Seems like the game has peaked on XFire

Pelaaja

Originally posted by Forumtalker

Originally posted by gervaise1

Trends is what XFire is good for assuming that people play about the same number of hours a day - and at the moment the average is usually 5-6 hours which is pretty much what WA has said.

 

TREND 1 : the ramp up during EGA is exactly what you would expect.

 

TREND 2 : Pretty flat around the 75,000 hour line between 20th - 31st December. Ups and downs due to weekends, Christmas Eves, server downtime etc.

- no big rise on Christmas Day. Remember all the chatter about how Santa Claus would be delivering the game to lots of people.

- people who really wanted the game either pre-ordered or bought it on the 20th December.

TREND 3 : Small blip, 5-8%, after the 'New Year's' weekend. You can see the current XFire high water mark and it is post launch. This could be due to people having more time to play over the New Year weekend but the simple answer is additional post-Christmas sales - people going shopping with Christmas money, BestBuy gift cards etc. etc. As mentioned above the average hours per person played has usually been in the 5-6 hour mark (as I type it is 10,853 people and 56822 so just over 5 again).

- Now if the 75,000 hour line relates to EA's 1M+ registrations they talked about in their press release just before Christmas then - maybe - the 5-8% blip equate to an extra c. 100k registrations. (I believe EA would have reported 1.5M is they had 1.5)

Note: EA sells boxes to Amazon, BestBuy, GAME, etc. so registrat   

TREND 4 : no big, huge increase in play time post launch. No 'word of mouth' monster sales rise. No upward trajectory.

- Going forward people will probably play less as the shiny wears of, so even a 'flat line' could be the due to increased subscriptions (average hours per person played is a key factor)

- There sould be a downward trend as the average hours at the moment is largely unchanged; to early to say though.

- the 19th January will be a key date.

 

I agree 1st billing cycle will be telling.

 

Trend #1

Out of 20,766,204 registered users of Xfire 10,555 players playing SWTOR represent .0508% of total Xfire registered players playing SWTOR.

 

Trend #2

.0508% of Xfire users playing SWTOR represent 55,000-101,000 of total SWTOR sales (1-2 million estimate)

 

XFire is actually more forgiving to the game than torarena statistics. It shows about 17% decline from highest peak to last peak. Didn't want to calculate from highest peak to lowest valley, because I think it would make the graph go down too steep.

Does anyone know the relativity between Y-Axis of the graphs there? Because they aren't directly concurrent users.

gervaise1

Forumtalker posted:

Trend #2

.0508% of Xfire users playing SWTOR represent 55,000-101,000 of total SWTOR sales (1-2 million estimate)

 

Be careful talking about sales. EA sales will be higher as they sell boxes to Amazon, BestBuy etc. etc. - and they get some money for these sales it is true. (One analyst post I read reckoned that if they sold 2M boxes + digital downlloads they would make $60M).

Registrations is the key and at the moment the only number we have is what EA have announced: 1M+. And on that basis XFire users could represent 1% of the playerbase; 0.5% if it is 2M registrations.

If it was close to 2M I feel pretty sure EA would have said 'nearly 2M' - as Trion did with Rift (nearly 1M). I reckon they would have said 1.5M if they had 1.5 as well. Not sure below that.

There are lots and lots of uninformed wild guesses doing the rounds at the moment. We will know more after EA give their quarterly earnings presentation. 

MMOGamer71

Originally posted by gervaise1

Forumtalker posted:

Trend #2

.0508% of Xfire users playing SWTOR represent 55,000-101,000 of total SWTOR sales (1-2 million estimate)

 

Be careful talking about sales. EA sales will be higher as they sell boxes to Amazon, BestBuy etc. etc. - and they get some money for these sales it is true. (One analyst post I read reckoned that if they sold 2M boxes + digital downlloads they would make $60M).

Registrations is the key and at the moment the only number we have is what EA have announced: 1M+. And on that basis XFire users could represent 1% of the playerbase; 0.5% if it is 2M registrations.

