Trophy Rarity Question

[djb5f]

ok, I see what you are saying.  I am taking a broader interpretation of difficulty and you a more narrow one I think.   To me taking the time to set up boost sessions and grind that much plays into overall difficulty but I see where you are coming from.  Pure difficulty is hard to measure because basically most anything can be boosted, exploited, glitched if you really wanted.  There are whole websites dedicated  to getting the trophy as easy and cheaply as possible.

[thegirlruka]

ok, I see what you are saying.  I am taking a broader interpretation of difficulty and you a more narrow one I think.   To me taking the time to set up boost sessions and grind that much plays into overall difficulty but I see where you are coming from.  Pure difficulty is hard to measure because basically most anything can be boosted, exploited, glitched if you really wanted.  There are whole websites dedicated  to getting the trophy as easy and cheaply as possible.

That's what I'd refer to as tedious rather than outright difficult. I refer to needing skill as difficult while needing time is tedious.

[djb5f]

I definitely understand where you are coming from.  nonetheless, you can't really argue the correlation between rarity and difficulty.  it is the #1 factor as trophies that are rated with a high difficulty are far away much more rare than trophies that are rated easy.  it is the #1 factor, with other contributing factors being time and interest.

 

To me, the rarity is actually a better representation of challenge to get a trophy than a trophy difficulty rating that often doesn't have many people voting and not tailored to individual trophies.

 

Not a perfect science, but without question, my rarest trophies (<1%) were absolutely the hardest to get for gamers.  They wouldn't be under 1% otherwise.

[ProfBambam55]

so i've just been through my profile again this time ranking trophies in order of a) achievers, b)rarity and then c) by calculating some of the rarities with the old system (or the PSN way if your prefer; dlc achievers/base game owners)...here's what i found:

a) is a good indicator of how rare, by definition, a trophy is...this stat exists so no need to have it in our rarity %...

b)although a good indicator, in general, it gets confusing when DLC only has a few trophies...if i was looking at my trophy log for TLoU (grounded mode) , Metal Gear Rising (DL-VR missions) or Mirror's Edge (90 stars in time trials) i would think "great, 97% rare...i'll do these first..."

c) wow, all my DLC trophies are ultra rare...

and there lies the problem...so i found myself rereading all 14 pages of the thread in search of a solution...up to this point the following has been suggested...please bear in mind the two things that mess up everything in attempting to do rarity %s justice...1: we have dlc trophies added to our profiles regardless of ownership...2: there is no way of actually knowing the true number for DLC owners (yes, base game as well although earning one trophy is so close to 100% accurate that it's kind of not worth debating)...:

-calculating DLC rarity based on an average taken between %s for base game owners and dlc owners...simpified it would look something like this > DLC achievers/DLC game owners = X, DLC achievers/base game owners = Y, (X + Y)/2 = rarity...this is interesting but brings up the point that instead of dealing with relatively concrete figures by saying that earning one trophy indicates ownership, we are now dealing with a phantom number of owners...it would also be kind of ambiguous for people new to the stat..."what does rarity mean?"...and would no doubt bring up as many arguments by the same people who think using base game ownership for DLC achievers is unreasonable...

-lowering the % scale for DLC trophies...meaning instead of ultra rare being <5% it could be <1% or something like that and so on for further degress of rarity...this is also quite interesting but would no doubt have critics saying "why is DLC ultra rare 1% and not 5%" and basically just have the same debate as now but with a different instigator...

-removing DLC stats from rarity altogether...this solves our debate to a certain extent but would inevitably bring up "why don't DLC have rarity stats?"...my trophy log not having DLC rarity would be strange indeed but perhaps better this than an inaccurate figure...

-having something that allows us to toggle DLC rarity for our displayed stats...i imagine this kind of thing being like the clean/dirty leaderboards or the "show DLC" option in our trophy advisors...we could have a "CLEAN: without DLC" and "DIRTY: with DLC" for our profiles...interesting and fun but still not a solution since we would need to debate which rarity to display on the DIRTY section...i would vote for the current stat in this case since it represents rarity figures the best IMO but this is also beside the point...it would be interesting to have the option of seeing rarity based only on base game totals though, no?...

