Jump to content

Geometric Mean for DLC Trophies


Sly Ripper

Recommended Posts

56 minutes ago, voleman79 said:

And what about trophies that reguire 2 or more controllers or special controllers like Move, PS Eye or even second console (Vita with ad hoc trophies)? Or online trophies in general - you have to pay for internet connection and PS+. That is all hidden behind "paywall". Hey, these trophies should also be separated and rarity should be counted from confirmed people who have second controller etc.

Exactly, and this ideology is ridiculous. To quote a perfect post from the last time(?) this was discussed:

 

Quote

hey guys!

Did you know that 100% of people own an elephant?!

Its true!

I know this because i counted all the people who i saw own an elephant, and didnt count anyone else, and 100% of them own an elephant!

https://forum.psnprofiles.com/topic/24623-dlc-rarity-change/?page=9#comment-536264

 

 

I just don't understand how anybody could think that sort of tracking is okay or reasonable, especially when there are better alternatives like the very first method or this new one xD

Link to comment
Share on other sites

5 hours ago, MMDE said:

Just to iterate. You are making up numbers when you base them on something that you don't have supporting data for. If you say everyone who has played the game has played the DLC too, you're making up numbers, because you know that just because you've played the game, that doesn't mean you've played the DLC. Saying that the number of people who has played the DLC is somewhere between the amount of people who has earned a trophy for the DLC and who has for the main game is true, especially if inclusive between. But that doesn't mean it's in the middle, so again, you'd be making a somewhat estimated educated guess, in other words, making up numbers.

Buk isn't implying that all who have played the game have played the DLC. He's saying that all who have installed the trophy list for the base game have also installed the trophy list for the DLC because it is impossible not to do so. This is why the numbers are not made up.

Edited by kuuhaku
  • Like 2
Link to comment
Share on other sites

7 hours ago, kuuhaku said:

Buk isn't implying that all who have played the game have played the DLC. He's saying that all who have installed the trophy list for the base game have also installed the trophy list for the DLC because it is impossible not to do so. This is why the numbers are not made up.

 

Oh, I know that is what he and some other says, but that's not a very consistent view. Why? Well, what is it you're trying to express with the rarity? What is it supposed to represent? What kind of statistical meaning is it supposed to have? Is it supposed to say what percentage of people who got a trophy from those who has played the game? I'd say that is what it is. Don't confuse this with owners of the game, because I own tons of games and they won't count unless I've played the game. Hopefully you're still following me at this point. Next step is to realize that if we're going to be consistent with this idea, we can't count those who hasn't played that which can give them the trophy. If they haven't played the game/DLC, they shouldn't count.

 

The problem is, it is hard to tell who has played a DLC or not. Should we say everyone who has played the game has played the DLC too? We know that isn't true and it gives us very wrong and skewed statistics. We shouldn't count people we don't know if has played the DLC or not, so what we instead did was find a way to only count those we knew had played the DLC. In other words, only count those who had earned a trophy for that DLC. This means that we don't try to guess the amount or use numbers we have no basis for, but instead we use confirmed numbers.

 

There are serious inconsistencies in the way stats are counted on this site. Like how games on your profile that you haven't earned a trophy for doesn't affect your completion %, but you do count as an owner of said game. :S This is inconsistent. It should count on their profile IMO, because we know they've played the games.

 

What this new system does is just to guesstimate.

Edited by MMDE
Link to comment
Share on other sites

On 2/8/2017 at 11:37 PM, MMDE said:

A good example of how wrong this new system is all the PS+ games that has paid DLCs. A majority of those paid DLCs will be ultra rare or very rare now.

 

I was curious about this claim, so I looked up the trophies for a few PS+ games that I knew had DLC and I'm not seeing this.

 

Here's what I counted:

Driveclub -- (14 Very Rare, 55 Rare, 25 Uncommon, 1 Common)

Sound Shapes (North American List) -- (55 Common)

Never Alone (North American List) -- (4 Uncommon)

King's Quest -- (6 Very Rare, 24 Rare, 6 Uncommon)

 

Maybe there's some other games where this is happening and the first four that came to my mind aren't representative.  If so, I'd be interested in seeing the examples.

  • Like 1
Link to comment
Share on other sites

I thought I would have a look at my new Ultra Rare collection as it jumped from 62 to 83. Here is a list of all my Ultra DLC trophies and highlighted in yellow are the trophies that have become Ultra Rare since the change to Geometric:

 

SLEIA1F.jpg

 

Looking at the figures:

 

The original PSNP method (Sony) is all 0 - 1%. It's the reason it got changed.

 

The previous PSNP method (Old) shows all previous trophies as Rare at best.

 

The current PSNP method (Geometric) finds that middle ground through a formula and is pure guesswork, but looks most realistic.

 

This is how the majority of people will see it on here, you can argue about accuracy but no method is 100% accurate in determining who has played DLC.

