Seasonal evolution of players in NS2
Blrg
Join Date: 2013-09-01 Member: 187580Members, Reinforced - Supporter, Reinforced - Gold, Reinforced - Shadow
Join Date: 2013-09-01 Member: 187580Members, Reinforced - Supporter, Reinforced - Gold, Reinforced - Shadow
Hi there,
For some time I have wanted to take a look at the monthly evolution of NS2 players in a graph. The intention was to see in perspective the pattern of decreasing player numbers over spring and the gain of them over autumn.
I reduced the comparison to the last 3 years in order to reduce the noise of player fluctuations due to other factors (novelty of the game, big steam sales, big updates, etc). Also, take into consideration that values of January 2014 have been truncated to ease the readability of the graph (values of that month are written at the bottom of the figure)
I just was interested in seeing this and since I did it I thought that I could share it with you.
For some time I have wanted to take a look at the monthly evolution of NS2 players in a graph. The intention was to see in perspective the pattern of decreasing player numbers over spring and the gain of them over autumn.
I reduced the comparison to the last 3 years in order to reduce the noise of player fluctuations due to other factors (novelty of the game, big steam sales, big updates, etc). Also, take into consideration that values of January 2014 have been truncated to ease the readability of the graph (values of that month are written at the bottom of the figure)
I just was interested in seeing this and since I did it I thought that I could share it with you.
Comments
Having slightly higher numbers is nice and all, but that information is not that relevant I think.
boring stuff:
Don't get me wrong, I like a lot that we have higher numbers than last year and that was the main reason why I did this. It kinda started as a response to the people who only says "the game is dead, look at Steamcharts"... because things are more complex than that
What I personally find more interesting is that the decay of players is way lower than what we had last year (higher retention through the bad months). Look how in January 2015 we had more players than January 2016 and how this changes in the following months.
That doesn't mean that next year we will have more people than now. We can't get that information from this figure, maybe less people leaving now means that less people will be coming back, giving us more stable numbers through the year. BUT if we have a steady influx of new players and the retention has increased (how it seems to indicate the figure to a certain degree) then yes, it means that next year we could have higher numbers than now (altough certainly not hundreds of thousands of concurrent players...)
I personally think, and I am being completely subjective here, that the better experience for rookies have improved the player retention of the game. And also very important (although a polemic topic) is that the continuous updates have make many people to come back regularly to check what the update changed. For the better or for the worst, people that wouldn't be playing the game by now are playing a little every week to check the changes, even if they don't like them
If only there was a way to quantify the volume of new players/month (I don't dare to think players ID are available, otherwise, one could tell everything, new players retention, players cycle of life ...).
Yes, we have this info.
Can you say something about it? Give us some feedback about the situation of the game.
Sure. We get about 75-175 new players a day. During the recent sale, we had 460-600 new players a day. During past sales we had huge spikes in new players as well. The way we've been measuring retention is pretty coarse... but the definition goes -- did they launch the game again within a one week period on a different day from their first launch. That has been at about 50%. I'm actually going to be looking into this data more over the coming days and seeing what else I can glean from it.
Besides Remi, I also have some of the UWE data and a bunch of hive data. If you have any questions you think I could answer with this data we could discuss it. I have been very busy lately, so be patient.
If someone cand send me an exerpt of raw datas, I would like to try to work on player lifcycle.
For that, one need ID's and timestamps.
Athough I'm more used to relationnal Data Bases and SQL, I can also quite get some snip out of Excel.
No way I intended to disregard what you got out of the datas and the work it takes to rfine them before they can be of any use.
<span style="color:orange">PM</span> sent <= Actually not <span style="color:orange">that kind</span> of PM