Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Seems like the game has peaked on XFire

17810121369

Comments

  • Vato26Vato26 Member Posts: 3,930

    Originally posted by Robokapp

    Originally posted by Icewhite


    Originally posted by Robokapp

    claiming xfire is not representative as a sample is the same as claiming there's an xfire conspiracy against swtor. you have no evidence that xfire users got together and decided together to sabotage swtor.

    First sentence is a False Dichotomy, by the way.

    It isn't a claim, it's simply a basic rule of good statistics and polling practices.  No motives required.  This is Stats 110 stuff, available in any junior college stats class or a million places on the net.

    "4. How were those people chosen?

    The key reason that some polls reflect public opinion accurately and other polls are unscientific junk is how people were chosen to be interviewed. In scientific polls, the pollster uses a specific statistical method for picking respondents. In unscientific polls, the person picks himself to participate." http://www.ncpp.org/?q=node/4

    Why is it important?  People making vague claims from bad statistical data is where that phrase about "liars use statistics" was born.

     

    m'kay. and how were these people chosen?

     

    since xfire is older than swtor, they probably werent chosen based on swtor. Soooo...the sample space is therefore random.

     

    because the people werent chosen.

     

    your argument while legit, doesn't apply. it's assumed both xfire and swtor players are part of the 'computer players' category. not sure how more scientifically you can pick them.

    WARNING!

    Logic failure detected.

     

    "Not chosen" =! "random".

  • TyvolusTyvolus Member Posts: 190


    Originally posted by GrayGhost79

    Originally posted by Tyvolus
     


    Originally posted by GrayGhost79



    Originally posted by Distaste




    Originally posted by GrayGhost79

    Yeah, Xfire isn't a great measuring stick at all lol. Gamers do not equal MMO gamers first off so regardless of how many Xfire users there are it's moot to bring it up. Secondly it's used less and less by those playing online games as many anti cheat programs have conflicted with Xfire and it is fairly pointless considering whats needed out of what it offers is done better by so many other people. 
     
    I mean if we are to use Xfire WoW has 20k people playing a day and it's boasting 10mil subs, does that mean with 10k players a day SWTOR has 5mil subs? 
    Haters and fans of games need to stop using Xfire as a measuring stick. 
     
    Myth
    Xfire is good because it has a good sample size -
        Wrong. It's a mix of gamer types and most of which are not MMO players. Xfire has had enough issues in the past that most online gamers do not use it or at least turn it off when they play online. 
     
    Myth 
    Xfire can give you at least an idea of how popular an MMO is - 
       Wrong. It's so random who uses it and several games with large communities don't even have a Xfire pressence due to problems in the past and due to some that still exist. And again...... it's a bit random on who uses Xfire. And we won't even get into the discussion on whether devs/gms/mods/fans use Xfire to try and boost hype. 
     
     
    Bottom line, for the most part Xfire is fairly useless the good things that it offers are done better by so many others. Xfire is not an accurate measuring stick to be used for or against a game. If you use Xfire, good for you I guess but just keep in mind most gamers do not especially online gamers and those that do do not have it on all the time. 



    All I can say is that it was accurate at showing population trends with AoC, WAR, AION, and RIFT. Each time someone would say "but it's not accurate", but time and time again it was. I'm not saying that it is infallible, but it does show a trend and those trends have been accurate in the past.




    Actually it wasn't accurate. Population was rising when Xfire showed it trending down and did the opposite when it population was declining. 
    It's only accurate in showing trends from Xfire players. If you're concerned with what people on Xfire play it's an excellent tool. However thats all it shows and it's rarely been even remotely accurate outside of Xfire users. And yes that includes its accuracey showing trends. 
     
    Now I'm not saying Xfire has been wrong all the time, even a broken watch is right twice a day. 



     
    according to your "logic" it is quite possible SWTOR is not a top 5 xfire game, but could be the #1 MMO in the world then ? strange how wow was perched at #1 on xfire for a very long time and strangely enough was the #1 MMO in the world. I hope you are able to understand if swtor proves itself to be the top MMO in the world 1 year from now, it will be reflected accordingly by xfire #s. wow proved these will not be mutually exclusive, proving the xfire sampling is, to a degree....accurate enough.


    lol......... yes when you hit the 5-10 mil plus it is more than likely that Xfire will be able to accurately show the game has a lot of players. 
     