 

On a personal note: Not a single trophy in the list above do I disagree with. The satisfaction of seeing Uncharted 3 DLC trophies as Ultra Rare including treasure collecting and the Crushing difficulty trophies of Overseer and Triumvirate on there is immense. Also Pa-Pa-Pa Poker Ace, damn that feels good too. The Birth of Oni is now Ultra Rare, requirements are more difficult than Long Time No See in the main game - also Ultra Rare. Ground Control In Black Ops solo took me a lot of time and patience. If any of those multiplayer trophies were in the main game I'm sure they would be Ultra Rare.

 

So even though I don't necessarily agree with the new stat because it's guesswork, it feels good seeing how DLC rarity has changed in a positive way and I'm sure most others feel the same. I think this stat is here to stay.

  • Like 1
Link to comment
Share on other sites

On 2/8/2017 at 10:27 AM, MMDE said:

AFAIK, one side wants owners of the game to count as owners of the DLCs too. I know I just want consistency and I don't want invented numbers. How does this reach a compromise between the two sides? It doesn't do anything to close in on either, but instead equal or further away from both sides from where it already was at.

 

Well, as Larry David said, a really good compromise is one that leaves both sides equally dissatisfied.

Link to comment
Share on other sites

3 hours ago, FawltyPowers said:

I thought I would have a look at my new Ultra Rare collection as it jumped from 62 to 83. Here is a list of all my Ultra DLC trophies and highlighted in yellow are the trophies that have become Ultra Rare since the change to Geometric:

 

SLEIA1F.jpg

 

Looking at the figures:

 

The original PSNP method (Sony) is all 0 - 1%. It's the reason it got changed.

 

The previous PSNP method (Old) shows all previous trophies as Rare at best.

 

The current PSNP method (Geometric) finds that middle ground through a formula and is pure guesswork, but looks most realistic.

 

This is how the majority of people will see it on here, you can argue about accuracy but no method is 100% accurate in determining who has played DLC.

 

On a personal note: Not a single trophy in the list above do I disagree with. The satisfaction of seeing Uncharted 3 DLC trophies as Ultra Rare including treasure collecting and the Crushing difficulty trophies of Overseer and Triumvirate on there is immense. Also Pa-Pa-Pa Poker Ace, damn that feels good too. The Birth of Oni is now Ultra Rare, requirements are more difficult than Long Time No See in the main game - also Ultra Rare. Ground Control In Black Ops solo took me a lot of time and patience. If any of those multiplayer trophies were in the main game I'm sure they would be Ultra Rare.

 

So even though I don't necessarily agree with the new stat because it's guesswork, it feels good seeing how DLC rarity has changed in a positive way and I'm sure most others feel the same. I think this stat is here to stay.

No, there were plenty of DLC trophies using the original method that were 10%+ (like L.A Noire) or even higher.

PSNP has a much smaller userbase than the one that Sony use for their stats, so we wouldn't have to worry about them all being 0-1%

Link to comment
Share on other sites

12 minutes ago, madbuk said:

No, there were plenty of DLC trophies using the original method that were 10%+ (like L.A Noire) or even higher.

PSNP has a much smaller userbase than the one that Sony use for their stats, so we wouldn't have to worry about them all being 0-1%

 

We had the old system, the most requested feature was to change it. It's not coming back, it was useless.

  • Like 3
Link to comment
Share on other sites

1 minute ago, Sly Ripper said:

 

We had the old system, the most requested feature was to change it. It's not coming back, it was useless.

Oh, I realize that. I prefer the original method but this new system is one I can live with - so much better than the one before it. I was just correcting that statement.

Link to comment
Share on other sites

18 minutes ago, madbuk said:

Oh, I realize that. I prefer the original method but this new system is one I can live with - so much better than the one before it. I was just correcting that statement.

 

Sorry I should have written it better, the trophies I listed came back as 0-1% and the issue on a wider scale had been that too many DLC trophies came back as Ultra Rare when using main game owners. I didn't mean every trophy for every DLC was 0-1%

 

Link to comment
Share on other sites

  • 2 weeks later...
On 2/9/2017 at 7:13 AM, madbuk said:

You may not believe it but there are plenty of people who buy DLC in preparation and maybe never get around to actually doing it. As such, the "DLC owners" number will always be completely made up and wrong. And, since you're so afraid of those nasty made up numbers, you should be against that, right?

 

This new method is MUCH closer to being accurate than the nonsense we had before lol.

I actually do this quite a lot. I have quite a few GOTY and Ultimate Editions that I haven't actually done the DLC to but have played through some or all of the base game. Other times it is a season pass or something on sale but I haven't actually bought the game yet.

Edited by skateak
Link to comment
Share on other sites

On 2/2/2017 at 7:26 PM, dalailama1989 said:

 

I hardly think it will :D

 

I think the Geometric Mean is awesome! :)

Yeah, why would it, when it's still just a random number with no basis in fact? If anything, it's now more random than either of the old methods.

