EVE General Discussion

So they don't add anything to the game like I said. Null corps don't hang out in high sec, and have to go through low sec anyways, so no real reason to chase a nullbear into high sec, when you can effectively stop operations by battle in null/low sec. Like you said conflict would happen without it.

What the heck is wrong with you? You can do small scale PVP anywhere. Small scale PVP in high sec? The OP is about getting new players into larger scale activities, maybe even PVP like on the commercials, not small scale PVP . That is the silliest excuse for picking on new players I have ever heard. Besides, I remember a game without war decs. It also had small scale PVP.

I wouldn't mind new players having an area where they can do stuff where people don't have the free range of null sec to kill them while they do mining and PVE. Why not? I've known many players who started playing the game under those conditions. The game was fine then, and those players seem fine now. What benefit does war decing a new player do? What good reason is there to war dec a high sec corp? Only thing I've heard is to attack bots, which CCP could manage on their own.

And CCP should look at the risk v. reward issue you mentioned. It has been broken for a long time, and risk and reward is a perception so no one is really wrong if they argue on either side. If you have good numbers clearly showing how all riskier types of EVE activities pay off better than less riskier activities, I'd like to see them. I assume you don't have any evidence, and are regurgitating propaganda from a concept that died almost a decade ago. I really wish you could have experienced the old EVE.

Or in other words . . .



So, having two options for social interaction leads to more social interaction than only having one option . . . wow! Wuuuuuut?!

The quote doesn't say they want to discourage NPC corp membership. Nor does it really say they want to encourage player corp membership. What it says is that player corp membership isn't really an option for many players because they won't risk getting AWOXed by their corpmates, and that the change CCP is making to corporation mechanics will give such players the option to join a player corporation, albeit with friendly-fire DISABLED, but thusly still increasing their potential for social interaction by giving them an alternative to NPC corporation membership.

Compare:
"Hiking is fun. Swimming is fun. Bob won't swim with sharks in the pool. We'll let sharks be taken out of the pool. Bob will be able to have more fun."
"Hiking is fun. Swimming is more fun. Bob likes to hike. We should break Bob's legs and throw Bob in the pool."

See the difference?

Why might that be?

Why are CCP piranhas? What's wrong with them trying to increase player retention?

I have removed a troll post.


Reminder to keep on topic and be respectful. I've been in this thread too many times already.

I should have known procrastinating on the forums when I took a week off work to focus on study was a bad idea. It also doesn't help that it's for post grad statistics, especially with how its being discussed here.

First things first, there are two presentations that are relevant to this. The first is CCP Rise and the second is CCP Quant. The data sets that the both are discussing are observational sets. This is not experimental data, it will not be neat or easy to use and is by definition a little fuzzy. Fortunately we have this whole field of math dedicated to dealing with these kinds of problems.

Second, for the love of Bob please stop trying to make the data say things it is not saying. Just stop. It gets annoying. Please at least understand what the data actually represents and where it is coming from, how it was collectd and the methods that were used to analyse it.

TL;DR;

• CCP Rise didn't talk about NPC corps in his main presentation.
• CCP Quant's segmentation model didn't account for corp type, only behaviour.
• The segmentation model looked at tenure and game time, unsurprisingly those that did a lot in the game or were social stayed for longer.
• The decision tree did discuss corps and corp size for retention, this was after the player killed node. (Being blown up is MORE 'important' than if you are in a NPC corp)
• My takeway; get people to do more things faster to get them to stay.

The First Bit by CCP Rise (Ganking in first 15 days)

Right so first up they are looking at 80,000 users. And they are looking for deathers in the first 15 days and if the killers were nuked by concord et al. This group was split into killed legally, illegally or not. Retention was then looked at for each of these groups.

So from here in the talk we have 85% not killed (68,000), 13.5% killed legally (10,800) and 1% ganked (800). I am asssuming there was some rounding. And from here CCP Rise notes that Ganked were most likely to stay, followed by legally killed and then not killed. Thats it. Thats all the data says for this one. It doesn't say anything else. From here you might come up with a new hypothesis to test and a new test to design, but thats all (Sounds like it was a comparison of means I guess).

The Second Bit gets interesting by CCP Quant
https://www.youtube.com/watch?v=h-jfvjMoe9Y
bit.ly/199jryz

Now this one is interesting. Incidently correlation does matter for this sort of analysis, but it is not the main tools used. I believe he even mentioned Principle Componant Analysis (CPA) for dimension reduction and k-means for the actual clustering. After all what was presented was a multi factor model (Involving a bunch of fun dimension reduction and factor analysis. I suspect eigenvalues were involved too.). The aim of these techniques is to take known data, in this case behaviour stats, and use that to define and understand latent or unmeasurable attributes, like what kind of a player you are.

The bits of the presentation about segmentation are the most relevant to this discusion, and frankly it is very interesting stuff. This is the bit where we can look at player activity and groupings. You can also follow along with how different variables were used to define factors, and then how those factors were then used to identify segments (You may notice that kind of corp is not a variable by the way, these are behavioural variables).

Coming from the slides themselves, here are CCP Quant's key takeaways:



The part of the presentation about decision trees was also very interesting. The short of that one if that dying in a fire due to another player was a common factor in those that made it from the first three months to the fourth. However this one was also caveted.

Size of the corporation joined was another factor he mentioned as mattering for those that did get blown up (remember this is a decision tree). As well as mining for those who did not get their ship blown up. It seems that if you didn't explode, mining meant you were more likely to stay than those who were not.

However CCP Quant did not make any real hard and fast conclusions in his presentation on this. In fact he made a point of stating that this helped them identify new things to test and, I am assuming this part, based on variables they have identified to test for.

Honestly ya just might wanna lock this one down then. We've got to many combative people, myself included, large egos, stubborn "nu-uh"ers, and "well, my turrets bigger than yours, and throws harder shells at her".

The points on both sides have been made, discussed, beaten to death, and on and on.

I think wardecs would be much more interesting if they were a default of 2 days for 20mil

Yeah it would be nice to force people to be doing what they would normally be doing while they are wardeced but no.. Wardec allows you to interupt another corporations activities if you have enough firepower to scare them, it doesn't force them to line up and get shot unless their own greed makes them take the risk. With a smaller repeatable wardec it would allow a quick and over with show of force and for a much hated corp to be constantly in threat from it's many victims that want payback.

The smaller charge is less ISK wasted if the wardeced corp simply ducks out of sight and is more accessible to less thrifty corporations.

nothing wrong with me at all i never said anything about small scale PVP not being able to happen anywhere. i'm not making excuses for anyone picking on new players, i'm not a pick on anyone kind of person anyway.



because you can't create an area where you can make isk without risk that's why. there are many reasons for war decs you're trying to make out there isn't and that's just nonsense.



there is risk related to reward, how is bringing your corp into a wh, investing billions into it and reaping the rewards from the wh not risk VS reward ? you could be attacked after a few days and lose everything.
what about setting up in 0.0 and spending a lot of effort and time building your little empire and all the time reaping in the rewards of the area moons and so on only to be attacked and lose it all, this is not risk = reward?
i agree it's broken in some areas of the game but not everywhere.
there are many ways to make isk in EVE without any risk at all. so i've no idea why you'd assume i should have some metrics to back up a comment on the forums.
you really wish i could have experienced the old EVE? i'm playing since 2007 is that not old enough lol.

PS: actually played in 2006, in 2007 i decided to stick around.