    Like I said even a broken watch is right twice a day. Congratz on pointing out that even Xfire can see the moon without a telescope!
     
    Though I don't see SWTOR as #1 one year from now. 

    THANK YOU !!!!!!!!!!!!!!!!!! Like I and many others have been trying to get some of you people to understand... Xfire can and does reflect gaming trends. THANKS !! Now we just need to get the rest of them to come to terms with this. thanks yet again for proving my point.

  • UnlightUnlight Member Posts: 2,540

    Originally posted by Icewhite

    Originally posted by Unlight

    This is essentially all I'm trying to get to (thanks for simplyfing it, Robokapp).  What about this group of gamers introduces the bias necessary to disqualify them as a valid polling group?

    Simple.  They are required to select and run one specific piece of software in order to participate.  Self-selection, by definition.

    Draw any conclusion you like from the data you are presented with.  The result isn't in dispute, but the methodology most certainly (and correctly) is.

     

    I'm actually not defending the methodology because it's not a poll at all, as far as I know.  I don't understand how the data is being gathered and I don't know how it's being processed.  My question was related to the Xfire users and why they wouldn't make a suitable polling group for this game, were they to actually be polled

    I'm not drawing any conclusions about the results that this thread is discussing.  In fact, I think I've already stated that I wouldn't trust them.  I believe "nebulous" was the word I used.

  • PelaajaPelaaja Member Posts: 697

    Originally posted by Icewhite

    Originally posted by Robokapp

    m'kay. and how were these people chosen?

     since xfire is older than swtor, they probably werent chosen based on swtor. Soooo...the sample space is therefore random.

     It is decidedly non-random, consisting of (many) FPS gamers, and an scattered assortment of other gamer "breeds".  Using the XFire software itself is a self-selected process, by definition.  When you to present those results as representative of a much larger population (all gamers), you have made a basic statistical error.

    If you want a random sampling of the population All Gamers, one method would include Random Direct Dialing (as used by most political pollsters), or any one of a dozen other common schemes.  Stats 110 stuff, again.

    Hmm,

    when they make those RDCs and other gallups they don't choose people randomly. For example, in nordic countries everyone over 18 is eligible to vote. When they start gathering gallups, they start calling these non-random people eligible to vite) whom so forth consist of those who don't vote (FPS-gamers in this context), those who vote for Donald Duck and so on. In USA it's even more XFire type thing, because people have to register to vote (self-selected process).

    XFire can be used as a trend meter. It is accurate enough, when combined with other data people on these forums can mine from the net (financial reports, official announcements, polls of all sorts, personal experience).

    You should never underestimate the power of MMORPG.com (or it's users).

    image

  • uohaloranuohaloran Member Posts: 811

    Can we get a forum filter for any links or references to Xfire?

  • SovrathSovrath Member LegendaryPosts: 32,952

    Originally posted by Tyvolus

     




    Originally posted by Sovrath





    Originally posted by GrayGhost79






     




     

    It's only accurate in showing trends from Xfire players. If you're concerned with what people on Xfire play it's an excellent tool. However thats all it shows and it's rarely been even remotely accurate outside of Xfire users. And yes that includes its accuracey showing trends. 

     

    Now I'm not saying Xfire has been wrong all the time, even a broken watch is right twice a day. 






    To this point, I don't use X-Fire and would never use X-fire. So where is my voice in these statistics? Answer: Not present.

    Two other people at my work play mmo's and they don't use x-fire either. Where is their voice? Answer: Not present.

    So X-fire may be an excellent tool to track a certain demographic of people but it's not entering in my decisions and I suspect it's not entering in the decisions of thousands of other people like me.

    Perhaps it might help to figure out which demographic of "gamers" use x-fire and that might have more bearing on what it truly can track.




     

    "To this point, I don't use X-Fire and would never use X-fire. So where is my voice in these statistics? Answer: Not present.

    Two other people at my work play mmo's and they don't use x-fire either. Where is their voice? Answer: Not present."

    THATS WHY IT IS CALLED A STATISTICAL SAMPLE !!!!!!!!!!! Just like polling or a survey. Just like when a movie is pre-screened, they take a VERY small sample of movie goers and get their feedback -- or even a closed Beta, they take a small sample of gamers and work with them. Come on !!! are you serious right now ???

    ok deep breath now.

