Planetarion Forums - View Single Post

Mzyxptlk · 1 May 2009, 12:24

Alright, this post started as a two paragraph answer to your questions, but has grown into a full article as I wanted to clarify more and more. Enjoy the read. :P

Hypothesis
The goal of my research was finding out if tags have often been full, and if so, which level of tags have had an easier time to fill their ranks. The reason we wanted to know this was because we supposed that people would stop bothering to play the game if they couldn't get into a certain level of alliance, be it a top alliance, a decent alliance or training alliance. To complete the picture, I also considered the information about active non-alliance players.

Method
The data I used for alliances comes from two sources. Appocomaster was my primary source, he gave me the needed information for rounds 14-25 and 27-29. He also gave me the data for round 26, but this information was clearly erroneous because it was taken after havoc, while the data from r30 was missing completely. I used the PAwiki to fill the gaps. I think it's safe to say that round 30 is still fresh enough in our minds that these figures could not be tampered with, while the graph shows normal values for round 26. I am therefore confident that these two pieces of data are correct, despite not coming from the most credible of sources.

The information about non-alliance members comes from Appocomaster as well, even for round 26 (something worth keeping in mind). Information for round 30 is missing.

Because alliance names change on an almost per-round basis while the players do not (common knowledge), analysing the evolution of individual alliances would have been a waste of time. Furthermore, to categorise alliances I would have had to introduce a subjective element, in order to be able to decide which alliances were "better" or "worse than others.

Instead I opted to analyse rankings. This avoids the natural fluctuations in individual alliance's performance and member count, if alliance X doesn't play a certain round, a new alliance Y will take its place, often taking in a lot of alliance X's former members.

I initially refrained from putting ranks into categories for fear of losing information. However, when I displayed each rank separately, the graph became a unintelligible mess of colour. I mitigated the issue by categorising the ranks.

I divided into 4 groups of 5 because generally speaking, the top1-5 plays a separate game from the top6-10, and the top11-20 plays a separate game once again; I split this last group into 2 of 5 to keep consistency across the board and because I noticed that though in earlier rounds the top16-20 often had a reasonable number of players, in later rounds this was not the case. I did not wish to throw away this distinction. If you disagree with any of this, I would be interested in your reasoning.

it is worth noting that the peaks in rounds 22 and 27 coincide with these being free rounds. Though beyond the scope of this post, it worries me that the peak for round 27 is much smaller than the one for round 22.

Analysis
The data suggests that people prefer being in top alliances over being in decent alliances and that they prefer being in decent alliances over being in training alliances. It also demonstrates that top alliances do not have such strict recruitment policies as is often supposed.

http://img12.imageshack.us/img12/8968/image3wdb.png
http://img12.imageshack.us/img12/420/image1twv.png

It can safely be said that the bottom tags (11-20) have lost a significant number of members over the last 17 rounds, while the top alliances (1-5) barely lost any; the trend lines (not shown) are quite clear. The group closest to training alliances (active non-alliance player) shows an even worse drop, from 1851 in round 16 to 533 in round 29.

Conclusion
The conclusion we drew from the data was yet again that the alliance limit has had virtually zero impact on the game, or at least much less than other influences, such as whether or not a round is free, the fact that a certain alliance is playing or whatever other reason you can think of.

What worries me more than anything is the huge drop in non-alliance players. These are the people Planetarion needs, fresh blood, not the 1000 or so veterans that have been playing this game on and off since round 18.

1 May 2009, 12:24	#67
Mzyxptlk mz. Join Date: Aug 2005 Posts: 8,587	Re: Drastically lower alliance tag limit Alright, this post started as a two paragraph answer to your questions, but has grown into a full article as I wanted to clarify more and more. Enjoy the read. :P Hypothesis The goal of my research was finding out if tags have often been full, and if so, which level of tags have had an easier time to fill their ranks. The reason we wanted to know this was because we supposed that people would stop bothering to play the game if they couldn't get into a certain level of alliance, be it a top alliance, a decent alliance or training alliance. To complete the picture, I also considered the information about active non-alliance players. Method The data I used for alliances comes from two sources. Appocomaster was my primary source, he gave me the needed information for rounds 14-25 and 27-29. He also gave me the data for round 26, but this information was clearly erroneous because it was taken after havoc, while the data from r30 was missing completely. I used the PAwiki to fill the gaps. I think it's safe to say that round 30 is still fresh enough in our minds that these figures could not be tampered with, while the graph shows normal values for round 26. I am therefore confident that these two pieces of data are correct, despite not coming from the most credible of sources. The information about non-alliance members comes from Appocomaster as well, even for round 26 (something worth keeping in mind). Information for round 30 is missing. Because alliance names change on an almost per-round basis while the players do not (common knowledge), analysing the evolution of individual alliances would have been a waste of time. Furthermore, to categorise alliances I would have had to introduce a subjective element, in order to be able to decide which alliances were "better" or "worse than others. Instead I opted to analyse rankings. This avoids the natural fluctuations in individual alliance's performance and member count, if alliance X doesn't play a certain round, a new alliance Y will take its place, often taking in a lot of alliance X's former members. I initially refrained from putting ranks into categories for fear of losing information. However, when I displayed each rank separately, the graph became a unintelligible mess of colour. I mitigated the issue by categorising the ranks. I divided into 4 groups of 5 because generally speaking, the top1-5 plays a separate game from the top6-10, and the top11-20 plays a separate game once again; I split this last group into 2 of 5 to keep consistency across the board and because I noticed that though in earlier rounds the top16-20 often had a reasonable number of players, in later rounds this was not the case. I did not wish to throw away this distinction. If you disagree with any of this, I would be interested in your reasoning. it is worth noting that the peaks in rounds 22 and 27 coincide with these being free rounds. Though beyond the scope of this post, it worries me that the peak for round 27 is much smaller than the one for round 22. Analysis The data suggests that people prefer being in top alliances over being in decent alliances and that they prefer being in decent alliances over being in training alliances. It also demonstrates that top alliances do not have such strict recruitment policies as is often supposed. http://img12.imageshack.us/img12/8968/image3wdb.png http://img12.imageshack.us/img12/420/image1twv.png It can safely be said that the bottom tags (11-20) have lost a significant number of members over the last 17 rounds, while the top alliances (1-5) barely lost any; the trend lines (not shown) are quite clear. The group closest to training alliances (active non-alliance player) shows an even worse drop, from 1851 in round 16 to 533 in round 29. Conclusion The conclusion we drew from the data was yet again that the alliance limit has had virtually zero impact on the game, or at least much less than other influences, such as whether or not a round is free, the fact that a certain alliance is playing or whatever other reason you can think of. What worries me more than anything is the huge drop in non-alliance players. These are the people Planetarion needs, fresh blood, not the 1000 or so veterans that have been playing this game on and off since round 18. __________________ The outraged poets threw sticks and rocks over the side of the bridge. They were all missing Mary and he felt a contented smug feeling wash over him. He would have given them a coy little wave if the roof hadn't collapsed just then. Mary then found himself in the middle of an understandably shocked family's kitchen table. So he gave them the coy little wave and realized it probably would have been more effective if he hadn't been lying on their turkey.