-having a box to say "i own/do not own this DLC"...also cool but relies way too heavily on user participation...human error overload...wait, 1000 people forgot to check the box since despite being tracked never actually use this site...let's just flag them and let the mods sort it out...really?...how do the mods know if they've actually played the DLC or not?...because they've earned a trophy?...isn't that already how things are?...

-show both stats...since both the previous and current method of displaying rarity are ultimately flawed (not the fault of the site community by any means) having them both allows us to choose which of the two has meaning to us...they could either both appear appear on our trophy cards, one, or none (dealer's choice)...since, due to their ambiguity, these stats are kind of for personal reference, this solution is sure to please all, no?...haha...

-remove rarity from our profiles...this would stop the debate dead in its tracks but c'mon the rarity stat...THE RARITY STAT, MAN!...since being involved in this thread its meaning has degraded to just about worthless IMO but others would disagree...

-make your own website which tracks rarity however you like it!...haha...this is good and kind of like it is now...it reminds me of the 10 years that i lived in japan...should i say something like "it's interesting that people here do this" of a given scenario, the defensive response was always "if you don't like it you can go back to your country"...despite finding the situation of not being able to calculate DLC rarity accurately frustrating, i'm confident a happy middle ground can be found...

-and then there was this : P = ((T2)/(B*D))0.5...dude i have no idea what that means but i vote you as the guy to figure out the situation mathematically...edit: ah, multiply them together and do the square root...i like how you've written it for this case...how do the stats compare?...working on figuring out how the geometric mean equation applies to the totals...could it be?...could it be?...

-edit: i also just had the thought of creating a ratio of base game owners:dlc owners and then rating the DLC on this average but it basically makes it like just calculating base game owners...example: 591,052 users have earned at least one trophy in the TLoU base game, 7,413 have earned a trophy in the grounded dlc...that would be a stat of 1.25% of base game owners also earning a trophy for this dlc...that would make the complete grounded mode trophy 1.25% (from 100%)rare and grounded plus 1.02% (from 82.11)...this is all good except i'm guessing most DLC trophies would then follow suit and become quite rare despite their simplicity...the average of base:dlc owners would also be kind of a phantom number since people who earned a trophy in the base game didn't necessarily play the dlc...math fail...

so this kind of summarizes where we're at...feel free to add other suggestions that i may have missed...i'm pretty tired at this point and sifting through countless "the stat is/was the best" of varying degrees is not as easy as it may seem...i'm hoping to inspire someone to figure this out...cheers...

Edited October 18, 2016 by ProfBambam55

[Se7en]

the average of base:dlc owners would also be kind of a phantom number since people who earned a trophy in the base game didn't necessarily play the dlc

You’re missing here the point that in case of many games, PS4 version of TLoU for example, DLCs with trophies are already included in the base game. So any game owners not playing these DLCs is not the reason to exclude them from calculations. Well, no one suggests that rarity of online trophies should be counted only based on the number of people who have actually played online content and got at least one multiplayer trophy BTW, much kudos for your GTA4 multiplayer guide, was extremely helpful.

Edited October 19, 2016 by Se7en_Rus

[ProfBambam55]

ok, i'm missing the point...i'm currently thinking of a way to calculate rarity in a way that can be applied to each trophy and each game and am more than willing to look into more specific cases to see if it works once a formula has been found...this subject has actually become the topic of a few lengthy conversations i've had with people over the past couple days...this is interesting stuff and worst case scenario is we learn a little bit more about numbers and math...

i think mekktor was onto something here with his geometric mean...first, a quick link for those interested in geometric mean...frankly i had only ever stumbled upon it when dealing with banks and had heard of it used in population growth stats but had to do some research to see if it was applicable in our case:

https://www.mathsisfun.com/numbers/geometric-mean.html

next, and once again i am not very smart here but will try to brainstorm my understanding of this in hopes that someone with an actual background in math and not just a love of numbers will correct me...i will try to explain in layman's terms why this could work for trophy rarity...please bear in mind that we are looking for a formula that would be universal...also consider the issues we are having, that have been stated as clearly as possible in the post above, in calculating rarity properly and have, up to this point, made the stat inconsistent or inaccurate...

