Trophy Rarity Question

[Doom]

Re-reading what Sly suggested I think I misunderstood what he meant.

He probably suggested to calculate the % using (Game Owners+DLC 'Owners')/2.

So it would be (DLC Trophy Achievers*100)/((Game Owners+DLC 'Owners')/2).

This would more or less double the rarity values of the old method, so half of the ultra rare would still be ultra rare (the one that were lower than 2.50).

In this way, I think, the % would still be too similar to the old method.

[DrBloodmoney]

Just remove the stat.

No one can understand that rarity doesnt mean difficulty.

Just take it away.

Job done - no more of these threads.

[Doom]

I don't care if people misunderstand what the rarity value means. It's not my problem.

I just want to remove all those 100% that look extremely dumb.

[FawltyPowers]

Re-reading what Sly suggested I think I misunderstood what he meant.

He probably suggested to calculate the % using (Game Owners+DLC 'Owners')/2.

So it would be (DLC Trophy Achievers*100)/((Game Owners+DLC 'Owners')/2).

This would more or less double the rarity values of the old method, so half of the ultra rare would still be ultra rare (the one that were lower than 2.50).

In this way, I think, the % would still be too similar to the old method.

 

Well with that here are the updated results for Uncharted 3 - Shades DLC

 

1. (7552 * 100) / ((667857 + 20851) / 2) = 2.19% (Ultra Rare)
2. (3744 * 100) / ((667857 + 20851) / 2) = 1.09% (Ultra Rare)
3. (6832 * 100) / ((667857 + 20851) / 2) = 1.98% (Ultra Rare)
4. (14666 * 100) / ((667857 + 20851) / 2) = 4.26% (Ultra Rare)
5. (7017 * 100) / ((667857 + 20851) / 2) = 2.04% (Ultra Rare)
6. (5255 * 100) / ((667857 + 20851) / 2) = 1.53% (Ultra Rare)
7. (5467 * 100) / ((667857 + 20851) / 2) = 1.59% (Ultra Rare)
8. (15793 * 100) / ((667857 + 20851) / 2) = 4.59% (Ultra Rare)
9. (7594 * 100) / ((667857 + 20851) / 2) = 2.21% (Ultra Rare)
10. (5321 * 100) / ((667857 + 20851) / 2) = 1.55% (Ultra Rare)

Definitely too close to the old method. Really like the previous suggestion.

 

As for dropping the stat, why would you do that? It's just a guideline to give you an idea of how challenging the trophy is, that's all. There's so many different factors involved in a trophies rarity:

• Difficult due to a high level of skill involved
• A lot of grinding required
• Luck
• Multiplayer trophy, easy to boost, next to impossible legit
• PS+ free game, people hiring games for a couple of nights, never playing again

I mean to be honest you can keep going and going with this list and because of this it will always be subjective.

 

DLC stats were previously weighted towards Ultra Rare, now they are weighted towards Common. A nice balance somewhere in the middle would be nice and the stat mentioned earlier on the surface looks good, I need to try it out on other DLC's.

 

Once again it's just a guideline to give you an idea of how challenging a trophy could potentially be.

[Doom]

Definitely too close to the old method. Really like the previous suggestion.

I know, but it was a fluke.

The value calculated by dividing the population (game owners) by the sample (DLC owners) probably doesn't mean anything, statistically speaking.

Edited November 6, 2015 by Doctor Doom

[fisty123]

I don't agree with dropping it altogether, many members aim  for the ultra rare trophies to make there gaming more fun, some ultra rare trophies are not hard but could take a very long time, or someone decided to get trophies on a really poor game it was never meant for ultra rare trophies to be hard trophies and if members do not like easy trophies being ultra rare then go and unlock them and boost the percent up

[Djeheuty]

Well with that here are the updated results for Uncharted 3 - Shades DLC

 

• (7552 * 100) / ((667857 + 20851) / 2) = 2.19% (Ultra Rare)
• (3744 * 100) / ((667857 + 20851) / 2) = 1.09% (Ultra Rare)
• (6832 * 100) / ((667857 + 20851) / 2) = 1.98% (Ultra Rare)
• (14666 * 100) / ((667857 + 20851) / 2) = 4.26% (Ultra Rare)
• (7017 * 100) / ((667857 + 20851) / 2) = 2.04% (Ultra Rare)
• (5255 * 100) / ((667857 + 20851) / 2) = 1.53% (Ultra Rare)
• (5467 * 100) / ((667857 + 20851) / 2) = 1.59% (Ultra Rare)
• (15793 * 100) / ((667857 + 20851) / 2) = 4.59% (Ultra Rare)
• (7594 * 100) / ((667857 + 20851) / 2) = 2.21% (Ultra Rare)
• (5321 * 100) / ((667857 + 20851) / 2) = 1.55% (Ultra Rare)

Definitely too close to the old method. Really like the previous suggestion.

