Wednesday, August 13, 2014

Is EVE A PvP Game?

A couple of weeks ago, Mike Azariah asked an interesting question on Twitter...

He followed that up with a blog post last week with his opinion and asking for others to try to sum up what type of game EVE is in one or two sentences.  I think answering the first question, whether EVE Online is a PvP game, or a game that has PvP as a part of it, goes a long way in determining how people look at EVE.

Perhaps not surprisingly, as someone who has appeared on four killmails over the past three years (two kills, two deaths), I believe that EVE is a science-fiction world simulator that uses PvP as the driving force for its economic model, and not a PvP game.  But is that opinion just formed from my unusual playstyle?  I don't think so. 

CCP, despite relatively recent efforts to hide EVE's subscription numbers, still is one of the most open gaming companies where game data is concerned.  Players have taken the information from EVE's APIs (and now public CREST) and created web sites that provide game data to the community going back years.  I'll use two trusted sources of information for this post: Wollari's Dotlan Maps and Chribba's

The first question I'll address is: what is the ratio of player ships killed to NPC ships (including rogue drones) killed.  If EVE is a PvP game, I would expect a very low NPC to player ships killed ratio.  If EVE is a game (or a simulation) that only has PvP as an element of the virtual world, I would expect a high ratio.  The results may surprise some non-EVE players.

In July, the ratio was 296 NPCs killed for every ship a player lost.  I use the word "lost" instead of "killed" because the data available at Dotlan does not separate out ships lost to other players versus ships lost to NPCs.  The only region that had a ratio I would expect from a PvP game that contained NPCs is low sec, with a ratio of 27:1 in July.  The other two areas, high sec and null sec, both have ratios I would expect in a PvE game.  Indeed, those K/D ratios are a lot better than I had playing my ranger in Everquest 2 on a PvE server.

Because peace has recently broken out across null sec, I added the period from January 2013 to June 2014 to the chart.  That period encompasses two major null sec conflicts, the Fountain War in the summer of 2013 and the Halloween War in the autumn and winter of 2013-2014.  The overall NPC kill to player death ratio was 248:1 during that time period.

I've included the above graph to show the impact of the two major wars on the NPC kill ratio.  Interestingly enough, the blue line representing null sec is almost the inverse of the average concurrent user mark.  When the ratio goes up, the ACU goes down.  But even with that relationship, the lowest the ratio went during the period was down to 215 NPCs killed for every player-owned ship lost.

Having explored the relationship between NPC and player deaths, I next decided to try to match up the data from Dotlan and the data from  What does the average play session for the average player look like?  I know that for a very long time that Xfire showed that the average time spent playing EVE for a player was between 3-4 hours a day.  So I decided to calculate for a 3-hour play session.  I then calculated the total number of hours played each month using the average concurrent user numbers from EVE-Offline and calculated the average number of NPC kills and player ships lost per account hour played for each month.  Once I had the average per hour, I multiplied the result by three to get the average numbers for a three-hour play session.

Surprised?  According to the results, players kill an NPC on average every five to six minutes.  Killing another player's ship?  The average player in EVE doesn't kill another player's ship during an average play session.

In the first graph, the average number of ships lost looks like zero, but that is not quite true.  This last graph shows that since the ship rebalancing that arrived in the Rubicon expansion, the average number of ships lost in a three-hour play ranged between 0.12 and 0.17 ships.  I should remind everyone that the figure doesn't mean that is how many ships are killed in PvP.  Because Dotlan does not provide a breakdown between PvP and PvE losses, the number of PvP kills per 3-hour play session is less.

When I think of a PvP game, I think of a game like Planetside 2.  If you are not fighting other players, you are not playing the game.  With a ratio of 250 NPC deaths to every player ship killed, EVE players do an awful lot of fighting against non-players.  Indeed, many players rarely, if ever, fight other players.  But for me, the final argument is this: can I go up to someone and, with a straight face, tell them that a game in which the average player in a 3-hour play session will most likely only kill NPCs and not kill another player or have another player kill him, is a PvP game?

I can tell someone that EVE Online has areas specializing in PvP and give advice on where to locate a good fight.  That, I have no problem with.  But after looking at the numbers, can I honestly tell someone EVE is a PvP game?  No.


