How is achievement percentage calculated?

[GC13]

Sorry for having to ask this: there appears to be a low signal:noise ratio when searching for this answer.

Steam keeps tracks of the global completion percentage for achievements, saying that, say, 67.3% of players have such-and-such an achievement. How is that calculated? Is it 67.3% of owners, people who have ever installed the game, or people who have ever started the game up?

[A Gelatinous Cube-Z-]

Presumably, it's calcuated by owners.

[Irregular Seconds]

It's what's on the tin: The percentage of people who own 'x' game and have completed achievement 'y'. I cannot put it in more simple terms than this.

[GC13]

In that case, I'm surprised that so many people who buy games have actually played them.

[Heartburn]

They rate them by people who own the game by people who have done the achievements. So if 2 people own a game but 1 finishes it. That means 50% have completed it.

[Irregular Seconds]

Originally posted by GC13:

In that case, I'm surprised that so many people who buy games have actually played them. :prisoner:

Actually, it's a of people who buy the game and do not actually play it, as much as it is people who buy the game and play it. This varies game-to-game quite a bit.

[GC13]

I can't catch your meaning there, Irregular.

Anyway, I don't think it's out of everyone who owns the game. Payday 2 just launched this morning, and 73.8% of its owners have played it already? I doubt that very much. It has to use a different number as a divisor.

[bvguthrie]

Originally posted by GC13:

I can't catch your meaning there, Irregular.

Anyway, I don't think it's out of everyone who owns the game. Payday 2 just launched this morning, and 73.8% of its owners have played it already? I doubt that very much. It has to use a different number as a divisor.

That doesn't surprise me. The sort of person who buys a game on launch day seems likely to want to play right away. If they didn't, why not wait for a sale?

[GC13]

Well, it surprises me. It sold a lot in pre-orders, and it was still relatively early in the morning in the USA when I checked.

[Guilherme (jgcasa)]

I guess the percentage of achievements a person has is related to the actions whitin the game (you can finish a game and has 42% of achievements about that game). I know it isn't the subject, but I find it, related to specific users very low the overal percentage we have relating to achievements.

Now, for specific achievements let's say, a rare achievement, it should be calculated considering the number of owners (the number of people with that achievement divided by the total of copies sold of that game).

About how fast some achievements are acquired, you have achievements just to start a game (so to speak, the easy ones)...you'll have higher percentage for that ones (It can be the case with Pay Day 2m this morning). THe rare ones takes a lot longer to get.

[Troik]

sorry for resurrecting this ancient thread, but it's basically the first result on google for "steam achievement percentage calculation"

came here looking for the exact answer to that. I suspect it's owners as well, but only counting people who had started the game ( a stat steam tracks anyway) would be much better. Sometimes you buy a bundle for a few games but the other games don't interest you, so you'll never play the game, simply duluting the percentages for all the people that actually play the game.

There are games on Steam that give you an achivement the second you start the game. that would be the definitive test for this, if 100% of players had that achiement, then it would only count actual players, but I suspect the percentage will be lower, proving that it's calculated by owners.

[jimjam]

Originally posted by Troik:

sorry for resurrecting this ancient thread, but it's basically the first result on google for "steam achievement percentage calculation"

came here looking for the exact answer to that. I suspect it's owners as well, but only counting people who had started the game ( a stat steam tracks anyway) would be much better. Sometimes you buy a bundle for a few games but the other games don't interest you, so you'll never play the game, simply duluting the percentages for all the people that actually play the game.

There are games on Steam that give you an achivement the second you start the game. that would be the definitive test for this, if 100% of players had that achiement, then it would only count actual players, but I suspect the percentage will be lower, proving that it's calculated by owners. :smartsam:

I came here for the same reason, but I don't think it is calculated by owners. Reason being, I can get achievements for games that I don't own when I play them using family sharing. So it must be calculated in some other way?

[cSg|mc-Hotsauce]

Originally posted by jimjam:

I came here for the same reason, but I don't think it is calculated by owners. Reason being, I can get achievements for games that I don't own when I play them using family sharing. So it must be calculated in some other way?

It is tracked by actual ownership. Sharing does not impact global stats.

[Iceira]

Im not sure how you can not understand it, steam have all stats, bought item and who trigger them, i did not say they play the game , some use SAM.
if i was you i will not trust 2 cent on steam user did it legit.

do note even game devs update game and screw them up. so things once worked, dont work anymore is also possible.

why do you not get the math part nr of achive per game / owner of game / and who trigger them.

[Iceira]

Originally posted by GC13:

In that case, I'm surprised that so many people who buy games have actually played them. :prisoner:

i only have old info, from consol info years ago, back then avarage gamer spent under 8hour on a game, that why they included cheat menu, problem is also so many games today that user switch or get a new or dump the other game , if user dont feel like its worth doing or to hard.

this has result in games thats dump down, so game devs might get better score and more buy them. ( this just backfire then user figure this out ) but again why diffculty level matters.

point is game devs try to cover 8hour as minimum ( but not a decad later or more , i think a game devs should be lucky if a user spent more then 2hours in them. this is not a joke, you barely scrates the surface, but i stick to old 8hour in a game then you have a good idea whats its about.

im sure user can do it with 3-4 hours , but then you have experience from other games, and use previous experience in what to do.

and again why 2 hours is in some (elite players that simple say okay fine game not what i looking for next game )

this post sounds odd
but i bet you can see truth in it.
you can also see then they do find a game they use way more the 100 of hours in them or 1000.


steam also do a mistake here, then forget ppl can have owned the disk version, before steam digital version, so hours spent on game is actual not correct.
sure not steam problem but stats is misleading.

take Deadlock2 shrine wars , im sure i have way more hours in it from the past ( we could use a option box as own as disk cd/dvd also ( i think many other buy a digital version , and why hours in a game cant be count as pleayed time. )

[Judgmental Amaterasu]

Originally posted by GC13:

Sorry for having to ask this: there appears to be a low signal:noise ratio when searching for this answer.

Steam keeps tracks of the global completion percentage for achievements, saying that, say, 67.3% of players have such-and-such an achievement. How is that calculated? Is it 67.3% of owners, people who have ever installed the game, or people who have ever started the game up?

I mean without randomly staring at the data it is likely number of people that have played w/o the achievement / the number of poeple that have played the gave with the achievement. Free weekends and the like included.