 

I think you took the wrong game to make examples. I mean uncharted 3 dlc's really deserve ultra rare status due to how hard/time consuming/luck based/multiplayer only they are all. I'm sure that if they were in the main game they would have been all from very rare to ultra rare and seeing them all common or uncommon is misleading and not rewarding as they should be. Maybe try with another game with easier dlc trophies (Star Wars force unleashed comes in mind but there's plenty) and if it comes out they are all ultra rare as well then this won't be a good method.

[Sly Ripper]

Re-reading what Sly suggested I think I misunderstood what he meant.

He probably suggested to calculate the % using (Game Owners+DLC 'Owners')/2.

So it would be (DLC Trophy Achievers*100)/((Game Owners+DLC 'Owners')/2).

This would more or less double the rarity values of the old method, so half of the ultra rare would still be ultra rare (the one that were lower than 2.50).

In this way, I think, the % would still be too similar to the old method.

 

Not the average of owners, the average of each rarity.

[FawltyPowers]

I know, but it was a fluke.

The value calculated by dividing the population (game owners) by the sample (DLC owners) probably doesn't mean anything, statistically speaking.

 

Regrettably yes, it seems as though the DLC I picked at random was a fluke. The greater the number of DLC owners the more obvious it becomes.

[Doom]

EDIT: This was written before reading the latest post Sly made.

Don't consider the second column from the right of the image I posted.

---------------

I think another possible solution is to do the math using the "Average Completion" of the base game alone.

I can't propose any 'real life' math because I think that Average Completion also takes into account DLC trophies.

But the formula could be like this:

(DLC Trophy 1 Owners * 100)/(Total Game Owners/100)*Average Completion)= ... %

Obviously in this formula the Average Completion should be calculated without DLCs because you are trying to infer how many people could have the DLC and got no trophies in it.

Even though the Average Completion of the base game as a value to infer how many people could have the DLC is still a random guess.

Anyway, the values would end up being just slightly higher than what Sly suggested. So, still pretty low but somewhat better.

I tried it using the base game stats for Dead Space 1 plus the DLC stats of Dead Space 2.

I know it's stupid, but it was the only way I could use a 'clean' Average Completion (the game has roughly the same amount of owners... kinda).

The results, compared, would look like this...

I can't think of any other valid approximations, for now.

------------------

Not the average of owners, the average of each rarity.

Ah, you mean the average rarity of the main game trophy list?

How would the formula look like, if you don't mind?

Edited November 6, 2015 by Doctor Doom

[StrickenBiged]

If there's no way to track real DLC owner we should only use the official method used by Sony. That's the better way since they created trophies and they make the rules. Also IMO all the free DLCs like The Heist Update in GTA V should be counted with all the game owner since everyone has them.

 

A good point. But how would PSNP be able to distinguish between trophies attached to free-DLC and trophies attached to paid-DLC? That's an awful lot of flagging for an admin to do.

 

Yes I know and that's why we should blame Sony for that. I mean how the hell they aren't able to track people that buy DLC from the PSN Store? It's just stupid. IMO they know the real amount of people that have a game but they add DLC trophies to all users as they know it will sell a bit more 'cause of trophy hunters.

 

I honestly don't think Sony can do this without some major re-engineering with the way PSN works. From what I've read online and around abouts, the PSN overall is a Frankenstein's Monster of a thing. Whereas XBL was developed as an integrated service combining online MP, the store and achievements, Sony was caught out by digital stores and achievement systems and had to bolt these types of features onto what had originally been envisaged as simply a platform for online multiplayer gaming back in the PS2 days. From what I understand, PSN, the PSN store and the trophy system simply don't "talk to" one another in the way they'd need to for this to be possible. 

 

As I understand it, this is why you have to Sync your trophies with PSN, whereas XBL syncs your achievements immediately once they are earned. Your PS device has to log in to and access a completely separate service to update your trophies, which doesn't happen with XBL.

[Sly Ripper]

Ah, you mean the average rarity of the main game trophy list?

How would the formula look like, if you don't mind?

 

The average of the old and current method.

[LeetWolf2]

I feel sorry for Sly, he must be sick of seeing these threads all over the place. They always come to the same answer, that people wanted it this way so that's why it is this way. I personally don't mind either way as although individual rare trophies are great, I focus on completing DLC to get 100% overall completion for a game and the only thing I truly care about being rare is the platinum.

[Manuelito]

The average of the old and current method.

That's alot better. At least DLC trophies wouldn't look to rare or to common.

