Ganking and PVP: Numbers in perspective

[Kezrai_Charzai]

@Xucca

@Kezrai_Charzai

ganked group are the most likely to stay subscribed afterwards

I’d like to look a bit deeper into this.

Is there any context?

I linked the actual presentation above, here’s a link to the post that contains it:

Ganking and PVP: Numbers in perspective

So yes, when I referred to it as “ganking makes people stay”, it’s because that’s the interpretation PvP players have been putting on it for 7 years now. What CCP actually said, specifically, was that people assumed ganking drove people away, and so they looked at the data, and found that in a very specific subset of “ganks” looked at, that people who were ganked were “slightly” more likely to continue playing than people who died ‘legally’. Without any numbers, and with a fair bit of fuzzines…

The part where they go over the killed/legal/ganked numbers is in the first few minutes so is easy to catch. And yes it’s a Fanfest 2015 presentation so you may not be interested.

The ‘context’ is they were looking to see if suicide ganking drives players away. They checked 80,000 new accounts over first 15 days of play. Something like 85.5% lost no ships, 13.5% lost ships “legally”, and 1% were “ganked”. And the “ganked” group were “most likely to stay subscribed” according to the presentation.

There’s a lot of interpretation room there because it’s unknown how they chose the 80,000, how the “legals” were killed (suspect baiting, for instance), or whether they had any way to differentiate between actually new accounts, or new accounts created by existing/previous users.

@Xucca

Being ganked is not a primary reason for quitting.

So I had actually assumed, when I started gathering numbers, that the numbers would in fact show that ganking isn’t a huge problem. It may not be a problem, but turns out the number of ganks is quite a lot higher than I originally expected them to be.

(Yes I’m aware you don’t like numbers analysis, however with half a million players per year cycling through EVE I think a numbers/statistical approach has validity. As does personal experience.)

@Xucca

A very different thing is, that there are not many alternatives regarding PVP

This is one of the problems I’ve been pointing out forever. If EVE is a “PvP centered” game, why is there so little PvP relative to total game activity? Why are there so many players going well out of their way to optimize their “high sec ganking” techniques? It seems like the worst kind of low-hanging fruit for a game supposedly built around PvP.

[Aiko_Danuja]

@Lucas_Kell

gankers refine it down so they can blap newbies to death

Every ganker I know has zero interest in blingy haulers, freighters, and faction battleships. They are only interested in blapping new players, dragging them into a chat channel to mock them and humiliate them and emotionally torture them. Most gankers use an overview which hides any character over 30 days old.

[Ramona_McCandless]

Bew hew its too hard to gank gankers.

[Aiko_Danuja]

If it’s so easy, why don’t you do it?

Oh wait, you don’t even play the game.

[Aiko_Danuja]

You don’t even play EvE.

It was clearly too challenging for you.

[Ramona_McCandless]

So why is he still here?

Isnt trying to make people quit against forum rules or anything?

OMG if it isnt please tell me, Ive just had a great idea.

[Aiko_Danuja]

He’s still here because he has nowhere else to go.

[Ramona_McCandless]

He’s looking for something “even more unique”.

[Aiko_Danuja]

You would think a guy who has mastered the “simple” mechanics of EvE Online would be able to muster more than a couple hundred mediocre kills over the course of a fifteen year career. This casts doubt upon his claims.

[Ramona_McCandless]

Well I mean he mustve got bored with this boring game and drifted off half way through playing.

Thats why he never posts on the forums of the game anymore, too boring.

[Aiko_Danuja]

EvE is solved. It is far too easy and simple for a brainiac like me. I could easily master the markets and the PvP/PvE interactions in order to establish myself as the leader of a dominant alliance… if I wanted to. Your puny activities are far beneath my potential. I am a great genius with many exciting important things to do, which is why I continue to post in the forums of a game I quit playing years ago, mostly to remind everyone about how easy the game is and how boring and how successful I could be if only, if only!

[Uriel_the_Flame]

I think he might be here specifically for you (Aiko and Ramona), he enjoys these back and forths and can’t resist partaking in such just to experience your presence. I think this might be a sign of true love, no matter how many times you belittle him, he just keeps coming back and wanting more.

