Update platinum rarities

[Bezenko]

I know this has been an issue for ages. People hack trophies, rarities go up, hacker is removed and rarity stays up. I'm not sure if this applies to every game, but it's at least very noticeable in less common games and games with very few platinum achievers. I know at least once, years ago, this issue was addressed and all rarity-values were updated, if there is no permanent solution, could we at least get another update like that?

 

Here are a few cases where the rarity is completely off due to this issue:

Bandfuse: Rock Legends - 0.84%, should be 0.10%

https://psnprofiles.com/100-club/2276-bandfuse

MotoGP 10/11 - 0.50%, should be 0.38%

https://psnprofiles.com/100-club/850-motogp-1011

MotoGP 09/10 - 0.60%, shuld be 0.52%

https://psnprofiles.com/100-club/265-motogp-0910

 

Also, I don't know if this is related or another issue, but for some games (MotoGP 09/10 is one):

Leaderboard shows 38 achievers on first page, page 2 continues from 51. 100% club also shows 38 first achievers, but 50 last and 44 fastest.

[steel6burgh]

Rarity seems all messed up to me.  it seems as if only people who own dlc are being counted and I disagree with that.  Dlc is part of a game.  A game is incomplete without it and if you fail to buy the dlc and play it that is lacking on your part as a lot of people go out of their way to complete lengthy dlc.  In the time it takes a person to finish all their dlc they could have easily had 20 or 30 more platinums maybe 100 or 200 if they're playing ratilaka games.  

[NathanielJohn]

29 minutes ago, steel6burgh said:

Rarity seems all messed up to me.  it seems as if only people who own dlc are being counted and I disagree with that.

 

DLC rarities are computed based on the geometric mean of the rarities with and without including DLC owners. See this forum thread:

Supposedly this method was chosen because it produces more "realistic" rarities, even though it gives a number that measures nothing at all (neither the rarity including DLC owners nor the rarity excluding them).

Edited January 27, 2020 by NathanielJohn

[thefourfoldroot]

To be honest, it's pointless anyway until the site starts auto updating profiles that haven't signed in for a while. 

Just try updating someone's profile from the leaderboard and see how outdated all the stats are...

[Havok_Knight]

On 1/27/2020 at 2:21 PM, thefourfoldroot said:

To be honest, it's pointless anyway until the site starts auto updating profiles that haven't signed in for a while. 

Just try updating someone's profile from the leaderboard and see how outdated all the stats are...

It'd be nice to have all profiles updated on here, even if at least only once, and I mean all profiles cause that would give the most accurate stats possible. But with there being 60 million of them, that isn't possible, even if like 30% of them had no games or trophies it would still take 115,000 profile updates per day for a year to just get them all on here, once. Then there would have to be a few other things done, but it'd be cool to have that info.

[BlazikieronUS]

Payday infamy 25 isn’t ultra rare so I agree

[AvengedEvil]

1 hour ago, Havok_Knight said:

But with there being 60 million of them, that isn't possible, even if like 30% of them had no games or trophies it would still take 115,000 profile updates per day for a year to just get them all on here, once.

 

You know that this site only tracks 4.2 million profiles, don't you?

[visighost]

While I'm sure there's reasons why rarities are not accurate, I find it annoying when I start a game with a low platinum rate only to find out down the line it's actually unobtainable and that the score is legacy from cheaters. Would it be possible to add a "glitched" flag to games where completion is impossible due to glitches (and not, obviously, to server closure - another issue entirely)? It could also be used as a way to auto-remove cheaters. And if added as a visual cue somewhere at the top (akin to the region markers?), it would effectively warn people.

[Havok_Knight]

1 hour ago, AvengedEvil said:

 

You know that this site only tracks 4.2 million profiles, don't you?

You know that reading comprehension is a basic skill and that I specified in my first sentence that I wanted all profiles for the most accurate stats, which would include the ones not tracked on here. 

[AvengedEvil]

9 hours ago, Havok_Knight said:

You know that reading comprehension is a basic skill and that I specified in my first sentence that I wanted all profiles for the most accurate stats, which would include the ones not tracked on here. 

 

You know that you this site also shows PSN rarity, which, as far as I understood, would be exactly what you're looking for.

Edited January 30, 2020 by AvengedEvil
Typo

[Havok_Knight]

7 hours ago, AvengedEvil said:

 

You know that you this site also shows PSN rarity, which, as far as I understood, would be exactly what you're looking for.