A good point. But how would PSNP be able to distinguish between trophies attached to free-DLC and trophies attached to paid-DLC? That's an awful lot of flagging for an admin to do.

Something like this is going to require a bit of time at the start since and admin gotta flag all the free DLCs but after that it would be easier since there's not alot of free DLCs coming out every week..

I personally don't mind either way as although individual rare trophies are great, I focus on completing DLC to get 100% overall completion for a game and the only thing I truly care about being rare is the platinum.

I used to care about my average rarity before I started playing DLCs. No it's only a mean less number for me.

[Doom]

The average of the old and current method.

Ah! Now I see.

Wouldn't that mean that you need a trophy to be under 5% in both old and current method for it to ever be UR like that?

Not that I particularly care about URs, but I'm trying to see things from both sides of the discussion.

[FawltyPowers]

So using an example:

 

Main Game Owners: 667857

DLC Owners: 20851

DLC Trophy Achievers: 7552

 

Old Method = 7552 / 667857 = 1.13%

Current Method = 7552 / 20851 = 36.22%

New Method = ((7552 / 667857) + (7552 / 20851)) / 2 = 18.67%

 

Definitely a middle ground

[iKiLLG0dZ-]

I quite like the new formula.

[starcrunch061]

Ah! Now I see.

Wouldn't that mean that you need a trophy to be under 5% in both old and current method for it to ever be UR like that?

Not that I particularly care about URs, but I'm trying to see things from both sides of the discussion.

 

Why?

 

If a trophy was 1% with the old method, and 9% with the current, wouldn't that be UR at that point (with an average of 4.5%)?

 

Incidentally, I still don't think the stat is that great. Every 100% trophy now can, at most, turn into 50%. I suppose that will make people who play a TON of DLC happy, but honestly, what percentage of a trophy hunter's trophies are DLC? 5% seems like a pretty liberal estimate. 

 

Still, if it shuts off these "debates", I suppose it's better than the current state of things.

Edited November 6, 2015 by starcrunch061

[Doom]

If a trophy was 1% with the old method, and 9% with the current, wouldn't that be UR at that point (with an average of 4.5%)?

Ah yes! That's right!

I suck with math

Anyway. This should be an example using Dishonored

Edited November 6, 2015 by Doctor Doom

[starcrunch061]

Oops - of course, the average would be 5%, but still, you get the point. I think I'm just assuming that 0% will average with the current percentage to give the new percentage for all trophies, but of course, that's unreasonable.

 

I've been sitting at work thinking about this compromise, and for the life of me, I don't understand what it tells. I'll grant that the extremes (i.e. the old method and the new method) are heavily flawed, but at least both methods had a semblance of purpose. The average just seems to say two wrong make a right.

[Doom]

It's an attempt to try and 'normalize' the 2 methods because they both dish out extreme % almost always.

[Djeheuty]

I've been sitting at work thinking about this compromise, and for the life of me, I don't understand what it tells. I'll grant that the extremes (i.e. the old method and the new method) are heavily flawed, but at least both methods had a semblance of purpose. The average just seems to say two wrong make a right.

IMO the average is the best compromise, both two extremes are actually unreal: in the first case we were assuming everyone who owns the game also owns the dlc, which is unreal; now we are assuming that only who earns a trophy in the dlc owns it, which is wrong, especially in case of dlc with few trophies. In fact the more trophies a dlc have the more accurate the owners number will be. We could add a multiplier to count this, so a dlc with just 1 trophy gets a *0.1 and a dlc with 10 trophies *1.0 (result won't be changed because owners number is rather accurate itself). Just an idea...

[Doom]

IMO the average is the best compromise, both two extremes are actually unreal: in the first case we were assuming everyone who owns the game also owns the dlc, which is unreal; now we are assuming that only who earns a trophy in the dlc owns it, which is wrong, especially in case of dlc with few trophies. In fact the more trophies a dlc have the more accurate the owners number will be. We could add a multiplier to count this, so a dlc with just 1 trophy gets a *0.1 and a dlc with 10 trophies *1.0 (result won't be changed because owners number is rather accurate itself). Just an idea...

But technically only if the trophies are unmissable you could have an accurate value of owners.

[Djeheuty]

But technically only if the trophies are unmissable you could have an accurate value of owners.

I know, I'm talking about probability. You can't have the exact value of owners, but chances are most players will unlock at least 1 trophy out of 10, while dlc's with 1 trophy are the extreme case of inaccuracy.

[Manuelito]

But technically only if the trophies are unmissable you could have an accurate value of owners.

That's still incorrect because if there's an unmissable trophy for completing mission 1 of a DLC and I don't complete it because the DLC is rubbish I still not count toward the owners. You can't have an accurate owner because of the stupid methods Sony use.