    'yes" it's a statistical sample Which means that X-Fire takes a sample of a certain demographic. I and others are not part of that demographic therefore we aren't being counted as we are not part of the demographic that partakes in X-Fire. whcih means that using X-Fire to predict trends becomes a bit more dicey. What it can do is predict the trends of the types of players who use X-Fire.

    Did you not see the part that I highlighted or were you too busy being incredulous?

    Like Skyrim? Need more content? Try my Skyrim mod "Godfred's Tomb." 

    Godfred's Tomb Trailer: https://youtu.be/-nsXGddj_4w


    Original Skyrim: https://www.nexusmods.com/skyrim/mods/109547

    Try the "Special Edition." 'Cause it's "Special." https://www.nexusmods.com/skyrimspecialedition/mods/64878/?tab=description

    Serph toze kindly has started a walk-through. https://youtu.be/UIelCK-lldo 
  • TyvolusTyvolus Member Posts: 190


    Originally posted by dubyahite
    You guys are still asking for reasons why it's a bad polling group? Self selected sample as Icewhite said. Read his post. Take a stats class.

    yeah, that self-selected sample being a bunch of gamers, with interests ranging the entire spectrum of games -- MMOs, RTS, RPGs, FPS, etc. you act like xfire is used by soccer moms --- its a sample of GAMERS who play all kinds of GAMES. seems like a darn good bunch of people to determine trends in GAMING to me.

  • YamotaYamota Member UncommonPosts: 6,593

    Originally posted by catlana

    Originally posted by Robokapp


    Originally posted by teakbois


    Originally posted by Unlight


      It sounds to me like they are just a typical subset and nothing really sets them apart from the rest of us.

    But they arent a typical subset.  Its like going back 5 years and saying myspace users were a typical cross section of internet users.

    stop with parallels and tell us what's atypical about them please.

    Xfire was developed by several fps "stars". Xfire originally focused on fps games for users with non dedicated communication (ala could not afford their own server).  Thus Xfire users tend to be younger and less affluent. More established players / guilds tend to use more secure channels (ala vent, etc) for privacy reasons.  Xfire is a subset of users. 

    Pure conjecture on your part. I installed XFire when I installed AoC a couple of years ago and it has since then been bundled with alot of non FPS games. And I dont use XFire for communicating, I use it to track my gaming time which I think is the most common use.

    So I see no hard evidence showing that XFire gamers would be "atypical" gamers. Specially since it is accurately showing both WoW and LoL as being among the most popular games currently being played, which they are.

  • YamotaYamota Member UncommonPosts: 6,593

    Originally posted by Tyvolus

     




    Originally posted by dubyahite

    You guys are still asking for reasons why it's a bad polling group? Self selected sample as Icewhite said. Read his post. Take a stats class.



     

    yeah, that self-selected sample being a bunch of gamers, with interests ranging the entire spectrum of games -- MMOs, RTS, RPGs, FPS, etc. you act like xfire is used by soccer moms --- its a sample of GAMERS who play all kinds of GAMES. seems like a darn good bunch of people to determine trends in GAMING to me.

    This.

    Add to this the fact that it is accurately showing the most popular (western) games as being very popular. It even showed the decline in WoW numbers not long ago, one that was confirmed by Blizzard.

  • IcewhiteIcewhite Member Posts: 6,403

    Originally posted by Tyvolus

    THANK YOU !!!!!!!!!!!!!!!!!! Like I and many others have been trying to get some of you people to understand... Xfire can and does reflect gaming trends. THANKS !! Now we just need to get the rest of them to come to terms with this. thanks yet again for proving my point.

    Another early thing really you must learn just to pass that 110 level stats class:  Anecdote != Data.

    Self-pity imprisons us in the walls of our own self-absorption. The whole world shrinks down to the size of our problem, and the more we dwell on it, the smaller we are and the larger the problem seems to grow.

  • SovrathSovrath Member LegendaryPosts: 32,952

    Originally posted by Tyvolus

     




    Originally posted by dubyahite

    You guys are still asking for reasons why it's a bad polling group? Self selected sample as Icewhite said. Read his post. Take a stats class.



     

    yeah, that self-selected sample being a bunch of gamers, with interests ranging the entire spectrum of games -- MMOs, RTS, RPGs, FPS, etc. you act like xfire is used by soccer moms --- its a sample of GAMERS who play all kinds of GAMES. seems like a darn good bunch of people to determine trends in GAMING to me.

    But what about non "self proclaimed" gamers?