If it was close to 2M I feel pretty sure EA would have said 'nearly 2M' - as Trion did with Rift (nearly 1M). I reckon they would have said 1.5M if they had 1.5 as well. Not sure below that.

There are lots and lots of uninformed wild guesses doing the rounds at the moment. We will know more after EA give their quarterly earnings presentation. 

Yeah was using conservative numbers 1 million confirmed, 2 million rumored.

http://www.gamasutra.com/view/news/36509/Analyst_Star_Wars_The_Old_Republic_Could_Sell_3M.php 

Will we see 3 million out the door?  Only Europe and NA releases so far.

 

Edit:

Hate smart phone typo errors.

gervaise1

Originally posted by Robokapp

Originally posted by gervaise1

- the 19th January will be a key date.

 

I guarantee it'll be a peak. or at least higher than the friday before it since its a  friday.

 

the 23rd is the interesting one imo. because its the first non-weekend day after subscriptions kick in.

Agree - good point.

Someone should post the number of people playing tomorrow, the 12th, alomg with their average play time. That will be the thing to compare it to.

Pelaaja

Please, explain this graph to me:

What is that flat line in the middle of the night? At all other times there's somekind of variation on the graph but not at that time.

It happens frequently:

gervaise1

Forumtalker posted:

<<Yeah wa using conservative numbers 1 million confirmed, 2 million rumored.

http://www.gamasutra.com/view/news/36509/Analyst_Star_Wars_The_Old_Republic_Could_Sell_3M.php

Will we see 3 million out the door? >>

 

 

I find it really scary that the 'prefessional' analyst seemed to be using nothing more than Amazon pre-orders (worldwide?). And as the trend was better than Star Craft 2 and Cataclysm shouldn't the number out the door have been above 4M. 

EA quarterly results soon. Will be interesting to see what they say about:

- 'sales' :  digital + boxes sold to retailers

- 'registrations'  :  people who have bought the game and started the 30 days;

- 'subscriptions' : people who have decided to pay extra to play beyond the 30 days.

 

gervaise1

Pelaaja:

The flat line is probably people leaving their characters logged on; what it was assumed to be on a DAoC specific graph that was run for some years - and that broke down the EU servers and NA servers so you could tell when 'night time' was. I assume the graph you posted is NA + EU combined and with a 9 hour time difference between central europe and the pacific coast it makes it hard to say when night is.  

(Don't know for sure though).

Pilnkplonk

Originally posted by Forumtalker

 

Common sense tells me that any third party software to track what gamers are playing whether it be TOR or Microsoft Chess online IS NOT an accurate reflection of time played.

It is not Steam.

Obviously, the accurate reflection is only given by a company who made the game and in whose interest it is to portray the game as successful to the public. Just like the only way to audit a bank fairly is to have it done by the bank's own staff because, hey, they know it best.

Are you putting us on?

Vato26

Originally posted by Forumtalker

Originally posted by Yamota

Originally posted by Pelaaja

It's on a slight decline in XFire (56281) which is 2876 away from the lowest after launch (53405).

Could someone expain to me some graphs on SWTOR Arena.

There's something that seems a definite cut on lowest valleys. For example EU monthly seems to have it around 100k and US at 125k. Only times when the graph penetrates these thresholds seems to be when they have shut servers down.

Just a coincidence or error in retrieving data probably?

Interesting that this SWTOR Arena is pretty much showing the same trend as XFire. Highest during last weeks of Decembre and declining a bit now. So much for XFire being useless for showing trends. But I am sure it is all a coincidence.. for sure.

It's useful to show trends when you want it to show the trend that supports a position.

Yup, the people that try to apply X-fire statistics outside of X-fire are doing it for their self-serving agenda.

Facts still remain:

1. 
   X-fire is only useful in interpereting trends within the X-fire community.
2. 
   X-fire statistics are completely invalid when trying to apply them to outside the X-fire community.