[Ramona_McCandless]

Thats what I thought about BF2

[Kezrai_Charzai]

Just wanted to post a wrapup here. This little ‘quickie review’ went very sideways on me. I had good input from several sources on issues that affect any analysis of high-sec ganks from Zkillboard (other people have tried this before).

Once it was clear I couldn’t easily pull well-defined data I enlisted a friend who works on a type of heuristic data analysis project to see if we could come up with some better “guesstimates”. Issues that arose:

• Should corvettes count? Pods?
• How to reflect ‘newbie’ ganks vs. ganks of T2/T3/exhumers/freighters/Marauders etc.
• What counts as a gank vs. non-gank combat? Can a 1v1 be called a gank?
• Does an unfitted (or bait) ship count for a gank? (Since it was clearly a throwaway expendable or bait)

For this type of overview (raw size of the ganking issue), my approach was that a ‘gank’ was any legitimately fitted ship engaged in PvE activity in high-sec, being killed by aggressors it had effectively zero chance of winning against.

We looked at numbers in the solo, 2-4, 5+ and 10+ brackets. We looked separately at the records of prolific gankers and gank corps. We looked at the activity of high-gank systems. The problem was drawing all this data together - and I drew the line at writing code to pull our own data and sort our own database because of limited time.

To get truly useful data we’d need CCP-level access: to differentiate which kills were legal, semi-legal (suspect baiting etc.) and illegal (Concord responds). It would also be best to look at the comparative age/SP level of the actors, and how long after how many ganks occur did activity cease on the account. Total cost of the gank relative to targets total wealth would be nice too.

Best guess from what we looked at is, the actual size of the total high-sec ganking issue is likely in the range of 3-7 times larger than what gets listed as ganks on Zkillboard. (Again, no fault to Zkill on this - digging further showed just how messy this problem can be.) My own personal ‘best’ estimate showed about 4.5 times as many ganks.

Compared to OP, this would move high-sec ganks to about 36,000 per year, 24,000 pilots, losing @60 trillion ISK per year. This would also move ganking up to around 4-7% of all PvP activity in high sec. (Again, very rough estimate.)

My original intent in the OP was to get a handle on the size and importance of the actual in-game ganking as opposed to the scale and passion of the discussions about ganking. I freely confess I had assumed the discussions were making a mountain out of a molehill. That ganking isn’t as big an issue, and doesn’t drive as many players (especially newbies) away from the game as most discussions revolve around. In the scope of “Top 10 problems facing EVE player retention” I would have placed ganking near the bottom - and potentially not even in the top 10.

The “likely” higher numbers would indicate that the discussions are more relevant than I thought. Although I found little “clear newbie” ganking, the total high sec ganking on mid-high value targets is significant. It means that EVE could potentially be losing in the range of 5-10,000 players per year over frustration from ganking. (The “EVE value” of any player who might quit over a gank is a separate discussion.)

The most important aspect of this would be that these are mostly players who have already committed to EVE beyond the first couple months, and played enough to start getting some value built up.

For me personally, this review has moved ganking from “low in the top 10 problems” up to maybe the 7 position. Maybe 6.

My thanks to all those who contributed info and suggestions and pointed out issues along the way. I may make another post or two just to reflect some numbers and further thoughts, but for now I’ve looked as far as I intend to.

(Edit: adding some references to high-sec for clarity. Keep in mind that it’s also valid to discuss whether ‘ganks’ can occur in low, null or WH space and what effect they may have on retention.)

[Io_Koval]

My best guess is that without data, best guesses aren’t going to change anyone’s opinon. Even with data, people still will fit the numbers to their opinions anyway.

Thanks for the effort you did put looking into it, though.

[Altara_Zemara]

@Kezrai_Charzai

Compared to OP, this would move ganks to about 36,000 per year, 24,000 pilots, losing @60 trillion ISK per year. This would also move ganking up to around 4-7% of all PvP activity. (Again, very rough estimate.)