there was a suggestion to use the arithmetic mean(average) to calculate trophy rarity...this means: DLC achievers/DLC game owners = X, DLC achievers/base game owners = Y, (X + Y)/2 = rarity...this implies that there is a middle point between the number of dlc and non-dlc owners that we can use to calculate an average...but a middle point is not what we are looking for...due to the huge difference in the ownership figures and also because the true owners of dlc content is unknown, the ratio between the two has the potential to be much more accurate...how do we calculate that ratio?...you guessed it, geometric average...

so let's have a look at the formula...please refer to the above link for more clarity...mekktor has already laid it out...we will take the square root of achievers/dlc owners*achievers/base game owners...another way of expressing square root is as an exponent 0.5 or to the power of 0.5 if you prefer...we can also combine achievers making it exponent 2 (or to the power of 2) since it is constant for both cases...the simplified formula becomes: (achievers²/(dlc owners*base game owners))^0.5...brilliant...

 

i've attached a simple chart displaying all of the current methods of calculating with a few games' DLC...if we look carefully at the proportions, the geometric value is the most consistent throughout all of the games and kind of keeps difficulty in check...at the bottom i've included a section which shows the flaw in the method though...what happens when there is a game that has %100 DLC achievers...the max stat is capped by the ratio to the base game owners...i would have to say that from what i've seen on this site DLC achievers are more rare than not so this stat also kind of hints at that...mango's solution of changing the rarity scales is very similar to this in a sense but now we would have less arbitration since the geometric average would change the trophies' ratio for us rather than having to guess what would be a suitable categorization...due to our unknown factors inaccuracy seems pretty much inevitable in getting a consistent rarity stat...this formula is the best i've seen so far though...any opinions?... 

Trophy Chart.pdf

[Sly Ripper]

I like the geometric mean. TLoU Survivor at 12.18% and Grounded at 11.20%, Survivor+ at 9.21% and Grounded+ at 9.20%... makes a lot of sense.

[Hemiak]

Looks good to me. Anything is better than having 100% common trophies.

My absolute favorite thing about this thread however, is how positive and supportive everybody has been. Instead of getting all heated and devolving into an argument, people have been calmly discussing the various methods and working together. If it finally ends up changing the state for the better, that's just icing on the cake.

[Satoshi Ookami]

Looks good to me. Anything is better than having 100% common trophies.

Exactly this. 

And if Sly supports it, the better

[ProfBambam55]

haha...cool...i have a database set up on my PC now and a few days off between big contracts at work so would be more than happy to document some game's trophies' stats like in the above pdf if there would be community interest...i would welcome private messages listing games to calculate to see how the rarity totals would work out...i can commit to doing about 50 games (to begin with, no idea how long it will take) should that many requests come in....suggestions, improvements, and further formulae to consider are always appreciated if it leads us to not having to debate this again...i also apologize for trying to take the lead in this, as i fear being seen as an over-bearing tyrant, but felt that someone should act as a sort of mediator in trying to understand all of this and to show that an attempt is being made for all sides to be heard...that being said, i am relatively new to this debate, and despite trying to be unbiaised, may have missed some valid suggestions, critique, corrections, and thoughts...i'm sure this doesn't need to be said but please feel free to attack anything i've posted in trying to find a crowd-pleasing solution to calculating rarity totals...

edit: games that will be added to the list:

- max payne 3

- gta iv

- naughty bear

- batman: arkham city

- the last of us (PS4)

- dmc: devil may cry

- wipeout hd

- the walking dead

- mirror's edge

- alice: madness returns

- red dead redemption

- uncharted 2

- uncharted 3

- call of duty: black ops

- burnout:paradise

- sniper elite 3

- driveclub

- HITMAN (2016)

- helldivers

- ac iv: black flag vs ac iv: freedom cry

- rocket league

- walking dead: season 2

- king's quest

*p.s. looking to have a variety of popular/unpoplar titles, games with single/multiple dlc packs, etc...

