And as you said, the PSN accurate information is available elsewhere if I want to look at it.

Edited May 19, 2021 by MrHealthy

[x7eight]

If there's dlc trophies in a list it should be counted for everyone regardless if you own the dlc or not, the trophies are there and you can't do anything about it besides earn them or don't.

 

It makes no sense having it any other way regardless of who it pisses off.

 

If a trophy list has dlc trophies why treat them differently?  Unnecessarily complex for no real reason.

Edited May 19, 2021 by x7eight

[HusKy]

40 minutes ago, Shikotei-kun said:

- Number of people who own the base game = nrOwnDLC

Issue: Not everyone owns the DLC in reality

Result: Heavily inflated rarity across the list

 

Let's call it "number of people who have the list in their profile". Doesn't matter if you own the game or not. The whole idea of "I own the base game but not DLC" is flawed because it can't be verified either. The only thing we know for certain is that there's a trophy list in someone's profile.

 

And by "heavily inflated", do you actually mean "accurate"? Just asking.

Edited May 19, 2021 by HusKy

[Shikotei-kun]

1 minute ago, HusKy said:

Let's call it "number of people who have the list in their profile".  [..] The only thing we know for certain is that there's a trophy list in someone's profile.

That is what I indeed meant by nrOwnBase. They (at some point) owned the game and got it in their lists. Doesn't matter if they borrowed the game disc, or used their account on a friend's console. They had access to the game.

 

 

1 minute ago, HusKy said:

And by heavily inflated, do you actually mean accurate? Just asking.

Heavily inflated would mean than the rarity percentage is far lower (towards Ultra Rare) than should be.

If the goal of the rarity of DLC must be based on the number of people who own the DLC (or have access or whatnot), then using the base game owners (by definition equal or larger than DLC owners) it is highly inaccurate.

[Xylobe]

42 minutes ago, Shikotei-kun said:

What PSNP is trying to do, is set the rarity of the DLC trophies based on players who own the DLC (regardless of trophy earned).

 

This is exactly what people are taking issue with in the first place-- it's inconsistent with official Sony statistics and it's inconsistent with every other statistic on this site. The problem isn't that we need a way to accurately estimate the number of DLC owners, the problem is that the number of DLC owners doesn't matter in the first place. Even if we knew the exact numbers for every DLC for every game, using them rather than the number of base game owners would still buck convention in all the wrong ways.

[HusKy]

7 minutes ago, Shikotei-kun said:

Heavily inflated would mean than the rarity percentage is far lower (towards Ultra Rare) than should be.

If the goal of the rarity of DLC must be based on the number of people who own the DLC (or have access or whatnot), then using the base game owners (by definition equal or larger than DLC owners) it is highly inaccurate.

 

> than should be

We don't decide what the rarities should be - players do. The whole notion of "this number looks good" and "that number feels right" is completely invalid.

 

> base game owners

There's no "base game owner". There's only "trophy list". Whether the user has access to trophy in base list or DLC list cannot be determined. And doesn't matter either.

 

> base game owners (by definition equal or larger than DLC owners)

Also wrong as showcased by countless number of games with free DLCs.

Edited May 19, 2021 by HusKy

[madbuk]

20 hours ago, djb5f said:

There is no way to count DLC owners and hence the entire issue.

 

At one point, it was using DLC owners who have at least one trophy earned but that is a much different count and skewed the statistics.  It skewed very high for games with very difficult DLC packs.  A really hard trophy could show 90% achieved or something silly.

 

I am fine with how it is now.  The formula is admittedly arbitrary but I use it as a guide to seeing how difficult a trophy is and it works in that regard.

 

The only other option is like psntrophyleaders where the DLC owners is the same as base game owners but that isn’t accurate and skews percentages very low, thereby inflating ultra rare trophies.

 

So any of the methods is broken because you have no idea how many people actually have the DLC.

 

 

How is it not accurate? 100% of Borderlands 3 players also have the Borderlands DLCs on their trophy list, and should be factored into rarity as a result. The DLC is part of the game, whether you've paid to unlock it yet or not is irrelevant.

[BeautifulTorment]

Lots of good points being made on the side to use base game owners as the value to calculate DLC rarity. Not so much on the other side.

 

There does not need to be a geometric mean, and there does not need to be an arbitrary "owners" value based on the number of people who popped at least one trophy in that pack. 

 

I'm just popping in here to voice my opinion on the matter to give support to the correct way of representing a DLC rarity, nothing more. Been beaten to death. Should be fixed to how it was originally, but I doubt it ever will.

[scemopagliaccioh]

Rarity equals effort, whether it’s because you got easily carried out by others more good than you online, whether because the game is dogshit  and nobody plays it, whether because the game is Grindy, or hard as hell, those are all things that require dedication to persevere, some like the online carry a little less, but you could argue that at some point people will get tired of you asking for help.

games like the Atellier thingy mentioned, no idea how they work but, the 20% sample of people is a LOT less than the 6%, so we’re talking about  a concentrated  lesser number of people who on average know more about the game.

Personally, I’m okay with using  base game numbers value on free updates. Paid, though? You’re better off stop deluding yourself and go buy that church play thing.

Edited May 19, 2021 by scemopagliaccioh