I find 100 or so a day…which is pretty much the same thing…to be a good estimate. But…

…this then brings to mind a point Macgybo made in his own recent estimates. He calculated that 15% of his ganks were ‘1 day old noobs’. But Macgybo only ganks targets in the hundreds of millions of ISK range. What is a genuine 1 day old noob doing carrying 1bn ISK worth of skill injector to Jita ? When I was a day old I’d never even heard of Jita.

Clearly, the entire ganking stats are skewed by the fact ( and it is one of the few things one can definitively state is a fact ) that a large portion of ‘noobs’ are not new at all. A genuine noob is still coping with running one character of one account. Whereas an oldbie creating Alpha accounts can conjure up 3 ‘noobs’ in the space of minutes.

Until such time as one can definitively show who is genuinely noob or not, stats about noob ganking are bound to be skewed too high precisely because a large portion of ‘noobs’ are not new at all.

Bear in mind that it only requires for each player to create one new character during their first year in Eve for the rate of genuine noobs in Eve to be equal to the rate of non-genuine noobs over that year. In other words even with a minimalist consideration, at least half the ‘noobs’ in Eve are not ‘new’.

[Scipio_Artelius]

@Kezrai_Charzai

This would also move ganking up to around 4-7% of all PvP activity. (Again, very rough estimate.)

Not sure of the math here.

Even assuming that an estimate of 36,000 ganks per year is correct rather than a guess, then from the monthly economic reports for the last couple of years:

Year	Total PVP Kills	36000 as percent of all pvp activity
2020	5,674,137	0.63%
2021	4,899,115	0.73%
2022	2,975,091	1.21%*

* 2022 contains data only for Jan-Aug so far, so the adjusted percentage on the whole year would be 0.81%.

[Kezrai_Charzai]

@Io_Koval

best guesses aren’t going to change anyone’s opinon. Even with data, people still will fit the numbers to their opinions anyway.

Yes, for some people, any data that conflicts with what they want to believe is irrelevant. For other people, any useable information at all is preferable to none.

As said in the OP, I was simply trying to get a handle on whether the discussions of ganking were trying to make a big issue out of a comparatively small one - not change any opinions.

While taking a stab at this, I learned quite a few things I hadn’t before, and changed my own opinions - which is really all you can hope for in these things. I just posted it for other people to review because I thought it was interesting.

@Altara_Zemara

Until such time as one can definitively show who is genuinely noob or not, stats about noob ganking

True enough. Which is why I came to the conclusion that I couldn’t make even an attempted valid estimate without access to CCP-level data - which would include some information on actual newness of characters or not.

Also, the analysis itself isn’t focused on new or “noob” players at all - I point out several times that they apparently make a small portion of ganks. It’s about ganking - not noob ganking. And ganking a 2 billion freighter from a 6 month old character might actually be more relevant than any effect on true noobs.,

@Scipio_Artelius

Not sure of the math here.

Well, thought I had made it clear more than once that the ganking review was looking at high sec only. However it won’t hurt to make that more clear. I’ll edit a few places to clarify.

Also:

• In general you shouldn’t mix data sources. If you start with Zkill numbers you should compare to Zkill totals, not MER totals. They won’t be using the same data set.
• For my own purposes in this review, I discounted MTUs/depots/structures but left in capsules and shuttles (although it may have been valid to remove those).

[Scipio_Artelius]

@Kezrai_Charzai

They won’t be using the same data set.

They definitely aren’t. The MER includes all PVP related kills while zkillboard doesn’t.

zkillboard stats include only PVP kills that zkillboard has access to, but zkillboard has a whole lot of additional NPC related kills that the MER doesn’t include. To get the most complete picture outside of CCP, both sets of data are needed (and still then, for some ships, it’s difficult to be confident in analysis, while for others, confidence can be close to 1 on the results when combining both datasets).

However, the edit to restrict to highsec is a good clarification (and not clear even from the OP whether only highsec data is considered, because the OP certainly includes numbers that involve non-highsec pvp losses).

But, if you are only using highsec, do you have the numbers handy? (it’s easy to extract from the MER, so I assume you did it to get to the 4-7%).

[Vokan_Narkar]

@Kezrai_Charzai

This would also move ganking up to around 4-7% of all PvP activity in high sec.

Implying that ganking is PvP. Which it isn’t. Ganking is seal clubbing (whatever that means).