Latest Diablo 3 News
DiabloWiki Updates
Page 3 of 5 FirstFirst 12345 LastLast
Results 21 to 30 of 43
  1. #21
    IncGamers Member
    Join Date
    Sep 2012
    BattleTag Fizoo-1366
    Posts
    313

    Re: Jay Wilson says AH was a mistake

    It would be pretty awesome if Mr. Wilson came out tomorrow and said "April Fools! We don't think it was a mistake at all!"

  2. #22
    D3 Amazon &
    Demon Hunter Moderator
    Dacar92's Avatar
    Join Date
    Nov 2003
    Location
    Chicago
    BattleTag Dacar92-1368
    Posts
    10,079

    Re: Jay Wilson says AH was a mistake

    Quote Originally Posted by Ivan E View Post
    Actually, that's a pretty stellar retention rate.

    Retention rate past first launch (not game launch, as in user buys/installs/launches game from their desktop) is about 50%. Only 10% of users even get to the end of a game they purchased. So 1/15 (your highest number, leading to the worst case) = ~7% retention, which is absolutely stellar considering the game is 11 months old. Sure, you could say that the scale we should use to measure D3 is different because it's a not a linear, one-story-playthrough game (most games are, and hence the retention stats I noted above reflect them). But most people I know actually bought D3, finished normal, and stopped playing there, and I'm guessing that most of the 10 million / 12 million / 15 million / whatever purchasers are actually more like my aforementioned friends, rather than the relatively hardcore that would care about the game's failures. So I think a retention rate of well over 5%, 11 months after launch, is a rousing success, considering that literally half of all game purchases are played for about 2 hours and then never touched again.
    Got any sources for those stats? Besides your anecdotal evidence? Believe me, I do put some value in anecdotal evidence. But your first couple sentences are stated as fact. I don't disbelieve you but I'd like to know where those data came from.
    Rose: You're an alien? You sound like you're from the North.
    Doctor: LOTS of planets have a North!!!

  3. #23
    IncGamers Member
    Join Date
    Sep 2011
    Posts
    1,218

    Re: Jay Wilson says AH was a mistake

    Quote Originally Posted by Dacar92 View Post
    Got any sources for those stats? Besides your anecdotal evidence? Believe me, I do put some value in anecdotal evidence. But your first couple sentences are stated as fact. I don't disbelieve you but I'd like to know where those data came from.
    Sure thing, if you'd allow me the opportunity for edification.

    The 10% is actually somewhat of a misnomer if you take its best-known source -- a report from Raptr and a freelance article that ran on CNN (hardly an authority on videogames), found here: http://www.cnn.com/2011/TECH/gaming....now/index.html

    In that article, the author extrapolates a 10% completion rate for Red Dead Redemption into "10% of users finish videogames," which is kind of silly, because obviously RDR is just one game.

    What's less silly is this very interesting (and pretty recent) "study" from Gamefront, available here: http://www.gamefront.com/why-dont-gamers-finish-games/

    As you'll see, ostensibly completion rates are not awful -- they're around 30% for many games. I think this article presents a great starting point, but want to make a few caveats to their data. Specifically, they themselves trace a path implying that shorter campaigns lead to greater completion rates, which makes sense. But their list excludes almost every game that's not a one-and-done game (Civ V is the obvious exception). And Civ 5's completion rate, which means finishing one game of Civ 5, ever, is about 27%. Which means that over 70% of Civ 5 purchasers didn't even finish one game, when the game is expected to be played for thousands of hours. (Of course, this says nothing about retention over time.)

    With that said, I have a few ideal edits to make to the Gamefront article. Inserting non-linear games, or games meant to be replayed repeatedly, into their data would probably torque those completion numbers down, for all the reasons we're well aware of (market saturation, gamers having less free time, emphasis on competition and multiplayer rather than games that you "complete", etc.). But, more importantly, inserting worse games -- not top sellers, not things that most people are very excited about -- is also going to bring those numbers down. If you look at the Steam achievement stats for games fewer people care about, or games that are easier to buy on a lark, the completion stats are much lower. Indie darling Legend of Grimrock has an 8.9% completion rate well after release (http://steamcommunity.com/stats/Lege.../achievements/).

    My other statement, about the 50% never launch again thing, came from a series of video interviews with Casey Hudson (of Bioware/Mass Effect fame). Unfortunately I can't find these, so you can disregard these as evidence. But I found this page pretty interesting: http://steamcommunity.com/stats/CivV/achievements/ -- all of the achievements that are well above 50% are those easily doable on the first playthrough -- the rate craters after, despite the fact that many of those other achievements are gimmes, and you just need to play "far" enough in a game to get them. City of Science is more or a less a gimme if you've played a game long enough, and it's got a rate of 55%. So maybe the re-launch rate is not quite 50%, but I'm guessing it's somewhere in there.

    Anyway, I wouldn't put much/any value in anecdotal evidence myself, so I wanted to provide some contextual information. Hope that helps.