Edited October 19, 2016 by ProfBambam55

[FawltyPowers]

I've not had chance to check out the formulas in the pdf you attached but I like the cross section of games you've listed.

 

I'm particularly interested in seeing the results for Uncharted 2 & 3, Red Dead Redemption, COD Black Ops and Arkham City. I remember the pain of completing those DLC's compared to the main game.

[ProfBambam55]

ok, so here is the list of games so far...i've put them in alphabetical order for easy scanning of a particular game...i've also highlighted the ones that would be ultra rare and games that have 2-trophy DLC...

i was supposed to have off work until monday but of course am going to end up working today so don't have enough time to share some of my observations of the stats...will do so when i get a chance later...feel free to pick them apart...i'm also open to calculating more games' totals should anyone have any further suggestions...so far playstation all stars and flow have been added to the list...just a quick reminder to try to keep in mind the unknown variables, that no stat we come up with will be "magic" or perfect, and also that we are trying to find a stat that hints at attempts vs success at the same time as overall rarity/difficult...look forward to hearing peoples' thoughts...

Rarity Chart.pdf

[MMDE]

My absolute favorite thing about this thread however, is how positive and supportive everybody has been. Instead of getting all heated and devolving into an argument, people have been calmly discussing the various methods and working together.

 

Isn't this contradicting? Calmly discussing is all about presenting and discussing arguments, and it can also be a heated discussion at the same time. I guess you mean people haven't fallen to name calling, ignoring arguments and facts presented, and misrepresenting etc? But this has nothing to do with if people feel passionately about the topic arguing calmly but where the result usually is a heated discussion.

 

I'm not sure this is entirely true about this thread though.

ok, i'm missing the point...i'm currently thinking of a way to calculate rarity in a way that can be applied to each trophy and each game and am more than willing to look into more specific cases to see if it works once a formula has been found...this subject has actually become the topic of a few lengthy conversations i've had with people over the past couple days...this is interesting stuff and worst case scenario is we learn a little bit more about numbers and math...

i think mekktor was onto something here with his geometric mean...first, a quick link for those interested in geometric mean...frankly i had only ever stumbled upon it when dealing with banks and had heard of it used in population growth stats but had to do some research to see if it was applicable in our case:

https://www.mathsisfun.com/numbers/geometric-mean.html

next, and once again i am not very smart here but will try to brainstorm my understanding of this in hopes that someone with an actual background in math and not just a love of numbers will correct me...i will try to explain in layman's terms why this could work for trophy rarity...please bear in mind that we are looking for a formula that would be universal...also consider the issues we are having, that have been stated as clearly as possible in the post above, in calculating rarity properly and have, up to this point, made the stat inconsistent or inaccurate...

there was a suggestion to use the arithmetic mean(average) to calculate trophy rarity...this means: DLC achievers/DLC game owners = X, DLC achievers/base game owners = Y, (X + Y)/2 = rarity...this implies that there is a middle point between the number of dlc and non-dlc owners that we can use to calculate an average...but a middle point is not what we are looking for...due to the huge difference in the ownership figures and also because the true owners of dlc content is unknown, the ratio between the two has the potential to be much more accurate...how do we calculate that ratio?...you guessed it, geometric average...

so let's have a look at the formula...please refer to the above link for more clarity...mekktor has already laid it out...we will take the square root of achievers/dlc owners*achievers/base game owners...another way of expressing square root is as an exponent 0.5 or to the power of 0.5 if you prefer...we can also combine achievers making it exponent 2 (or to the power of 2) since it is constant for both cases...the simplified formula becomes: (achievers²/(dlc owners*base game owners))^0.5...brilliant...