Pelaaja

Originally posted by gervaise1

Pelaaja:

The flat line is probably people leaving their characters logged on; what it was assumed to be on a DAoC specific graph that was run for some years - and that broke down the EU servers and NA servers so you could tell when 'night time' was. I assume the graph you posted is NA + EU combined and with a 9 hour time difference between central europe and the pacific coast it makes it hard to say when night is.  

(Don't know for sure though).

They're EU-daily and EU-monthly.

If your explanation is correct, there's certain people that are constantly leaving their characters online to do something

smh_alot

I think the flat line is confirmation of what I was saying earlier in this thread, namely that they're using the same method that the guys from the Rift charting sites were using: namely, that they compile the data based on the server status category instead of actual hrs played or ingame player population counts.

So, for the night you'll see a consistent flat line in the graphs if all the servers on the EU or US side have status 'light'.

The problem with that method is, that the status category isn't saying that much about actual player counts and hours - standard could in theory maybe range from 1000-2000 for example but 'heavy' from 2000-2500. Also the definition of the status category hasn't been a constant: 'very heavy' and 'full' indicated different player populations in the first days and week(s) than later on, when the player cap was increased step by step.

Metentso

Originally posted by Sukiyaki

Those "population" stats ...

Nice informative post Sukiyaki.

Do you know also if they increased the capacity of the servers, so what was "full" before might be just "heavy" now, for instance?

Guileplayer

Originally posted by Sukiyaki

Originally posted by Pelaaja

Please, explain this graph to me:

What is that flat line in the middle of the night? At all other times there's somekind of variation on the graph but not at that time.

It happens frequently:

Those "population" stats are based on the population tags of the server not the actual population activity or playtime. The numbers on the y axis are just standins and bear very vague relation to the actual population. At most it tells how many people are active at the same time and if there is an down or upward trend. But it does not even closely show the impact of them. Its pure speculative bullshit to say "SWTOR lost 20% of its player in 2 weeks" when this graph showed a drop by 20% of that "number" over two weeks. Its both ignorant of the methods the datapoints are tracked and also what the "number" on the y-axis even show. That "number" does not even represent the "number of player playing the game" but only and even just inderectly their online activity and then it is roughly categorized into broad brackets of player concurrency no one but Bioware even knows.

They are extremely vague to show the real population trends.

On swtorarena.com Low=1 Standard=2 High=3 VeryHigh=4(likely) Full=5(likely)

But no one even knows if  "High" has 1.5 times more player than standard in reality. A drop from 1.5 to 1 on the graph does not neccesarily imply a drop of 1/3 in reality.  These "numbers" are abstract as they use their own counting system not directly related to the real population. Thats one of the core issues here and the reason, why the graph is almost useless other, than to roughly predict how busy the server are relatively to the others or in general (by the unkown definition of Bioware) , second of course the trend of the status tags over the past, and third if its going up or down, but not even remotely accurate "how much".

 

An example for the rough brackets: lets say a server from 0-499 player online is "low" and 500-999 is "normal". If a server drops from 501 to 500 thus from standard to low then the value on that graph goes from 2 to 1 for that server. Even thought thats just a 0.2% difference in reality, the statistics says "loss" of 50% on the graph. But the latter is all what you see on the graph. You cant go backwards from that relative loss on the graph to the accurate "real" drop.

And more servertrends combined do not neccesarily increase the accuracy. If five server go from 500 to 499 thats again just a 0.2% loss in reality, on the graph however these 5 went down by 50%, just because 5*2(standard) changed to 5*1(low).

 

Anr example for the unknown range of the brackets: Lets again say Low=0-499 Standart=500-1999 High=2000-2499

In this case on the graph of swtorarena 2499 player on one server would be worth just 3 times as much as one single player on another! Even on the lower end 2000 player would be worth just 3 times as much as 499

No one knows if these "numbers" on swtor arena which just stand in for the tags low/standart/high etc have the same distribution as the population brackets of these "tags" defined by Bioware.

 

But as if that wasnt enough the highest ranges also is usally capped with a very small range. The "full" to "(very)heavy" range in almost all MMOs.