    With games becoming more open to the masses there are a lot of people who are signing into these things who are not gamers. They don't follow forums, they don't follow the latest news they just have an inkling to play a game. There are a lot of these types of players who play WoW and I suspect they might be interested in trying other games.

    So sure, x-fire is a great way to track die-hard gamers who can't get enough fo games and want to use the service. But it's not going to show the other side of the coin.

     

    Like Skyrim? Need more content? Try my Skyrim mod "Godfred's Tomb." 

    Godfred's Tomb Trailer: https://youtu.be/-nsXGddj_4w


    Original Skyrim: https://www.nexusmods.com/skyrim/mods/109547

    Try the "Special Edition." 'Cause it's "Special." https://www.nexusmods.com/skyrimspecialedition/mods/64878/?tab=description

    Serph toze kindly has started a walk-through. https://youtu.be/UIelCK-lldo 
  • IcewhiteIcewhite Member Posts: 6,403

    Originally posted by Yamota

    So I see no hard evidence showing that XFire gamers would be "atypical" gamers. Specially since it is accurately showing both WoW and LoL as being among the most popular games currently being played, which they are.

    Does not matter if they're "atypical" gamers or not.

    Are they represented in the exact same proportions and the same deviation as the parent population?  If not, they're going to produce biased results, a result that is different from what the full population would produce.  How different?  We'd need to figure the Sample Bias.

    I would assume none of you, or me, or particularly Xfire, would be willing to analyze their own sample in enough detail to calculate the total Sample Bias.

    All right, I can continue to try to educate here, but it appears people want XFire predictions to be "right" more than they want to learn why it's just never going to be good polling.  You can't overcome faith with mere mathspeak, so I'll bow out now.

    Self-pity imprisons us in the walls of our own self-absorption. The whole world shrinks down to the size of our problem, and the more we dwell on it, the smaller we are and the larger the problem seems to grow.

  • dubyahitedubyahite Member UncommonPosts: 2,483
    The statement that it accurately shows wow and lol as the most popular games doesn't mean squat or prove anything.

    What if the xfire sample is biased towards those games for some other reason you dot know of? If there is some reason that LoL users are more likely to sign up for xfire than other gamers, then that is a bias.

    This is the probl with a self selected sample. You have no idea what these peoples motivation for joining and using xfire is. If wow players are more likely than Eve players to use xfire, then eve players are underrepresented when comparing the two, even though xfire correctly reports wow as #1.

    Shadow's Hand Guild
    Open recruitment for

    The Secret World - Dragons

    Planetside 2 - Terran Republic

    Tera - Dragonfall Server

    http://www.shadowshand.com

  • ChimpsChimps Member Posts: 192

    People still use Xfire?

    It does nothing but lag my games and it's pointless, let's keep track of how little life i have.

    Haven't used it for years lol.

  • YamotaYamota Member UncommonPosts: 6,593

    Originally posted by Icewhite

    Originally posted by Yamota

    So I see no hard evidence showing that XFire gamers would be "atypical" gamers. Specially since it is accurately showing both WoW and LoL as being among the most popular games currently being played, which they are.

    Does not matter if they're "atypical" gamers or not.

    Are they represented in the exact same proportions and the same deviation as the parent population?  If not, they're going to produce biased results, a result that is different from what the full population would produce.  How different?  We'd need to figure the Sample Bias.

    I would assume none of you, or me, or particularly Xfire, would be willing to analyze their own sample in enough detail to calculate the total Sample Bias.

    All right, I can continue to try to educate here, but it appears people want XFire predictions to be "right" more than they want to learn why it's just never going to be good polling.  You can't overcome faith with mere mathspeak, so I'll bow out now.

    Like I said before, there is no way of knowing how representetive, or not, the XFire gamers are of the total population so saying it is representative would be as valid as saying it is not.

    I dont think anyone is going to claim that is will show completely unbiased results but with such a huge sample size and lack of any study showing the sample bias, as you put it, I choose to belive that it is largely accurate.

    It accurately showed WoWs dominance on the MMORPG market.

    It accurately showed Aion as being the second most popular MMORPG, in the west (prior to SW:TOR release)

    It accurately showed the huge surge in people playing AoC and Dungeon Fighter Online after they went F2P.

    It accurately showed the LoL rising star.

    And it, most likely, showed that SW:TOR is the second most played MMORPG, in the west, right now.

    The part of it being peaked might be premature to say and why I used the word: seemed.