  1. On an even more simple basis, I say a PVP game is one in which you HAVE to PVP. I can play EVE since 2006 and never have to fight another player. Some may argue that mining and selling minerals or market trading is some form of PVP, but try telling THAT to someone outside of EVE with a straight face.

  2. I'm not sure I should be surprised by these numbers, but I am. Granted, a large number of PvE (soft) targets are always available, but the *ratio* tells a hell of a story.

  3. I am as much in favor of stressing PvE's relevance in EVE as any other hisec carebear, but honestly I think that something is missing from those graphs; namely, how easy it is to kill a NPC, compared to how easy is to kill a player ship, and what's the mortality rate like.

    NPCS are on a essentially infinite supply and are easy to kill in large amounts, so obviously they will die a lot more than player ships, which are on limited supply and tougher to kill. A single mission runner could kill 400 or 500 NPCs in three hours, whereas a ganker maybe would get to gank half a dozen miners in that same time.

    I think that it would be more informative to see how much time is spent on those activities (how many time killing NPCs, and how many time killing players). It *still* would show that PvE is the single largest activity in the game, but likely not by a 250:1 ratio...

  4. If I were being facetious...

    The way to win FW is to shoot an NPC and circle an NPC station.
    The way to win Null-sec is to shoot at structures that can't shoot back, and then farm NPCs faster than your competitors.
    The way to win a hi-sec war dec is to deny the other person from shooting NPCs.
    People living in wormholes exist BECAUSE of the NPCs that live there.

    Of course that's not the whole story, but I doubt there's any capsuleers out there that haven't shot a red cross in their career.

    A good question would be: What would Eve be like if it was a pure PvP game?

  5. I don't really understand why you would think these metrics are informative in any significant way. Even if you could track the actual time spent in PVP versus PVE across the player-base it wouldn't say much--e.g., because "time spent" is going to be qualitatively different across different types of activities. Of course, EVE is a game which allows for a huge variety of play styles--including stuff like lost drone reclamation etc.! As a bit of sociology of the EVE community, it would be nice to see how every current player of EVE would answer this question. Who knows, maybe a majority would say they see EVE (or they primarily play EVE) as a PVE game.

  6. Heh, this strikes me as sort of the Douglas Adams argument that the population of the universe is zero. Theoretical infinite number of planets, but only a finite number support life, and any number divided by infinity is zero, ergo there is no life in the universe.

    And you left our the market, which I would argue is economic PvP. Blowing up spaceships isn't the only measure of players in conflict. You can say, "Well, even carebear WoW has markets," but what games has all the production and resource harvesting requirements behind it?

    Anyway, I will let you explain how EVE is not a PvP game to the guy whose freighter gets suicide ganked.

  7. Missions and sites have 50 frigates and then 10 BS which actually provide the bulk of the value. Meanwhile players do their best to avoid losing anything at all and attempting to get a fight from someone may be an hour long process in itself (not to mention that PVP at high levels gets TIDI'd and NPCing never does).

    As long as that's true I think this methodology is poor support for your conclusion.

  8. PvP provides no income of importance. Even if a player gets a great mod, they are more likely to use it themselves than sell it. So there is a well known army of alts supporting every PvP oriented player- either directly or through their corporation/alliance. Don't we all know that PvP player with the industry alt, the Incursion alt and the mission alt? At least two of those alts are killing scores of NPCs to support the PvP habit of the main character. Let's see your analysis when you take the bread winner alts into consideration please.

  9. Except, of course, we're not talking about city crime statistics and what part of town might or might not be dangerous, we are being pedantic about whether a video game where another player can shoot you at any point after you undock is really a PvP game.

    Saying a bunch of people didn't get shot really isn't all that relevant to the mechanics of the game, because I can shoot you if the mood strikes me. That pretty much defines a PvP game, that I can act against you without your consent.

  10. Eve is whatever you want it to be. For some, sure it is PvP, though I have no clue how anyone can afford to strictly PvP.

    But for most, it is PvE primarily. Anyone saying different is delusional.

  11. Maybe it's not a fully pvp game but it's a highly competitive game. You're always in competition with someone else even when you're ratting. So meh.

  12. So really the issue here is that we're not sharing definitions. One group says that if you're spending 80% of your time doing PVE then it's a PVE game. The other group says that since the whole time you're doing PVE there is the risk that PVP could happen, thus it's a PVP game. Both have reasonable (but opposing) definitions and thus can use the same date to draw opposite conclusions.