 

i've attached a simple chart displaying all of the current methods of calculating with a few games' DLC...if we look carefully at the proportions, the geometric value is the most consistent throughout all of the games and kind of keeps difficulty in check...at the bottom i've included a section which shows the flaw in the method though...what happens when there is a game that has %100 DLC achievers...the max stat is capped by the ratio to the base game owners...i would have to say that from what i've seen on this site DLC achievers are more rare than not so this stat also kind of hints at that...mango's solution of changing the rarity scales is very similar to this in a sense but now we would have less arbitration since the geometric average would change the trophies' ratio for us rather than having to guess what would be a suitable categorization...due to our unknown factors inaccuracy seems pretty much inevitable in getting a consistent rarity stat...this formula is the best i've seen so far though...any opinions?... 

 

I still think this is making up numbers, so it's very wrong IMO.

Looks good to me. Anything is better than having 100% common trophies.

 

Not making up numbers > 100% common trophies.

[FawltyPowers]

First glance is the geometric figures look decent. Trophies hitting Ultra Rare status deserve to be Ultra Rare, it's a far better balance.

 

Personally I don't mind if they were used to re-adjust trophy rarity but I'd have a hard time seeing the percentages displayed. I'd prefer to see percentage completion based off main game owners as this is accurate and meaningful.

 

The only issue you have then though is percentage statistics not matching rarity like it does in main game for instance:

 

UC3 - All Together Now

Old Stat (actual percentage) - 1.36%

Geometric - 5.89% (Very Rare)

 

It would look confusing having 1.36% displayed but with a rarity of Very Rare.

 

If you actually displayed 5.89% as a completion percentage then to me it's just a speculative figure. To me it's better to know as a percentage how many people have earned a trophy who own the game than how many people have earned a trophy who could possibly have the DLC based off a calculation best fit to accommodate any game.

 

Nice work though ProfBambam55. I'm going to try using Geometric myself for rarity.

[MMDE]

I'd prefer to see percentage completion based off main game owners as this is accurate and meaningful.

 

It is not accurate or meaningful... You're making up numbers, just like what was suggested earlier. Unless you know someone owns the DLC, you're making up numbers assuming someone is owning the DLC. In other words, it's not accurate. If you just want to base it on who owns the game, you may as well base each game's % on who owns the console, it makes as much sense, so it's not meaningful.

 

The reason it was changed was because most DLCs was then marked as Rare or Ultra Rare, pretty much always, unless of course most people owned the DLC.... So yeah, it was just wrong in every way, and it was mostly about "paying for ultra rare".

Edited October 20, 2016 by MMDE

[mekktor]

I still think this is making up numbers, so it's very wrong IMO.

 

Not making up numbers > 100% common trophies.

The current system uses made up numbers too, by assuming that the number of DLC owners is equal to the number of people who have earned at least one trophy.

 

Personally I don't mind if they were used to re-adjust trophy rarity but I'd have a hard time seeing the percentages displayed. I'd prefer to see percentage completion based off main game owners as this is accurate and meaningful.

This sounds reasonable, after all this method doesn't promise to somehow tell us the actual percentages. It's just an average of the old and new methods, with the hope that the results we get are more in line with what we expect from the word "rarity".

[MMDE]

The current system uses made up numbers too, by assuming that the number of DLC owners is equal to the number of people who have earned at least one trophy.

 

No, it uses confirmed numbers.

This sounds reasonable, after all this method doesn't promise to somehow tell us the actual percentages. It's just an average of the old and new methods, with the hope that the results we get are more in line with what we expect from the word "rarity".

 

Well, this old method didn't give us the results we expected from the word "rarity".... just saying! Most DLC trophies being rare or ultra rare made no sense, and whenever that wasn't the case it usually meant the DLC was free AKA when we actually counted correctly.... We're talking about people having many times the amount of DLC ultra rares than non-DLC ultra rares. It was completely broken.

Whenever people talk about counting people who own the game as owning the DLC, it just makes me want to say we should count all owners of the console as owning the game. Why? Try to make an argument that isn't special pleading and that doesn't apply in both cases!

 

I'm sure one could make arguments based on utility for counting DLC trophies differently though, but not really heard any good suggestions and arguments for those yet.