    But XFire has shown several times the trends in gaming and no-one is claiming it is the perfect polling tool, but it is the best we got.

     

  • DarLorkarDarLorkar Member UncommonPosts: 1,082

    Originally posted by dubyahite

    The statement that it accurately shows wow and lol as the most popular games doesn't mean squat or prove anything.



    What if the xfire sample is biased towards those games for some other reason you dot know of? If there is some reason that LoL users are more likely to sign up for xfire than other gamers, then that is a bias.



    This is the probl with a self selected sample. You have no idea what these peoples motivation for joining and using xfire is. If wow players are more likely than Eve players to use xfire, then eve players are underrepresented when comparing the two, even though xfire correctly reports wow as #1.


     



    You know whats foolish? Trying to make more of what it is than it is :)

    Take the above statement. If WOW and LOL ( have not checked so just for arguments sake if it is true) are showing as the most played on x-fire. Then trying to give reasons why it must be wrong or invallid, thats foolish.

    They very well prove exactly what they are. The most popular and most played games on x-fire.

    All you are proving here is your own bias against another way to see a sample of what games some gamers on x-fire are playing.

     

  • sancher36sancher36 Member UncommonPosts: 458

    I have used xfire but its not as popular as mumble, ventrillo and teamspeak used by the majority of mmo players.

    I would like to see statistics on these programs I mentioned :)

    Would love to see someone try and organise an ingame raid on xfire or guild pvp hehehe

     

     

  • IcewhiteIcewhite Member Posts: 6,403

    Originally posted by Yamota

    I dont think anyone is going to claim that is will show completely unbiased results but with such a huge sample size

    Oh...I..sample size, really?...no, must resist.

    Self-pity imprisons us in the walls of our own self-absorption. The whole world shrinks down to the size of our problem, and the more we dwell on it, the smaller we are and the larger the problem seems to grow.

  • TealaTeala Member RarePosts: 7,627

    Originally posted by zymurgeist

    It's just such a giggle. Ok the hours played has fallen relative to other games from when it was briefly #3 before they shut the servers down for maintenance. However the number of players using xfire playing swtor has risen. It's now in fifth place and the site says it peaked at fourth. What does this mean? Xfire does not support the OP's premise and indeed shouldn'tr be used to try and support any theory over the short term. It's not useful in comparing games because the relative populations of players using xfire in each game is different, and unpredictable. It can be an indicator of possible long term trends but is by no means an accurate predictor on it's own.

    SWTOR never hit #3 on xfire, if it did, it would say it did, it has gotten to #4.    With that being said, I have taken classes on statistics.    Anyone here ever heard of William Deming?  He is like the guy that showed that all you need to do to see how things are going, whether it was in manufacturing or people watching TV, was take a small sample of the thing you are studying to see a bigger picture. 

    For instance instead of checking 100 parts, you take 10 parts and check them.  If 1 out of 10 are bad, then you can bet that you'll have 10 bad parts out of 100.

    Same applies to things like the Nielson ratings.  If you sample poll 1000 people  and ask what show you watch on Monday night, and 600 out of the 1000 were watching show X, guess what, take that numberand apply to how many people have cable or TV's(a good guess since there is no actual number to show what households actually have a TV but we can assume X number do based on TV sells in the US) and with sample numbers you can predict how many people across the country watched show X.

    Xfire is no different.  It is a sample of gamers.  It shows which games are popular with the gamers using the xfire system.  Just as Nielson doesn't have a ratings box in everyones home to gather data, they can within an above average accuracy predict what shows are being watched and which ones are not.    Simple statistics.

  • TealaTeala Member RarePosts: 7,627

    Originally posted by zymurgeist

    Originally posted by Icewhite


    Originally posted by Yamota

    I dont think anyone is going to claim that is will show completely unbiased results but with such a huge sample size

    Oh...I..sample size, really?...no, must resist.

     I can't

    Swtor registered players more than a million, way more.

    Swtor players also using xfire to play today a bit more than ten thousand.

    That's one percent or less. That's a statistically insignificant population sample to anyone who has even the most basic understanding of the subject. It might as well be zero.

    Actually I do know the subject.  How many homes in the US have a Nielson box?   By your standard they would have to have 300k boxes in homes.  They don't.   Yet they accurately predict what shows are being watched and which are not.   Your argument is not holding up to the facts. 