Most game have those full server at launch which the second they have space again are just "heavy" or "very heavy" depending on game ranging from little as 20-50 player slots if not even less. Over time almost all switch to more lenient population because player simply player less by default than they do on launch irrelevant whether the games is godly or bad. But it also goes in two ways because it doesnt represent the number of player in the queues either.

So both these issues combined can easily result into a drop from average "full" per server to average "normal" per server on the status list, without hardly loosing a single player because they simply just play less per day thus do not overlap each other and thus do not increase the concurent user numbers the "status" is based on. A very common and expectable trend after any major patch or expansion or release. People in general simply spend much more time on newly released contents. That doesnt make them "more" or "less"

 

The  flat line as explained before is when all server are low=1 at nighttime. And that line then just equates directly to the number of existing server.

Best post i have read on mmorpg.com forums. And that is saying something. Sticky please.

Metentso

Seems like many SWTOR xFire users took the most famous fan advice and were "back to WoW" this weekend.

Pelaaja

Originally posted by Sukiyaki

Originally posted by Pelaaja

Please, explain this graph to me:

What is that flat line in the middle of the night? At all other times there's somekind of variation on the graph but not at that time.

It happens frequently:

Those "population" stats are based on the population tags of the server not the actual population activity or playtime. The numbers on the y axis are just standins and bear very vague relation to the actual population. At most it tells how many people are active at the same time and if there is an down or upward trend. But it does not even closely show the impact of them. Its pure speculative bullshit to say "SWTOR lost 20% of its player in 2 weeks" when this graph showed a drop by 20% of that "number" over two weeks. Its both ignorant of the methods the datapoints are tracked and also what the "number" on the y-axis even show. That "number" does not even represent the "number of player playing the game" but only and even just inderectly their online activity and then it is roughly categorized into broad brackets of player concurrency no one but Bioware even knows.

They are extremely vague to show the real population trends.

On swtorarena.com Low=1 Standard=2 High=3 VeryHigh=4(likely) Full=5(likely)

But no one even knows if  "High" has 1.5 times more player than standard in reality. A drop from 1.5 to 1 on the graph does not neccesarily imply a drop of 1/3 in reality.  These "numbers" are abstract as they use their own counting system not directly related to the real population. Thats one of the core issues here and the reason, why the graph is almost useless other, than to roughly predict how busy the server are relatively to the others or in general (by the unkown definition of Bioware) , second of course the trend of the status tags over the past, and third if its going up or down, but not even remotely accurate "how much".

 

An example for the rough brackets: lets say a server from 0-499 player online is "low" and 500-999 is "normal". If a server drops from 501 to 500 thus from standard to low then the value on that graph goes from 2 to 1 for that server. Even thought thats just a 0.2% difference in reality, the statistics says "loss" of 50% on the graph. But the latter is all what you see on the graph. You cant go backwards from that relative loss on the graph to the accurate "real" drop.

And more servertrends combined do not neccesarily increase the accuracy. If five server go from 500 to 499 thats again just a 0.2% loss in reality, on the graph however these 5 went down by 50%, just because 5*2(standard) changed to 5*1(low).

 

Anr example for the unknown range of the brackets: Lets again say Low=0-499 Standart=500-1999 High=2000-2499

In this case on the graph of swtorarena 2499 player on one server would be worth just 3 times as much as one single player on another! Even on the lower end 2000 player would be worth just 3 times as much as 499

No one knows if these "numbers" on swtor arena which just stand in for the tags low/standart/high etc have the same distribution as the population brackets of these "tags" defined by Bioware.

 

But as if that wasnt enough the highest ranges also is usally capped with a very small range. The "full" to "(very)heavy" range in almost all MMOs.

Most game have those full server at launch which the second they have space again are just "heavy" or "very heavy" depending on game ranging from little as 20-50 player slots if not even less. Over time almost all switch to more lenient population because player simply player less by default than they do on launch irrelevant whether the games is godly or bad. But it also goes in two ways because it doesnt represent the number of player in the queues either.

So both these issues combined can easily result into a drop from average "full" per server to average "normal" per server on the status list, without hardly loosing a single player because they simply just play less per day thus do not overlap each other and thus do not increase the concurent user numbers the "status" is based on. A very common and expectable trend after any major patch or expansion or release. People in general simply spend much more time on newly released contents. That doesnt make them "more" or "less"

 

The  flat line as explained before is when all server are low=1 at nighttime. And that line then just equates directly to the number of existing server.

Thanks for the explanation.

I'm sure they didn't have this link earlier, but now it's on the statistics page.

Anyway, thanks for a good explanation.

GMan3

Originally posted by Metentso

Seems like many SWTOR xFire users took the most famous fan advice and were "back to WoW" this weekend.

    Seriously, you are actually going to use xFire as your measure of how many players play SWTOR vs WoW?  How many people exactly USE xFire?  I know they claim in the tens of millions, but how many ACTUALLY use it?  I downloaded it shortly after it came out.  Created an account and everything.  Deleted it several days later and have never used it again.  That was several computers ago.  Never did delete my account though, so I guess their numbers may just be a little off.  They might also be a little off because people that do use it seldom ALWAYS use it.  Then there is the fact that not everyone uses xFire, there are several different programs like it after all.  So just using xFire is really inaccurate all told.  I guess if that is the only way you can get in an "I told you so" though, then go to it man.

troublmaker

Holy crap this thread confirms it.

Every single person that has SWTOR and XFire is still streaming on ToR.  That means that this game has a consistent fan base.

XFire proves this game is a hit.

smh_alot

Actually, regarding Xfire: this Sunday it reported 11.1k players for TOR, the Sunday before that 11.014 players and Sunday 2 weeks ago 10.8 players.

What I usually do when looking at tools as Xfire and Raptr is comparisons between the same day of the week, it has little meaning to compare a peak in the week with a valley in another week. If you want to determine trends, you have to compare the peaks with other peaks and valleys with other valleys from different time periods (although valleys can be influenced by server maintenance days).

I'm not saying that Xfire is the complete picture ofc, it's just one data gathering means of several with each their own degree of reliability and representation.

Yamota

Originally posted by smh_alot

Actually, regarding Xfire: this Sunday it reported 11.1k players for TOR, the Sunday before that 11.014 players and Sunday 2 weeks ago 10.8 players.

 

What I usually do when looking at tools as Xfire and Raptr is comparisons between the same day of the week, it has little meaning to compare a peak in the week with a valley in another week. If you want to determine trends, you have to compare the peaks with other peaks and valleys with other valleys from different time periods (although valleys can be influenced by server maintenance days).

 

I'm not saying that Xfire is the complete picture ofc, it's just one data gathering means of several with each their own degree of reliability and representation.

Yep, that is why I say it has peaked because if you look at the same days in the week you will see that it is pretty much un-changed. So unless this game does what no Themepark has done since the release of WoW it should start going down, but if that is in 1 month or 6 months is hard to tell.

Will be interesting to see the change in early february.

smh_alot

Actually, I was replying upon posters who're saying that SWTOR has taken a dive according to Xfire, which Xfire in its current figures says otherwise. But you're right, the really interesting results will be end of the month, after the 1 free month has ended.

Metentso

Originally posted by GMan3

Originally posted by Metentso

Seems like many SWTOR xFire users took the most famous fan advice and were "back to WoW" this weekend.

    Seriously, you are actually going to use xFire as your measure of how many players play SWTOR vs WoW?  How many people exactly USE xFire?  I know they claim in the tens of millions, but how many ACTUALLY use it?  I downloaded it shortly after it came out.  Created an account and everything.  Deleted it several days later and have never used it again.  That was several computers ago.  Never did delete my account though, so I guess their numbers may just be a little off.  They might also be a little off because people that do use it seldom ALWAYS use it.  Then there is the fact that not everyone uses xFire, there are several different programs like it after all.  So just using xFire is really inaccurate all told.  I guess if that is the only way you can get in an "I told you so" though, then go to it man.

I think I said "SWTOR xFire users"?

smh_alot

Originally posted by Tombill


Originally posted by smh_alot

Actually, I was replying upon posters who're saying that SWTOR has taken a dive according to Xfire, which Xfire in its current figures says otherwise. But you're right, the really interesting results will be end of the month, after the 1 free month has ended

The thread was "peaked".

You choose to ignore the fact that SW TOR is still in its launching month and sold 400.000 extra copies over the last 2 weeks., without change in the different stats.

Sales that will no longer be made in 2 weeks time.

In view of the server status, that tool from above and Xfire, the game should have shown growth in all those stats with a 400K boom of new players.

So it has "peaked" indeed with an added number of more than 400.000 players. Call it the Star Wars franchise.

 

Heh. The OP was from around Dec 27th, a period of time that player figures only just climbed above 9-9.5k and ranged between 9.5-10k, even with whole hordes of people enjoying buckets of spare time due to time off from work or school.

The weeks after that OP, the Xfire figures kept climbing to the figures it has now, regular weeks again with no peaks from having free days. I know of Rift that it showed a slowly continuing decrease already in the same period of time after launch.

Like said, all those semanthics aside (imo the OP was just wrong in its assumption or premature, no offense), the interesting part will be after the free month. Usually player figures show a 15-30% decrease from what they had in player numbers in the first week(s), all ongoing sales included.

Yamota

Originally posted by Tombill

Originally posted by smh_alot

Actually, I was replying upon posters who're saying that SWTOR has taken a dive according to Xfire, which Xfire in its current figures says otherwise. But you're right, the really interesting results will be end of the month, after the 1 free month has ended

The thread was "peaked".

You choose to ignore the fact that SW TOR is still in its launching month and sold 400.000 extra copies over the last 2 weeks., without change in the different stats.

Sales that will no longer be made in 2 weeks time.

In view of the server status, that tool from above and Xfire, the game should have shown growth in all those stats with a 400K boom of new players.

So it has "peaked" indeed with an added number of more than 400.000 players. Call it the Star Wars franchise.

You are assuming that during the last two weeks, where around 400k units were sold, none of the existing playerbase quit or played less than they did during the first week. A false assumption I would say because for sure there were people who did just that.

Yamota

Originally posted by smh_alot

Originally posted by Tombill

Originally posted by smh_alot

Actually, I was replying upon posters who're saying that SWTOR has taken a dive according to Xfire, which Xfire in its current figures says otherwise. But you're right, the really interesting results will be end of the month, after the 1 free month has ended

The thread was "peaked".

You choose to ignore the fact that SW TOR is still in its launching month and sold 400.000 extra copies over the last 2 weeks., without change in the different stats.

Sales that will no longer be made in 2 weeks time.

In view of the server status, that tool from above and Xfire, the game should have shown growth in all those stats with a 400K boom of new players.

So it has "peaked" indeed with an added number of more than 400.000 players. Call it the Star Wars franchise.

 

Heh. The OP was from around Dec 27th, a period of time that player figures only just climbed above 9-9.5k and ranged between 9.5-10k, even with whole hordes of people enjoying buckets of spare time due to time off from work or school.

 

The weeks after that OP, the Xfire figures kept climbing to the figures it has now, regular weeks again with no peaks from having free days. I know of Rift that it showed a slowly continuing decrease already in the same period of time after launch.

 

Like said, all those semanthics aside (imo the OP was just wrong in its assumption or premature, no offense), the interesting part will be after the free month. Usually player figures show a 15-30% decrease from what they had in player numbers in the first week(s), all ongoing sales included.

It only climbed around 5% just to drop again so the population levels in XFire has not changed noticeably since when i started this thread.

Also I did say "seems" which implies that it is no way conclusive and you are right, it if after the free month where the game will be tested and we can see how big the drop is.

kitarad

Yes let us see after the free month but probably will drop it won't go up typically every game loses people after the free month.