  4. #24
    IncGamers Member
    Join Date
    Jun 2003
    Posts
    409

    Re: Jay Wilson says AH was a mistake

    I see the 1 million unique users in one day, 3 million unique users over the course of a month as being terrible. These are big statistical numbers thrown out by Blizzard designed to make our eyes glaze over and make us think things are great and going well. As with most statistical presentations, what is touted is designed to hide what they don't want you to see. So what is Blizzard hiding.

    First of all, the numbers are for the game as a whole. How many realms are there: USA, Europe, Asia. Let's just use these three. For the sake of argument, let's also assume an equal distribution of players across these 3 realms. Already, you can take the 1 million and 3 million and divide them by 3 for your chosen realm: 333,333 unique players per day; 1 million unique players per month.

    Second, now that we have the numbers in terms of players per realm, 333,333 unique players per day is a misleading number. This number has repeat logins over the course of the month. A better unique players per day number is represented by 1 million unique players per month divided by 30, which is 33,333. This 33,333 average number doesn't contain any repeat/same-player logins over the course of a month. And putting this all together, what we have on a daily basis for each realm is, 300,000 players playing on and off over the course of a month, with 33,333 players logging in once never to login again over the course of a month.

    There is a reason Blizzard once again removed the ability to see player counts. The population is about 300,000 active players per realm, with zero information on how many hours per day any of these 300,000 are playing. Anyone want to guess what the peak concurrent connections per realm is when the population is 300,000 active players? Not good, which is why Blizzard removed the ability to check.

  5. #25
    D3 Amazon &
    Demon Hunter Moderator
    Dacar92's Avatar
    Join Date
    Nov 2003
    Location
    Chicago
    BattleTag Dacar92-1368
    Posts
    10,079

    Re: Jay Wilson says AH was a mistake

    Quote Originally Posted by Ivan E View Post
    Sure thing, if you'd allow me the opportunity for edification.

    ....clipped....

    Anyway, I wouldn't put much/any value in anecdotal evidence myself, so I wanted to provide some contextual information. Hope that helps.
    Interesting. I submit that Diablo 3 is different and since there is no hard evidence beyond which Blizzard releases themselves I think their numbers should be looked upon with a bit of skepticism. When is Diablo 3 considered "finished"? When you kill Hell Diablo? When you get to level 60? When you get to Paragon 100? Let's say I play 3 characters through to level 60 and kill D. Is that three finishes?

    For that matter, if I begin 4 games in Civ 5 and play through to the end and either win or lose each of those games, does that count as 4 finishes?

    I don't know. I feel like the numbers Blizzard releases are a little skewed and made to look a little better than they are.


    Quote Originally Posted by Galtrovan View Post
    I see the 1 million unique users in one day, 3 million unique users over the course of a month as being terrible. These are big statistical numbers thrown out by Blizzard designed to make our eyes glaze over and make us think things are great and going well. As with most statistical presentations, what is touted is designed to hide what they don't want you to see. So what is Blizzard hiding.

    First of all, the numbers are for the game as a whole. How many realms are there: USA, Europe, Asia. Let's just use these three. For the sake of argument, let's also assume an equal distribution of players across these 3 realms. Already, you can take the 1 million and 3 million and divide them by 3 for your chosen realm: 333,333 unique players per day; 1 million unique players per month.

    Second, now that we have the numbers in terms of players per realm, 333,333 unique players per day is a misleading number. This number has repeat logins over the course of the month. A better unique players per day number is represented by 1 million unique players per month divided by 30, which is 33,333. This 33,333 average number doesn't contain any repeat/same-player logins over the course of a month. And putting this all together, what we have on a daily basis for each realm is, 300,000 players playing on and off over the course of a month, with 33,333 players logging in once never to login again over the course of a month.

    There is a reason Blizzard once again removed the ability to see player counts. The population is about 300,000 active players per realm, with zero information on how many hours per day any of these 300,000 are playing. Anyone want to guess what the peak concurrent connections per realm is when the population is 300,000 active players? Not good, which is why Blizzard removed the ability to check.

    I can go along with some of your math, but let's focus on one segment. 333,333 unique players per day on each realm means 333,000 different players, not repeat log ins. Otherwise it wouldn't be a unique log in. Unique means 333,333 players logging in from different accounts per day. That would not count repeat log ins. Still, 1 million unique players per day is not very exciting for the Blizzard team and I would guess they are seeing those numbers slowly decline over the months.
    Rose: You're an alien? You sound like you're from the North.
    Doctor: LOTS of planets have a North!!!

