Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
X-fire numbers plummeting
This discussion has been closed.
It looks like you're new here. If you want to get involved, click one of these buttons!
Comments
See here:
http://www.mmorpg.com/discussion2.cfm/thread/336517/Seems-like-the-game-has-peaked-on-XFire.html
However Xfire means nothing when it deals with a games population.
Says another person who doesn't understand even the rudiments of statistical modeling. It absolutely does. XFire is statitiscally represenetative of the gaming community to an incredibly high degree of certainty because it is wholly representative of the gaming community due to its large size. XFire has a total user population of 21,000,000 total gamers with 250,000 logged in.
To measure the 20 million MMO players in Europe and North America to a 99% +/- 1%% confidence , you only need a sample size of 16,700 gamers. Considering most research uses 95% and 4%... You'd need just 600.
The currently logged size of X-Fire is drastically higher than what is needed to use the data with absolute confidence. Never mind the population IS drastically larger than what is needed for validity.
So, as you can see, the sample size is absolutely way past the threshold needed to use X-Fire as a predictive tool. No ifs. No ands. No buts. No whining, crying or BioDroning BS allowed. The subject is closed and you have lost because, desptie the liars who want to deny reality, X-Fire tells us what the PC gaming community is doing. And it does so with incredible accuracy.
And, just so you understand.. What ever X-Fire gives to its users is just to get them to use it... Their REAL customers are analysts and gaming executives, etc. who buy the data from X-Fire so they can see how they're doing in the market relative to their customers as well as in absolute terms. X-Fire isn't really about gamers. That part of the company, that which serves the gamers, is just bait. It's really about data mining and monetizing it's 'user' base. And they do that quite well.
Good pointsYou don't measure whole populations. Nor do you need to... The argument that it has to be done that way is, at best, false and, at worst, completely ignorant of statistics and sampling. XFire's population is, by itself, such a large sample that we have no reason to doubt it's sample size and the predictivity of the same.
I still think the biggest problem in education is that schools don't teach actual critical thinking (most of them teach a mockery of critical thinking that is nothing more than rejecting one's intellect and falling in line with tropes and lies), analysis and statistical methodoloy.
Where's your analysis that shows XFire users accurately represent the broad spectrum of MMO players. Isn't it possible that XFire users more closely represent the hardcore players rather than the full spectrum.
Do you think pollsters just randomly pick X number of people? Or maybe they actually pick people that represent a specific subset of voters? The same goes with Nielsens. They don't just randomly pick names out of hats, they do their best to make sure that everyone is as accurately covered as possible.
XFire represents anyone that signs up for it. That alone removes the likelihood that it is a fair representation of the gaming community as a whole.
You can challenge data based on
a) The sample not being large enough or
b) The sample not being random
XFire is certainly large enough sample to provide data, but it's not a random sample of PC gaming population.
Some games (FPS, RTS) require or have required the use of tools like XFire to stay connected with friends, and players playing those games have installed XFire for that purpose. Some clans in FPS games even require their members to use XFire. Whereas people who play, let's say previous Bioware games like Mass Effect or Dragon Age which are single-player only, have never needed to install XFire for those games. The sample isn't random. It's like claiming that you know USA's shopping habbits with incredible accuracy because you polled 1 million people who were shopping at a garden supply store.
XFire is still useful, but the sample isn't random sample out of all PC gamers, so there's going to be error and we don't know how large that error is. We can only quess.
EDIT: I didn't see the post above this post when posting /EDIT
In game I had the feeling we lost roundly 50% playing players and seems I was right.
I have to Agree with that X-Fire numbers only are true to X-Fire Users other than that it is an educated guess only. It only shows true numbers to X-Fire users.
Not everyone uses x-fire.
What a shock.
Im in a guild with 300 people and none have every used Xfire and neither have I. Xfire is a SMALL % of the population you cannot take an educated guess from such a small sample size.
There is little to no indicaton that X-fire users that play TOR wouldn't be representative for the game population. If nothing else, the trend is clear and not obfuscated by EA tweaking server load numbers. If EA were to encourage TOR players to start using Xfire it could cause a surge that is totally fake, but nothing looks abnormal about those numbers so far.
What it shows is a good cross section of the games playerbase. No it isn't a tell all tale that you cna determine every detail about the population, but it is a good indicator.
"I am not in a server with Gankers...THEY ARE IN A SERVER WITH ME!!!"
You have 300 players in your guild and your basing your opinion on that, yet you say that you shouldnt use a small sample size? Do you see the issue?
Also, in statistics you most often use a small sample size. As long as the sample is a random selection from the population you can pretty accuratly show trends.
Except that SWTOR is targeted to more casual players. Those casual players are MUCH more likely to NOT use this type of program while playing SWTOR. At the same time the more hardcore players ARE more likely to use this type of game tracking program. Therefore, the only thing this is showing is that the hardcore gamers are playing less often, not actually leaving the game.
Sorry, but the "random sample" of players you would need is NOT represented by XFire numbers.
"If half of what you tell me is a lie, how can I believe any of it?"
A month after Age of Conan was released
"Xfire numbers don't mean anything!"
A month after Vanguard was released
"Xfire numbers don't mean anything!"
A month after DC Universe was released
"Xfire numbers don't mean anything!"
See a pattern here?
The problem with this is that it is hard to tell. My thinking is that since its a chat client dedicated for gamers that your selecting a specific more dedicated group. Therefore, if xfire is used by more than casual gamers the trend would show that these gamers are not playing SWTOR anymore, but it neglects to show a potential (offsetting) rise in the casual population who dont use things like xfire (or even browse game forums).
The sample is not absolutly random and therefore should not be taken as fact.
Edit: Damnit, Gman beat me to the punch.
Why do XFire numbers HAVE to be all accurate to some people? Just accept that it isn't a random sampling of people and shows trends for those people that installed the software.
It is still very interesting information even if it only represents a certain type of player. Just don't make it out to be something it never claimed or tried to be.
It's possible that it is an accurate representation of the SWTOR population as a whole but it could also be way off in terms of who will or won't cancel their subscriptions.
I have yet to see anyone make a true analysis of XFire numbers versus numbers later released by the companies.
The problem isn't that XFire numbers are meaningless, because most people aren't saying that. It's the people claiming that XFire numbers somehow represent everyone, which they don't.
Actually, we have a WoW killer according to xfire -> http://www.xfire.com/games/lol/League_of_Legends/ on #1 the day WoW dropped from it...
About the whole yes/no validation of xfire numbers... As you can see from my signature, I'm using it, and it does show trends of what people play. It's not a 100% accurate number, but with all those people using xfire, the numbers might indeed be quite accurate...
Here is an example of why xfire numbers ae tough to judge:
When SWTOR announced the 1.7 subs, I figured 'maybe the percentageof xbox users is roughly the same for all games'. Using the ratio of actual subs to xfire users, I applied it to rifts xfire users. The result was that Rift has 125k subs. There is no way thats true.