I see that you still don't quite understand what I mean so I'll be more detailed. In order to get the most accurate stats possible, you would need to have all stats tracked, and then remove the profiles that never gained trophies (because playing a game once and getting the trophy list would inflate the rarity of the trophies) and removing the hacked/cheated profiles (which would also also change the rarities for obvious reasons). Now unless the PSN stats show what is specified (which they don't), then no, that sorting isn't what I am looking for, and the sorting on here isn't as accurate as it could be.

[DaivRules]

1 hour ago, Havok_Knight said:

Now unless the PSN stats show what is specified (which they don't), then no, that sorting isn't what I am looking for, and the sorting on here isn't as accurate as it could be.

 

In 2018, there were 80 million "PlayStation Network monthly active users", inevitably it's larger now.

 

 

 

Quote

removing the hacked/cheated profiles

 

I assume this would be based on our site rules for hacked/cheated profiles, since Sony takes no enforcement against hacked/cheated trophies. That job is going to take a whole lot of people a whole lot of effort and it'll never be complete. 

 

1. You indicated the time it would take to update all these and you're right. Sony would block that many profile update requests from any site like this fairly quickly.

2. You want to compare this site, full of people who know what trophies are and put in effort to earning trophies to every other* (*non-cheated, as you indicated) account out there, many of which are oblivious to trophies and those that do, generally don't care about trophies? What is gained about making that comparison? Most people want to compare themselves to people competing for the same thing they are.

Edited January 30, 2020 by DaivRules

[FawltyPowers]

You aren't going to be able to get 60 million accounts on here, you need PSN id's for that. The 4.2 million profile tracked on here is because those PSN id's have been manually added.

 

People are included in the stats even if they are at 0% in a game. If they weren't then the first trophy in Assassins Creed 2 would be at 100% as it's the first thing that happens in the game:

 

As it stands - The Birth of an Assassin has 647389 achievers out of 663615, a discrepancy of 16226.

[DaivRules]

33 minutes ago, FawltyPowers said:

People are included in the stats even if they are at 0% in a game. If they weren't then the first trophy in Assassins Creed 2 would be at 100% as it's the first thing that happens in the game:

 

As it stands - The Birth of an Assassin has 647389 achievers out of 663615, a discrepancy of 16226.

 

Ah yes, I was confusing peoples personal completion percentage statistic with the sites overall tracking. Thanks!

[Havok_Knight]

1 hour ago, DaivRules said:

 

In 2018, there were 80 million "PlayStation Network monthly active users", inevitably it's larger now.

 

 

0% games should not affect site stats at all. They're not counted for/against platinum percentage until they earn one trophy for said game, so those profiles don't really need to be removed.

 

 

I assume this would be based on our site rules for hacked/cheated profiles, since Sony takes no enforcement against hacked/cheated trophies. That job is going to take a whole lot of people a whole lot of effort and it'll never be complete. 

 

1. You indicated the time it would take to update all these and you're right. Sony would block that many profile update requests from any site like this fairly quickly.

2. You want to compare this site, full of people who know what trophies are and put in effort to earning trophies to every other* (*non-cheated, as you indicated) account out there, many of which are oblivious to trophies and those that do, generally don't care about trophies? What is gained about making that comparison? Most people want to compare themselves to people competing for the same thing they are.

To answer #2, this isn't about comparing trophy stats with other players, but about attaining accurate trophy rarities as the thread op indicated. Which anyone who does like and track trophies should want their rarities to be accurate. If removing cheated and 0% profiles made a trophy go from rare to ultra rate, what person wouldn't like to know their hard earned trophy was actually more rare than they thought?

[DaivRules]

21 minutes ago, Havok_Knight said:

attaining accurate trophy rarities

 

Okay, then at the sheer quantity of Playstation accounts and players, PSN stats aren't going to be that far off. Rounding errors, likely.

 

The OP is likely talking about the cheaters who are removed from Leaderboards still counting toward stats and the dormant accounts already on the site that had their updating paused during some site code maintenance. 

[Havok_Knight]

1 hour ago, DaivRules said:

 

Okay, then at the sheer quantity of Playstation accounts and players, PSN stats aren't going to be that far off. Rounding errors, likely.

 

The OP is likely talking about the cheaters who are removed from Leaderboards still counting toward stats and the dormant accounts already on the site that had their updating paused during some site code maintenance. 

Yes and no to this, if it was just flat numbers, rounding isn't too big of an issue, but when it comes to statistics and percentages, you can't really afford to round cause it could change too much, although if the rounding resulted in a 1% or less difference, it could still be acceptable.

 

Some examples of games percentage differences.

 

PS3

Infamous 833,274 tracked here

11.44% to 1.9% on PSN

God of War 3 718,666

16.11% to 2.6%

Mortal Kombat 9 459,367

0.62% to 0.1%

 

PS4

Monster Hunter World 245,771

5.59% to 1.2%

Horizon: Zero Dawn 489,482

32.12% to 6%

Marvel's Spiderman 269,838

49.09% to 8.9%

 

And these are just at a quick glance. Games that aren't popular won't have much of a change, but games that are very popular would have completely different trophy rarities.

[DaivRules]

25 minutes ago, Havok_Knight said:

although if the rounding resulted in a 1% or less difference

 

So is your proposal to get rid of the "PSNProfiles membership only" statistic and only have a singular statistic that has all of PSN?

 

I was saying that the PSN rarity, under the conditions you proposed (removing 0% and "cheaters"), that one wouldn't change. It would basically be what PSN is now so you can just substitute PSN's percentage and end up with what we have right now: the limited membership of this site compared to all PSN members at large. 

[cris3f]

21 hours ago, Havok_Knight said:

You know that reading comprehension is a basic skill and that I specified in my first sentence that I wanted all profiles for the most accurate stats, which would include the ones not tracked on here. 

 

You do even know what are the computational requirements and data bandwidth to fulfill that? And the privacy issues involved, not considering the people that choose not to share their trophy information?

 

All you have is what Sony provides and PSN Profiles provide each with their own flaws.

 

Sony could not care less about cheaters. Well they don't care about anything, surely not their customers, as long as they make money. Even on their side they should only be counting the number of times a trophy list is created, and consequently a trophy, and the number of times each trophy was synced. Nothing else, their percentage should be got from this. They have 0 interest on this, they will not spend a single cent trying to get any better accuracy.

 

As to here, even thought it's disappointing it's easy to understand that PSN Profiles can't afford the computational requirements and data bandwidth required to update the 4 million or so profiles here. Even more to do that regularly.

 

So we just have to accept things as they're. Far from perfect, but most probably what can be done.

 

 

[Walt the Dog]

On 1/27/2020 at 2:50 PM, Bezenko said:

I know this has been an issue for ages. People hack trophies, rarities go up, hacker is removed and rarity stays up. I'm not sure if this applies to every game, but it's at least very noticeable in less common games and games with very few platinum achievers. I know at least once, years ago, this issue was addressed and all rarity-values were updated, if there is no permanent solution, could we at least get another update like that?

 

Here are a few cases where the rarity is completely off due to this issue:

Bandfuse: Rock Legends - 0.84%, should be 0.10%

https://psnprofiles.com/100-club/2276-bandfuse

MotoGP 10/11 - 0.50%, should be 0.38%

https://psnprofiles.com/100-club/850-motogp-1011

MotoGP 09/10 - 0.60%, shuld be 0.52%

https://psnprofiles.com/100-club/265-motogp-0910

 

Also, I don't know if this is related or another issue, but for some games (MotoGP 09/10 is one):

Leaderboard shows 38 achievers on first page, page 2 continues from 51. 100% club also shows 38 first achievers, but 50 last and 44 fastest.

 

I do believe these two problems are related. My running theory is that the leaderboards have some sort of flag to hide certain accounts from view if they were found to have cheated by the site's standards, but while it'll change the rankings for that page accordingly, it doesn't drag in any records from the next page that should be able to fit, so you have weird instances like MotoGP 09/10 or MeiQ where you have pages that are clearly incomplete and which do not follow each other numerically.

 

Related to the first issue brought up, I believe that something like this is also what causes the mismatched rarities between actual achiever lists and the trophy's rarity. To an extent, someone else getting the plat fixes it, but it only gets rid of a couple achievers or so, not all of them. For example, when I got my Power Gig plat, the plat rarity went down from 0.98% to 0.94% since two of the hidden achievers stopped counting. Thing is that the site says there are 24 achievers, but there are only 13 actually listed in the count, so the actual rarity should be about half of what it is and were someone getting the plat an actual solution, the rarity would be a good deal lower than it is relatively speaking instead of damn near the same as before.

 

How this issue even happened, I have no idea. Short of looking at the source code for the site (which I believe only Sly can access), I cannot imagine how the database got screwed up this horribly that there's this level of inconsistency in the system. Would be nice to see it fixed, especially given that the cause of the issue was apparently fixed a little while ago, meaning there should be some concrete idea of what needs to be done to fix it, but I have little to no hope that it will be.