  6. #26
    IncGamers Member
    Join Date
    Sep 2011
    Posts
    1,218

    Re: Jay Wilson says AH was a mistake

    @ Dacar92

    To answer your questions -- Blizzard is using unique logins on a daily/monthly basis in JW's 1M/3M (respectively) figures. Obviously, this is not directly comparable to completion rates. However, my point was simply that, if we take even a generous completion rate of 40% for a given AAA title, the point is that you spend X hours (8, 10, 20, 40 for an RPG) on it, and you're done. Since completion is binary, you can more or less find "time spent" by just multiplying campaign length by the percent to have completed the game. This excludes the stragglers that haven't finished the game yet but are working on it, but includes the majority of players at a given point in time after release (when most people that wanted to beat the game already have). In any case, since we know the completion rate is pretty low for games in general, we can apply that to compare with D3. JW/Blizzard haven't said much about who "completes" D3 and how often, but if 7% of players are still logging in 11 months after release, that probably compares pretty favorably to games where 40% of players beat them in 20 gameplay hours after release, and never touch them again, at the very utmost. That's all I was trying to say.

    As for your Civ 5 question, that statistic was calculated by just looking at the achievement you get when winning a game -- 27% or whatever of all Steam accounts having purchased Civ V have that achievement, which is pretty dismal. (Though Civ 5 is a phenomenal seller for Firaxis/2k, which is kind of weird given that fact.)

    I'm sure Blizzard's spin is on their numbers, but the entire crux of my argument is that even a 7% retention rate after 11 months is pretty great, considering most games have a 40% retention rate even before players have completed them, and probably somewhere like 1% thereafter for the truly obsessive. Granted, most games are not Diablo-likes that are intended to have tons of replay value, but from a "value" perspective, D3's retention rate is pretty impressive, given that most games retain very little of their userbases over this long of a period.

  7. #27
    IncGamers Member Azzure's Avatar
    Join Date
    Aug 2011
    BattleTag Azzure-1989
    Posts
    1,356

    Re: Jay Wilson says AH was a mistake

    At least 100,000-200,000 of those are farmers. But still, a very respectable figure.

  8. #28
    IncGamers Member
    Join Date
    Feb 2006
    Posts
    5,373

    Re: Jay Wilson says AH was a mistake

    http://steamcommunity.com/stats/Torc...I/achievements

    80% of players* didn't even finish it despite the "this is wut Diablo 3 shuda been" hype. People have absolutely absurd expectations about player retention when it comes to games.

    I'd love to see more detailed stats released from all over the place, though. Diablo, Borderlands, Path of Exile, etc. Trying to gleam player retention from press releases and steam player concurrency rates + basic progression rates gets a bit old.

    *Don't actually know if steam statistics can accurately represent the entirety of the whole Torchlight 2 playerbase.

  9. #29
    Europe Trade Moderator krischan's Avatar
    Join Date
    Aug 2003
    Location
    Hamburg, Germany
    Posts
    33,195

    Re: Jay Wilson says AH was a mistake

    Quote Originally Posted by Ivan E View Post
    As you'll see, ostensibly completion rates are not awful -- they're around 30% for many games.
    I think the term completion rate depends a lot on what people regard as completion. If it's finishing inferno, then the completion rate for D3 is low. I never achieved it, for example. However, Blizzard wrote themselves that inferno was designed as being almost impossibly tough, so things were as Blizzard wrote when regarding the average player (not to be confused with those who still participate in D3 forums ). So I don't think it's a good definition of completion.

    You may also regard the game as completed upon defeat of Diablo in normal mode and I think that's not a worse definition than the first one. The rate is probably 80-90% then.

    Obviously, such an ambiguous term is no reliable means to make a judgement about the game, just like it doesn't make sense to make a judgement on a dart simulator after you are done with a game of darts in perhaps 15 minutes. You will probably play hundreds of dart games with it.

    Beyond that, a wary person might speculate that the term could have been defined in a certain manner in order to produce a certain result. The conclusion is, if there is no precise way to define it, but only ones which are subject to interpretation and opinion, it's not useful for an objective evaluation, just a meaningless, volatile number which can be produced in any way need. You may use it for a personal impression on the game, however.
    D3 Trading Forums: Europe - America
    Diablo Wiki / Arreat Summit / ATMA / Forum Rules / Adria
    You know I'm born to lose / and gambling is for fools / but that's the way I like it, baby / I don't want to live forever!

  10. #30
    IncGamers Member
    Join Date
    Sep 2011
    Posts
    1,218

    Re: Jay Wilson says AH was a mistake

    Quote Originally Posted by krischan View Post
    Beyond that, a wary person might speculate that the term could have been defined in a certain manner in order to produce a certain result. The conclusion is, if there is no precise way to define it, but only ones which are subject to interpretation and opinion, it's not useful for an objective evaluation, just a meaningless, volatile number which can be produced in any way need. You may use it for a personal impression on the game, however.
    Good points! However, I want to offer the following nuance:

    1. The completion rates from the Gamefront article are all very clearly defined as "beating the game," and the included games are largely narrative-driven and have a "story mode" that you can beat. Civ 5 is the exception, so it's really weird that it's on the list. But almost every recent AAA game has a Steam achievement for "beating" it.

    2. For Diablo, I was discussing "retention rate" and not "completion rate." My argument is solely that Diablo's retention rate seems pretty high, relative to other games, because other games have low completion rates. If most of your purchasers don't complete your game, your retention rate must be paltry -- which makes Diablo's 7%ish retention rate after 11 months pretty impressive to me.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •