Sunday, January 18, 2015

Is PVE Safer in High Sec, or Null Sec? - Some Evidence that Null is (Much) Safer

Is it more dangerous to do missions in high security space, or rat in NS? One might well assume that NS would be more dangerous (and presumably more profitable as a result). I mean, that's the point of risk versus reward in HS versus NS, right? In fact, however, this common belief is probably wrong, especially as of 2014-2015. To test this common belief, I did a small study of NPC kills in top systems in HS versus NS and compared them with PVE ships losses and isk values in those systems. If ratting in NS is more dangerous than missioning in HS, what we should clearly see from the data is that more ratting ships are dying in NS compared with mission ships in HS. However, this is not what we find...


Here is how I did the analysis: I started on Dotlan's immensely useful summary of statistics for various activities in EVE. I focused on NPC kill statistics from December 2014. I took the top 8 null sec systems for NPC kills, which in the month of December combined to a total of 4,133,166 NPC kills. These top 8 systems, conveniently, also represent different areas of space, including (at the time) CFC space, NA space, B0T space, and PBLRD space. Importantly, I included carrier losses in these systems as a "PVE" loss unless the carrier losses were part of an obvious fleet fight (of which there was only one such fight it appeared), even though some of the carrier losses were likely not ratting carriers

Then, I took three of the top high security systems for NPC kills in the month of December. I picked the three systems whose combined NPC kill count most closely matched the combined kill count of the top 8 null sec systems, yet also represented different regions of high security space (so as to help avoid biasing the sample, given that PVE systems near Jita have far more PVE ship losses than systems further away--if you want to see how much you can bias the sample toward high sec, only include PVE losses in systems around Osmon and you will see high sec numbers jump way up). These three systems are: Sheroo, Inaya, and Tvink, for a total of 4,164,725 NPC kills. So, the sample from high sec has just slightly more NPC kills (31k more)--not enough of a different to be statistically significant.

I then compiled the number and isk value of losses from each of these systems for the month of December. I only included losses of obvious PVE ships (so, I didn't including losses of mining ships—incidentally, there were far more losses of mining ships in these high sec systems than there were in the null systems, as one might expect). 

Methodologically, this little study has weaknesses. Ideally, a much better study would be to look at far more systems over the course of a year in both NS and HS. If anyone wants to do that, go ahead!

Here are the results, with both number of PVE ships killed as well as the total isk value (in billions of isk):

(One interesting thing to note is that the bulk of PVE losses in these null systems came from CFC systems. This suggests that PVPers might hunt CFC space more frequently than NA. space in particular, or that the CFC might be worse at defending their ratters than other sovereignty holders--or both.) 


In these high security systems, 69 PVE ships were destroyed in this month for 4.16m NPCs killed, while only 44 PVE ships were destroyed in those 8 null sec systems, and for a mere value of 23.8b isk for the same number of NPC kills. The most common PVE ships to die in HS were tengus and marauders, while in NS the most common PVE ships to die were (by far) ishtars, gilas, and VNIs. Interestingly, the average marauder to die was worth more isk than the average carrier to die--so if you like getting lots of big kills, gank marauders in HS, don't waste your time fishing for carriers in null. Plus, there were only a few carrier losses, while there was a few dozen marauder losses.

If we can extrapolate from the mere isk lost, we could say that high security space is almost 300% more dangerous for PVE! 
However, besides the mere fact that, as the numbers suggest, more PVE ships worth quite a bit more isk die in high security systems than null, what conclusions should we draw?

There are many general conclusions one might draw from this data set about risk versus reward in HS compared to NS, assuming the numbers are real and generalizable:

1. Risk versus reward in PVE activities in HS and NS are fine as they currently are. (In other words, the numbers above are "where they should be").

2. PVE in NS should be less dangerous while PVE in HS should be more dangerous.

3. PVE in NS should be more dangerous while PVE in HS should be less dangerous.

4. PVE in both NS and HS should be less dangerous.

5. PVE in both NS and HS should be more dangerous

And so on...

Let me put my cards up front and put my argument in as simple terms as I can. I have no real opinion about risk and reward in high sec. The conclusion I favor is this: PVE in null security space should be more dangerous. In fact...

PVE in null security space is broken.

I don't much care about risk versus reward in HS, because the risk of PVE in high security space does not seem broken, whereas the risk versus reward in null sec clearly seems broken based on the numbers above. Lots of ships are dying in HS--especially expensive ships--whereas significantly less ships in NS are dying, and at far less of a isk cost. Isn't null sec supposed to be more dangerous? I think it should be. (Disagree? I am open to hearing counter-evidence or reasons to the contrary.) If you are more likely to lose a tech 2 fit, plain-old marauder while doing missions in HS than you are to lose a carrier while ratting in NS, something is wrong.

That all said, I am also not opposed to conclusion 5--that both HS and NS should be more dangerous. Risk is good for EVE, and if anything the game needs more of it, everywhere. But, clearly, the bigger problem is with NS.

Finally, that all said, I think the attitude some players have--that HS is somehow this "safe" area while NS is somehow this extremely "dangerous" place--is completely off-track, and, at best, an out-of-date left-over from EVE Online pre-2010. HS is, in fact, far more dangerous than it used to be, while much of NS--partly due to the incredible intel and coordination of many null sec empires--is far safer than it has ever been for activities like PVE (and mining). When you look at the numbers, there's little other conclusion to draw. 

Well, assuming you agree with me that the status quo is NS needs to change, what changes should be made to fix the lack of risk? Making intel harder to acquire is one obvious way, through mechanisms such as local chat delay when entering systems. This is a big question, and probably a good one to ask those running for the next CSM...


  1. There is an alternative: people in HS are generally dumber. You won't see anyone ratting in nullsec with 30+ reds in local. You see idiots keep ratting/mining with 30+ flashy gankers in local linking kills and praising James 315.

  2. High sec dwellers tend to go for max DPS set ups or DPS plus max repair. Little or no buffer tank. If you want to attract a gank, just add shiny....

  3. Since the perception is that high sec is safer people take more risks in high sec.
    In null I don't rat with neutrals in local, in high sec I have no choice but most neutrals won't be looking to gank me either.
    In null I assume it's only a matter of time before I make a mistake and get caught. In high I don't expect to die so I can bling fit my ship more.