Edited January 30, 2020 by Walt the Dog

[Havok_Knight]

1 hour ago, DaivRules said:

 

So is your proposal to get rid of the "PSNProfiles membership only" statistic and only have a singular statistic that has all of PSN?

 

I was saying that the PSN rarity, under the conditions you proposed (removing 0% and "cheaters"), that one wouldn't change. It would basically be what PSN is now so you can just substitute PSN's percentage and end up with what we have right now: the limited membership of this site compared to all PSN members at large. 

It wouldn't be the same and I'll give a small example to show why.

 

PSN just tracks trophy earned vs profiles with the trophy list and they don't care how or when you got the list or trophies on them. So for example reasons, say there was a game with 100 tracked players and only 4 of them had the Platinum, that would give it 4% rarity (4/100), and that rarity would not change unless more people played it or got the Platinum.

 

Now if PSNP (this site) had access to those same stats, and found out that five of them were 0% percent profiles, then it would raise the percent from 4% to 4.2% (4/95). Then later found out one of those profiles was cheated and removed them, the percent would change from 4.2% to 3.1% (3/94). 

 

Statistically this site should have lower percentages when compared to PSN because we filter things out, and PSN doesn't. I wouldn't be surprised if PSN kept trophy data from profiles they ban themselves.

 

And none of this would be possible because of tech limitations like cris3f pointed out, it is all just potential statistics. The closest we could get would be to have an "average" percentage that took PSN and PSNP stats and found the median for them.

 

Tl;dr PSN would read it as 4%, PSNP would read it as 3.1% because we filter things out. So having accurate data to draw from would change things for everyone.

[Milktastrophe]

18 minutes ago, Havok_Knight said:

 

Now if PSNP (this site) had access to those same stats, and found out that five of them were 0% percent profiles, then it would raise the percent from 4% to 4.2% (4/95). 

 

Why should you filter out 0% profiles? How do you differentiate between someone who booted the game to add the list without actually playing vs someone who has played it but isn't good enough to earn a single trophy? If you don't include the latter then the statistic is not accurate because it's not counting everyone who has played the game.

 

An example of removing 0% profiles not working the way you're intending is Attacking Zegeta. If you only count profiles with at least one trophy then the Winner trophy will likely have 100% rarity (up from the 4% PSNP has and the 1% PSN has). This game does not hand out trophies for free so the current rarities don't mean only 4% or 1% of people have actually played the game, it means that only a few people that have played have actually beaten the first level and met the trophy requirements.

 

Filtering out 0% profiles can result in wildly inaccurate rarities 

[Havok_Knight]

6 minutes ago, Milktastrophe said:

 

Why should you filter out 0% profiles? How do you differentiate between someone who booted the game to add the list without actually playing vs someone who has played it but isn't good enough to earn a single trophy? If you don't include the latter then the statistic is not accurate because it's not counting everyone who has played the game.

 

An example of removing 0% profiles not working the way you're intending is Attacking Zegeta. If you only count profiles with at least one trophy then the Winner trophy will likely have 100% rarity (up from the 4% PSNP has and the 1% PSN has). This game does not hand out trophies for free so the current rarities don't mean only 4% or 1% of people have actually played the game, it means that only a few people that have played have actually beaten the first level and met the trophy requirements.

 

Filtering out 0% profiles can result in wildly inaccurate rarities 

You would then take into account the last time the game was played as that is something tracked on here. If someone hasn't earn one trophy on a 0% list in a month, good chance they aren't playing it.

 

Edit: Also keep in mind this site already filters out 0% lists for games, so it wouldn't change much.

Edited January 30, 2020 by Havok_Knight

[Milktastrophe]

8 minutes ago, Havok_Knight said:

You would then take into account the last time the game was played as that is something tracked on here. If someone hasn't earn one trophy on a 0% list in a month, good chance they aren't playing it.

 

Edit: Also keep in mind this site already filters out 0% lists for games, so it wouldn't change much.

The 0% games may be filtered out of personal statistics, but they're surely not filtered out for trophy rarities.

 

I don't agree that someone giving up on a game is reason to artificially inflate the rarity, making them more common.

[Havok_Knight]

5 minutes ago, Milktastrophe said:

The 0% games may be filtered out of personal statistics, but they're surely not filtered out for trophy rarities.

 

I don't agree that someone giving up on a game is reason to artificially inflate the rarity, making them more common.

If someone gave up on a game without getting any trophies, then it would be no different than someone never playing the game, what would you suggest if someone just put a lot of 0% lists on their profile just because they could and never intended to play them?