Edited October 20, 2016 by MMDE

[FawltyPowers]

It is accurate and meaningful MMDE in that when you install a game you install everything, main game and all DLC's (I don't agree with it but there you go). Game completion is based off all trophies. For example:

 

If I complete just the DLC trophy Sphinx' Riddle for the Catwoman DLC in Arkham City it would be:

 

15 / (1620 - 180) = 1.04%

 

So surely when calculating the number of people who earned it would be:

 

Number who earned the trophy / Owners of the game

 

Absolute accuracy would be as I mentioned earlier and that would be for Sony to have an install flag or date when you install DLC. This will never happen though.

[mekktor]

No, it uses confirmed numbers.

Well, this old method didn't give us the results we expected from the word "rarity".... just saying! Most DLC trophies being rare or ultra rare made no sense, and whenever that wasn't the case it usually meant the DLC was free AKA when we actually counted correctly.... We're talking about people having many times the amount of DLC ultra rares than non-DLC ultra rares. It was completely broken.

You could use "number of planets in the solar system". That is a confirmed number too. This number which you are saying is confirmed does not even belong in the equation - it's taking the place of the number of DLC owners. By doing that, the result is just as made up as this proposed method.

You're right that the old method doesn't give us meaningful results and I don't think it should be used, at least not without some sort of scaling of rarity categories. That's why we've been looking at other options.

Basically:

Old method - most technically correct option, results are correct by some definitions of rarity, dlc trophies too rare when compared to base game trophies

Current method - based on assumptions, results not correct by any definition of rarity, dlc trophies too common when compared to base game trophies

Proposed method - a compromise between the two, results not technically accurate but hopefully in line with definition of rarity

Edited October 20, 2016 by mekktor

[starcrunch061]

My absolute favorite thing about this thread however, is how positive and supportive everybody has been. Instead of getting all heated and devolving into an argument, people have been calmly discussing the various methods and working together. If it finally ends up changing the state for the better, that's just icing on the cake.

 

Amen to that. 

 

Like others, I agree that the word "rarity", as applied to DLC trophies right now, just doesn't make any sense. Something has to give. Even if, ultimately, the compromise only fits with Ambrose Bierce's definition, it still would be better than the current system, which makes zero sense.

 

However, although I love numbers, and think Mekktor has done some good work in checking up on data, I still like the idea of removing all rarity designations (UR, VR, etc.) from DLC whatsoever, and using the number of base owners as the computation for rarity percentage. 

[mekktor]

Whenever people talk about counting people who own the game as owning the DLC, it just makes me want to say we should count all owners of the console as owning the game. Why? Try to make an argument that isn't special pleading and that doesn't apply in both cases!

I'm sure one could make arguments based on utility for counting DLC trophies differently though, but not really heard any good suggestions and arguments for those yet.

I'm not sure what you're asking. Are you saying to provide an argument which supports using the old method? If so, I don't think we should use the old method on its own. I also don't think we should use the number of psn users, or the population of the planet as was discussed a few pages ago.

At least all of these have a clear meaning though which is in line with some interpretation of the word rarity, and they are all still better than using "how many people achieved the trophy out of how many people earned a trophy in that dlc pack", which is an absolutely meaningless ratio.

Edit: Oh I see. You're asking for an argument which supports using the number of base game owners for dlc calculations without also supporting using the number of psn users for all calculations. Why is it not allowed to support both? If you want to know how rare a trophy is amongst all gamers, that is a valid statistic. I guess my argument would be... By using the number of base game owners, we get unique information that can't already be found in the "achievers" stat.

However, although I love numbers, and think Mekktor has done some good work in checking up on data

Credit has to go to ProfBambam55 for this. Edited October 20, 2016 by mekktor

[MMDE]

You could use "number of planets in the solar system". That is a confirmed number too. This number which you are saying is confirmed does not even belong in the equation - it's taking the place of the number of DLC owners. By doing that, the result is just as made up as this proposed method.

You're right that the old method doesn't give us meaningful results and I don't think it should be used, at least not without some sort of scaling of rarity categories. That's why we've been looking at other options.