  • AdamTMAdamTM Member Posts: 1,376

    Originally posted by Teala

    Originally posted by zymurgeist

    It's just such a giggle. Ok the hours played has fallen relative to other games from when it was briefly #3 before they shut the servers down for maintenance. However the number of players using xfire playing swtor has risen. It's now in fifth place and the site says it peaked at fourth. What does this mean? Xfire does not support the OP's premise and indeed shouldn'tr be used to try and support any theory over the short term. It's not useful in comparing games because the relative populations of players using xfire in each game is different, and unpredictable. It can be an indicator of possible long term trends but is by no means an accurate predictor on it's own.

    SWTOR never hit #3 on xfire, if it did, it would say it did, it has gotten to #4.    With that being said, I have taken classes on statistics.    Anyone here ever heard of William Deming?  He is like the guy that showed that all you need to do to see how things are going, whether it was in manufacturing or people watching TV, was take a small sample of the thing you are studying to see a bigger picture. 

    For instance instead of checking 100 parts, you take 10 parts and check them.  If 1 out of 10 are bad, then you can bet that you'll have 10 bad parts out of 100.

    Same applies to things like the Nielson ratings.  If you sample poll 1000 people  and ask what show you watch on Monday night, and 600 out of the 1000 were watching show X, guess what, take that numberand apply to how many people have cable or TV's(a good guess since there is no actual number to show what households actually have a TV but we can assume X number do based on TV sells in the US) and with sample numbers you can predict how many people across the country watched show X.

    Xfire is no different.  It is a sample of gamers.  It shows which games are popular with the gamers using the xfire system.  Just as Nielson doesn't have a ratings box in everyones home to gather data, they can within an above average accuracy predict what shows are being watched and which ones are not.    Simple statistics.

    Nielsen ratings are horribly inaccurate and have been for years. They feature an error-margin so big that its not even funny.

    This is not a reliable way to sample content, neither is Xfire.

    At least Nielsen is using a specific way to select an "average" household and include different demographics. Which is still horrible.

    image
  • TealaTeala Member RarePosts: 7,627

    Originally posted by AdamTM

    Originally posted by Teala


    Originally posted by zymurgeist

    It's just such a giggle. Ok the hours played has fallen relative to other games from when it was briefly #3 before they shut the servers down for maintenance. However the number of players using xfire playing swtor has risen. It's now in fifth place and the site says it peaked at fourth. What does this mean? Xfire does not support the OP's premise and indeed shouldn'tr be used to try and support any theory over the short term. It's not useful in comparing games because the relative populations of players using xfire in each game is different, and unpredictable. It can be an indicator of possible long term trends but is by no means an accurate predictor on it's own.

    SWTOR never hit #3 on xfire, if it did, it would say it did, it has gotten to #4.    With that being said, I have taken classes on statistics.    Anyone here ever heard of William Deming?  He is like the guy that showed that all you need to do to see how things are going, whether it was in manufacturing or people watching TV, was take a small sample of the thing you are studying to see a bigger picture. 

    For instance instead of checking 100 parts, you take 10 parts and check them.  If 1 out of 10 are bad, then you can bet that you'll have 10 bad parts out of 100.

    Same applies to things like the Nielson ratings.  If you sample poll 1000 people  and ask what show you watch on Monday night, and 600 out of the 1000 were watching show X, guess what, take that numberand apply to how many people have cable or TV's(a good guess since there is no actual number to show what households actually have a TV but we can assume X number do based on TV sells in the US) and with sample numbers you can predict how many people across the country watched show X.

    Xfire is no different.  It is a sample of gamers.  It shows which games are popular with the gamers using the xfire system.  Just as Nielson doesn't have a ratings box in everyones home to gather data, they can within an above average accuracy predict what shows are being watched and which ones are not.    Simple statistics.

    Nielsen ratings are horribly inaccurate and have been for years. They feature an error-margin so big that its not even funny.

    This is not a reliable way to sample content, neither is Xfire.

    At least Nielsen is using a specific way to select an "average" household and include different demographics. Which is still horrible.

    You may see it as horrible, but the industry doesn't, because it works.

  • TealaTeala Member RarePosts: 7,627

    Originally posted by zymurgeist

    Originally posted by Teala


    Originally posted by zymurgeist

    It's just such a giggle. Ok the hours played has fallen relative to other games from when it was briefly #3 before they shut the servers down for maintenance. However the number of players using xfire playing swtor has risen. It's now in fifth place and the site says it peaked at fourth. What does this mean? Xfire does not support the OP's premise and indeed shouldn'tr be used to try and support any theory over the short term. It's not useful in comparing games because the relative populations of players using xfire in each game is different, and unpredictable. It can be an indicator of possible long term trends but is by no means an accurate predictor on it's own.