 

As far as I can tell, the only people that cared about the old way (i.e. The only factually correct way) were people who didn't buy DLC and worried that their e-peens were shorter than people who did. ?

  • Like 3
Link to comment
Share on other sites

@Sly Ripper Are the rarity percentages calculated on page load, or are they saved in a database and updated every 24 hours or so? If it's the former, why not implement a user preference for how they want the percentage to be calculated? I'm guessing it's the latter and that's why this option has not been considered.

 

This has always been a divisive issue on this site and it seems like the only way to make everyone happy is to let everyone choose what they want to see. 

Link to comment
Share on other sites

1 minute ago, simula67 said:

@Sly Ripper Are the rarity percentages calculated on page load, or are they saved in a database and updated every 24 hours or so? If it's the former, why not implement a user preference for how they want the percentage to be calculated? I'm guessing it's the latter and that's why this option has not been considered.

 

This has always been a divisive issue on this site and it seems like the only way to make everyone happy is to let everyone choose what they want to see. 

 

They're saved

Link to comment
Share on other sites

Cool! It's not exactly a fully accurate representation, but that is unattainable anyway, and any other method will get flack either way. Kudos for coming up with this metric, it seems as fair as we could possibly make it with the information we have. 

I have to admit to feeling embarrassed for not knowing what the geometric mean was (I looked it up and understand now) since I've taken so many advanced statistics classes in uni (and still am, as I'm working towards being data analyst), but oh well, I get to say I learned a new thing because of gaming I guess haha :)

 

That said, perhaps a suggestion so as to satiate those that complain that this estimate is not actual reality (which is a point I sympathise with, I just realise there's no way to make it actually reality): While still using the Ultra-Rare --> Common nomenclature, perhaps add a line that explains that the numbers you're about to see have been estimated in stead of letting people who haven't read this thread assume that they're real. Like a disclaimer at the top of the DLC list, where the number of bronzes, silvers and golds in the DLC are listed.

For instance, in their movie rankings, Rottentomatoes.com adds a disclaimer above the list that the scores have been adjusted. 

Edited by QuentinCle95
Link to comment
Share on other sites

 

On 2/22/2017 at 2:12 PM, ONUOsFan said:

 

There is only one method that is based on verifiably accurate numbers, and that's the way it was done in the beginning. Anything else is guessing and pandering to people who care that someone else's DLC trophies are "artificially" inflating (or deflating, to be accurate) rarity numbers for other people and making them look bad somehow (I guess?).

 

This is flat wrong.  The current method is based on accurate numbers as was the previous method.

 

Unfortunately, as has been pointed out by many, it's not possible to tell exactly how many people actually own the DLC... both methods rely on calculating this to some extent (using actual numbers as the inputs).

 

The old way took the number of people who had achieved at least one trophy as the number of people who own the DLC.  This is not the same thing as the number of DLC owners (it will in every case underestimate it), but it can be assumed to be a simple shorthand for it and is simple to use as the baseline.

 

The new way takes the geometric mean of the previous number and the number of game owners.  This also isn't the same thing as the number of DLC owners (it's not very clear whether it will overestimate or underestimate more), but it can also be assumed to be a simple shorthand for it and is simple enough to use as the baseline.

 

You can prefer one way or the other, but it is an outright lie to claim that one is based on "verifiably accurate" numbers and the other isn't.  If you mean the actual number of DLC owners as the "real" number, then neither is, since we don't know that number.  If you mean that the inputs are real, then both are.

  • Like 2
Link to comment
Share on other sites

3 hours ago, dmland12 said:

 

 

This is flat wrong.  The current method is based on accurate numbers as was the previous method.

 

Unfortunately, as has been pointed out by many, it's not possible to tell exactly how many people actually own the DLC... both methods rely on calculating this to some extent (using actual numbers as the inputs).

 

The old way took the number of people who had achieved at least one trophy as the number of people who own the DLC.  This is not the same thing as the number of DLC owners (it will in every case underestimate it), but it can be assumed to be a simple shorthand for it and is simple to use as the baseline.

 

The new way takes the geometric mean of the previous number and the number of game owners.  This also isn't the same thing as the number of DLC owners (it's not very clear whether it will overestimate or underestimate more), but it can also be assumed to be a simple shorthand for it and is simple enough to use as the baseline.

 

You can prefer one way or the other, but it is an outright lie to claim that one is based on "verifiably accurate" numbers and the other isn't.  If you mean the actual number of DLC owners as the "real" number, then neither is, since we don't know that number.  If you mean that the inputs are real, then both are.

 

No, I meant to say that the total number of game owners is the only number we know for sure - the way it was done originally. Anything else is just guesswork, because there's no way to know how many people owned/attempted the DLC. 

Edited by ONUOsFan
  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...