Basically:

Old method - most technically correct option, results are correct by some definitions of rarity, dlc trophies too rare when compared to base game trophies

Current method - based on assumptions, results not correct by any definition of rarity, dlc trophies too common when compared to base game trophies

Proposed method - a compromise between the two, results not technically accurate but hopefully in line with definition of rarity

 

Old method (the way PSN does it): based on imaginary number of owners of the DLCs, assuming everyone who owns the DLC owns the game. We know this is wrong. Earlier arguments for it has been that they can purchase the DLC, but just as well, everyone who owns the console can purchase the game too. In other words this is a terrible argument. Trying to get around this is just special pleading unless you can argue really well for it.

 

Because of the nature of statistics, you want to represent something because you just want to or it's usefulness, not necessarily because you want consistency or something like that. This is why I say you could make arguments for utility, a statistics usefulness and how wanted it is etc. I just haven't seen the good arguments for it or the good solutions.

 

The way it is, no numbers are made up, it's all based on confirmed data. A person who got a trophy for the DLC does own the DLC. If they don't own a trophy for the DLC, you don't know if they own the DLC or not.

[thegirlruka]

Old method (the way PSN does it): based on imaginary number of owners of the DLCs, assuming everyone who owns the DLC owns the game. We know this is wrong. Earlier arguments for it has been that they can purchase the DLC, but just as well, everyone who owns the console can purchase the game too. In other words this is a terrible argument. Trying to get around this is just special pleading unless you can argue really well for it.

 

Because of the nature of statistics, you want to represent something because you just want to or it's usefulness, not necessarily because you want consistency or something like that. This is why I say you could make arguments for utility, a statistics usefulness and how wanted it is etc. I just haven't seen the good arguments for it or the good solutions.

 

The way it is, no numbers are made up, it's all based on confirmed data. A person who got a trophy for the DLC does own the DLC. If they don't own a trophy for the DLC, you don't know if they own the DLC or not.

The only way that the current method of DLC calculation makes sense is if the method is also applied to the way that base game rarities are calculated. The way that base games are currently calculated is based on the number of users that have installed the trophy list regardless of whether they have earned a trophy in the game. When installing a trophy list, the dlc trophies are also installed. Both should require a single trophy to be counted or both should use the number of installed trophy lists as picked up by the site. The way it is now is inconsistent.

 

Alternatively, we could use the geometric mean. I'd be fine with any of these three methods but would prefer that it is based on whomever installs the trophy list (the old method.) I am not fine with the inconsistent way that rarities are calculated right now.

Edited October 21, 2016 by kuuhaku

[Se7en]

Most DLC trophies being rare or ultra rare made no sense

And what sense do they have to make except that the fewer people earned trophies from any particular DLC the more rare those trophies are? Please, not again with that difficulty thing.

 

Old method (the way PSN does it): based on imaginary number of owners of the DLCs, assuming everyone who owns the DLC owns the game. We know this is wrong.

Nope, there’s another assumption here, that every game player, aka person with the trophy list, can earn all these trophies, either from base game or from DLCs.

Actually ownership has nothing to do with that, as anyone can borrow or rent base game or GOTY disc with all DLCs and play it on weekend, for example. Those people don’t own neither base game nor DLCs, but as soon as they started game and it loaded trophy list on their PSN account, they are included in stats.

 

Earlier arguments for it has been that they can purchase the DLC, but just as well, everyone who owns the console can purchase the game too.

You are right, as soon as every console owner (or better say user) will automatically get trophy lists for all PlayStation games just by logging into his/her profile, trophy rarity has to be based on a total number of said console users. But for now rarity for every base game trophy is calculated based on a number of people who have trophy list from that game on their profile, while rarity for every DLC trophy is calculated based on a number of people who earned at least one trophy from each individual DLC. Which is inconsistency #1.

Moreover, DLC trophies aren’t totally separate, they are included in game trophy list and counted towards 100% completion of game. So if those DLC trophies are presumed to be integral part of a complete game, calculating their rarity should be based on the same number of people that used for rarity of other trophies from that game. Which currently is not and that’s inconsistency #2.