    SWTOR never hit #3 on xfire, if it did, it would say it did, it has gotten to #4.    With that being said, I have taken classes on statistics.    Anyone here ever heard of William Deming?  He is like the guy that showed that all you need to do to see how things are going, whether it was in manufacturing or people watching TV, was take a small sample of the thing you are studying to see a bigger picture. 

    For instance instead of checking 100 parts, you take 10 parts and check them.  If 1 out of 10 are bad, then you can bet that you'll have 10 bad parts out of 100.

    Same applies to things like the Nielson ratings.  If you sample poll 1000 people  and ask what show you watch on Monday night, and 600 out of the 1000 were watching show X, guess what, take that numberand apply to how many people have cable or TV's(a good guess since there is no actual number to show what households actually have a TV but we can assume X number do based on TV sells in the US) and with sample numbers you can predict how many people across the country watched show X.

    Xfire is no different.  It is a sample of gamers.  It shows which games are popular with the gamers using the xfire system.  Just as Nielson doesn't have a ratings box in everyones home to gather data, they can within an above average accuracy predict what shows are being watched and which ones are not.    Simple statistics.

     It hit #3 the morning it went down for maintenance. You need to go back to statistics class. You should concentrate on the part where they talk about what makes a statistically relevant sample.

    No it didn't.  If it hit #3 it would say it did on xfire.  The furthest it rated was #4.   Go see for yourself.  If it ever did hit #3 it would say it hit number 3. 

  • TealaTeala Member RarePosts: 7,627

    Originally posted by zymurgeist

    Originally posted by Teala


    Originally posted by zymurgeist

    It's just such a giggle. Ok the hours played has fallen relative to other games from when it was briefly #3 before they shut the servers down for maintenance. However the number of players using xfire playing swtor has risen. It's now in fifth place and the site says it peaked at fourth. What does this mean? Xfire does not support the OP's premise and indeed shouldn'tr be used to try and support any theory over the short term. It's not useful in comparing games because the relative populations of players using xfire in each game is different, and unpredictable. It can be an indicator of possible long term trends but is by no means an accurate predictor on it's own.

    SWTOR never hit #3 on xfire, if it did, it would say it did, it has gotten to #4.    With that being said, I have taken classes on statistics.    Anyone here ever heard of William Deming?  He is like the guy that showed that all you need to do to see how things are going, whether it was in manufacturing or people watching TV, was take a small sample of the thing you are studying to see a bigger picture. 

    For instance instead of checking 100 parts, you take 10 parts and check them.  If 1 out of 10 are bad, then you can bet that you'll have 10 bad parts out of 100.

    Same applies to things like the Nielson ratings.  If you sample poll 1000 people  and ask what show you watch on Monday night, and 600 out of the 1000 were watching show X, guess what, take that numberand apply to how many people have cable or TV's(a good guess since there is no actual number to show what households actually have a TV but we can assume X number do based on TV sells in the US) and with sample numbers you can predict how many people across the country watched show X.

    Xfire is no different.  It is a sample of gamers.  It shows which games are popular with the gamers using the xfire system.  Just as Nielson doesn't have a ratings box in everyones home to gather data, they can within an above average accuracy predict what shows are being watched and which ones are not.    Simple statistics.

     It hit #3 the morning it went down for maintenance. You need to go back to statistics class. You should concentrate on the part where they talk about what makes a statistically relevant sample.

     

    When if ever have you seen a poll that claimed a margin of error of less than one percent?  If you can cite one I can guarantee you the people who claim it are liars.



    A margin of error is always a possibility.   But I have used xfire to see what MMO are healthy for a long time and it has always been accurate to show us the health of a game.   Deny it all you want, spin it anyway you like, numbers do not lie.

  • catlanacatlana Member Posts: 1,677

    Originally posted by Teala

     

    You may see it as horrible, but the industry doesn't, because it works.

    Neilson strives very hard for a approximate disturbuted samplings / ratings similiar to the population (ie % high, medium, low income as well as set % of particular ethnic groups). None of that is present in Xfire anyway. So the point is kinda of mute.

This discussion has been closed.