Aerothorn on 8/11/2010 at 20:08
So glad you posted this, Koki - I've been looking for some sort of rough evidence that most people don't finish a game the first time (and, by extension, only a small minority finish a given game multiple times), and this seems to be the best thing around so far.
Koki on 8/11/2010 at 20:12
Quote Posted by ZylonBane
Do you also get about 45 minutes into movies, loudly declare "BORED NOW!", and walk out?
Hey, I did that on Lost in Translation
Eldron on 8/11/2010 at 20:24
Quote Posted by ZylonBane
Mandatory Achievements are just about the stupidest thing ever. Thank you, everyone's-a-winner culture.
Do you also get about 45 minutes into movies, loudly declare "BORED NOW!", and walk out?
No, but I often pick up movies on sales that I don't instantly watch.
ZylonBane on 8/11/2010 at 20:36
Quote Posted by Koki
Hey, I did that on Lost in Translation
That's because you are dumb.
Awesome movie.
faetal on 9/11/2010 at 10:42
These stats are meaningless unless they include information on what percentage of the figures include people who have run the game and also how recently they last did.
If someone has not run the game, then they shouldn't be part of the sample population. If someone last ran the game less than (e.g.) a month ago, then they shouldn't be part of the sample population. To say "well yeah but it tells us something" is to ignore the entire principle of statistics and population analysis.
What you could potentially do is to take games which were released within a certain time of each other and take directly comparable parameters, such as "completing the game" and compare them to each other, as this *may* tell you which of the two people are less likely to finish. But all of these untethered "well what do you know, very few people finish x game" observation are like reading tea leaves.
Koki on 9/11/2010 at 11:46
Quote Posted by faetal
To say "well yeah but it tells us something" is to ignore the entire principle of statistics and population analysis.
Oh the irony.
Have you considered that it might tell us something else than "people don't finish games"?
Eldron on 9/11/2010 at 11:52
They don't, but that doesn't mean they won't.
faetal on 9/11/2010 at 12:05
I think what I'm saying is that it tells us only about what *has* happened and provides little information on what hasn't. I certainly take your point that it's interesting, but bearing in mind the hidden variables, it doesn't feel like there's enough information about what the data are to be able to make any conclusions. I f someone is still playing a game, or hasn't played it at all yet, then their real behaviour and approach to the game is yet to be logged. Likewise, as mentioned earlier, if there is a sale on, then an influx of purchases floods the population with people who haven't played yet, or are still in mid-flow.
Would be interesting if you could specify a basement threshold for hours spent in game and *then* look at the stats.
nicked on 9/11/2010 at 13:20
With a lot of games you can exclude those who haven't played the game (or almost all) if the game has a "first base" achievement, like "Killed your first enemy", "Completed the tutorial", stuff like that. By removing the percentage that haven't completed these first steps, you can get a clearer picture of the stats of those who have played it. Still doesn't account for people in the middle of a playthrough though.