Edited October 21, 2016 by Se7en_Rus

[MMDE]

The only way that the current method of DLC calculation makes sense is if the method is also applied to the way that base game rarities are calculated. The way that base games are currently calculated is based on the number of users that have installed the trophy list regardless of whether they have earned a trophy in the game. When installing a trophy list, the dlc trophies are also installed. Both should require a single trophy to be counted or both should use the number of installed trophy lists as picked up by the site. The way it is now is inconsistent.

 

Alternatively, we could use the geometric mean. I'd be fine with any of these three methods but would prefer that it is based on whomever installs the trophy list (the old method.) I am not fine with the inconsistent way that rarities are calculated right now.

 

I can agree with you on that it should apply to base games too. I know the game doesn't count for your profile if you don't have any trophies for the game, but I think when the rarity is calculated it counts. If this is the case, this is an inconsistency that should be fixed!

And what sense do they have to make except that the fewer people earned trophies from any particular DLC the more rare those trophies are? Please, not again with that difficulty thing.

 

Nope, there’s another assumption here, that every game player, aka person with the trophy list, can earn all these trophies, either from base game or from DLCs.

Actually ownership has nothing to do with that, as anyone can borrow or rent base game or GOTY disc with all DLCs and play it on weekend, for example. Those people don’t own neither base game nor DLCs, but as soon as they started game and it loaded trophy list on their PSN account, they are included in stats.

 

You are right, as soon as every console owner (or better say user) will automatically get trophy lists for all PlayStation games just by logging into his/her profile, trophy rarity has to be based on a total number of said console users. But for now rarity for every base game trophy is calculated based on a number of people who have trophy list from that game on their profile, while rarity for every DLC trophy is calculated based on a number of people who earned at least one trophy from each individual DLC. Which is inconsistency #1.

Moreover, DLC trophies aren’t totally separate, they are included in game trophy list and counted towards 100% completion of game. So if those DLC trophies are presumed to be integral part of a complete game, calculating their rarity should be based on the same number of people that used for rarity of other trophies from that game. Which currently is not and that’s inconsistency #2.

 

1.

I haven't said anything about difficulty. It's more that the mark "ultra rare" and "rare" just didn't represent anything anymore. The trophies usually weren't "rare" among those who had the content they could get the trophy from, but few had the content they could get the trophy from. This is inconsistent with how the rarity was counted when it comes to base games. When it comes to the base game, it counts only confirmed data of who has the content they could get the trophy from. I could have the base game, just haven't booted it up. Should I count? I can purchase the base game, should I count?

 

2.

Yes, that's the problem. The DLC trophies are automatically added as related trophies to the base game when the base game trophy list is installed. This makes it hard to know who got the content that they can earn the trophies from, because this is what we essentially want.

 

(realized / potential) * 100

 

(earned the trophy / confirmed to have the content where you can earn the trophy) * 100

 

Knowing when someone actually got the DLC is the problem, but we know they do if they have earned one trophy from it.

 

Also, to be realistic, more of the people who actually get a DLC is likely to finish it and earn the trophies from it. We expect these trophies to have a lower rarity. Yet, with the old method (what PSN does), we get that most DLC trophies are Ultra Rare. Very misleading.

 

3.

Essentially, you don't want it to be based on who can potentially earn the trophies because they got the content they can earn the trophies from, but rather you just want to go by whatever PSN does to your profile when you boot it up. This site doesn't do that, because we want to see more interesting information. Also a game doesn't count on your profile until you've earned a trophy in it. You can delete games you haven't earned a trophy for.

 

Yes, people have argued for DLC trophies being "part of the game", but this is misleading if what we look for is who can potentially earn the trophies.

 

The question should be, what do we want it to represent? And then be consistent about it! If you just want the stats you seem to want, one may as well look at the PSN server numbers. I personally find those very misleading and useless when it comes to DLC trophies.

Edited October 21, 2016 by MMDE