That Darned PvP Stuff Again

The issue — or, I should say, the several and various critical issues — of Player versus Player combat in persistent world role-playing games is coming to the fore again, in all its contentious glory.

For some strange reason, this seems to be another area where persistent world developers and, especially, designers refuse to learn the lessons of past failures. In fine examples of the old saw, "Hope springs eternal," each new designer and/or development team is just positive they have the one true solution that makes non-consenting PvP work. Game after game touts itself in development as the answer to PvP, that they are the ones that will do it right first.

I’ve been hearing that since about 1989 and have yet to see it actually work in a persistent world. In fact, every time I’ve seen non-consenting PvP tried in various forms, it has been a miserable failure. Except once: Bartle’s MUD II, which can be counted as the for-pay online adventure/RPG game that started the industry. Why it worked in that game would take some pages to relate and that makes it fodder for another column.

So let’s explode some myths. Note that the following is relevant mainly to for-pay games and especially large-scale subscription games. In a non-pay game, players are willing to put up with just about anything, mainly because it’s easy to walk away from something if you don’t have to invest any money in it.

And when I use the word ‘non-consenting’ in this context, it means simply that at least one party to the combat has no choice of whether or not to fight; he/she is involved whether they will or no.

Myth One: "There are many hundreds of thousands, maybe millions, of players who want PvP in general, and unrestricted PvP in specific."

I had this discussion (again) recently with someone in charge of designing a fully PvP game and this hoary point came up - again. As we used to say when we were arguing as kids, "Oh, yeah? Prove it." The premise here seems to be that, because so many first-person shooters with Internet capability are sold, there must be a huge market for PvP in persistent worlds. More on that in Myth Two.

Quite aside from the fact that the argument mixes two different markets and player demographics, which is making the mistake of the old "apples and oranges" assertion, my experience has been: Not when they have to pay money and invest time for it. The available evidence would suggest just the opposite about non-consenting or unrestricted PvP:

Some 80%+ of all game play in commercial persistent worlds is either non-PvP or fully consenting PvP (duels, etc.);

Of the several third-generation persistent world RPGs to be released since 1997, only one featured raw, unrestricted PvP as part of the main servers (Ultima Online). When they reined back the killers and instituted mirror lands, one PvP and one not, in late 1999 and early 2000, UO’s subscriber base saw a resurgence. To be fair, institution of new content scenarios, improved customer service and an enhanced program to train new players had something to do with it, too. It is indicative, however, that about 80% UO’s game hours are now in the non-PvP areas of the game;

One relevant online survey of persistent world RPG gaming preferences can be found at http://www.andreasen.org/bartle/stats.cgi. That survey notes that as of October 20, 2001, only 21% of MUD/MMORPG players rate themselves as primarily killers, out of over 87,000 who have taken the survey (and note that surveys of less than 3000 people determine what shows we see on American TV, shows that are sometimes watched by tens of millions). Put another way, 79% of players rate themselves as primarily socializer, achiever or explorer players. When players rate themselves in various combinations of the Bartle player styles, the ‘Killer’ category remains sparsely chosen.

I would assume that these 87,000+ survey respondents are for the most part the hobbyists, the people most likely to play these games anyway. The hobbyists are our bread and butter; the subscription side of this industry can’t survive without them right now. If even they lean so far away from PvP and make up the smallest part of the market, why would anyone construct a game aimed at them?

Myth Two: "Those statistics are wrong. There is a huge market for unrestricted PvP; Look at the all those copies of Red Alert 2 and Quake that were sold, and those are all PvP games."

This is like saying, "There must be a huge market for stinkweed; look at all the oranges people buy!" There are critical differences between the two types of games:

1. Retail Hybrids such as Red Alert 2, et al, are not subscription-based. That makes them a different market right there. The minute you ask people to cut loose with money on a subscription basis, the whole nature of the game changes. In fact, the nature of the Hybrid customer can be described as "I want it for free or less.";
2. The Hybrids are not persistent. One of the main reasons people cut loose with additional fees for today’s MMORPGs is because of that persistence;
3. Hybrids are not massive-multiplayer. The ‘MM’ part of MMORPG is part of the draw. Right now, the most people you can normally shove into an Internet session of a Hybrid is 32 players. It is interesting to note, however, that the more people you can host in a Hybrid play session, the more popular the game seems to be…
4. Hybrids are zero-sum games, meaning if I win, you have to lose. After a while, 99% of the players realize that the other 1% will always beat them. They will then find another freebie game to play.

Myth Three: "Yeah, well, the reason non-consenting PvP is so unpopular is that no one has done it right yet."

This is one of my favorites. I believe I have seen every permutation of PvP possible in the last fifteen years and the only variations that have ever worked in a large commercial setting were of the consenting variety. One might present the argument that an exception was PvP in games on the old online services, such as GEnie, CompuServe and AOL pre-1996. And taken at face value, it certainly looks that way; several such games allowed players to attack each other without warning and that the PvP did draw together the players to punish the killer (see Myth Four below). We did, indeed, see this kind of activity.

Part of it was the emotional attachment to characters and social groups that we see in these games. However, there was a more important limiting factor that is often ignored in this particular debate. That limiting factor was the cost of playing the game, which was anywhere from $3 an hour to $19 an hour, depending on the year, online service and the time of day the play occurred. When a player could easily invest over $5,000 to build up a character over time (and I know of people that used to spend $2,000 per month to play on GEnie), you bet they gather like white corpuscles to destroy the few PKers and griefers who could afford to play. Money was the great equalizer, not design or style.

That emotional attachment to characters has not gone away and designs are still stuck in the hourly mode, in that they require many hours of play to advance. In these days of flat rate monthly fees, however, anyone that can push a lawnmower can afford to play and there are more griefers around. The reaction to this change has been a demand for less non-consenting PvP, not more (see Myth One above), to allow players to choose when and if to risk a prized character.

Yet today’s designers continue to ignore that important trend and to try to convince the gaming public that they, and they alone, have discovered the Rosetta Stone that allows them to decipher how to do non-consenting PvP in a way that everyone will want to take part in. The current flavor of the month is Faction versus Faction and the variations thereof, which is touted as promoting team warfare and, supposedly, team protection. The theory seems to be that if you join a Faction, you’ve automatically consented to the PvP. And there is justification to that argument.

However, I think the point will be moot when publishers see that "Faction versus Faction" is really just another phrase for "piling on" and that that drives customers out of the game. Sure as shootin’ what will happen is, one Faction will be far more organized than the others, gather teams of 5, 10 or 15 players and go hunting for smaller teams or solo players in a competing Faction, gank them fast and move on. They won’t be looking for a fair fight; they’ll be looking for victims. That’s what always eventually happens, because PvPers as a class of gamer are out to win. They don’t engage in combat with other human players to lose, after all, and if what it takes to win is to create unfair odds, that’s what they’ll do. Why shouldn’t they? That’s how you designed the game, isn’t it?

Myth Four: "Non-consenting PvP is necessary to create conflict, which creates drama, and that conflict/drama draws players together."

This argument derives from the well-known theater, movie and TV concept that ‘there is no drama without conflict.’ What it ignores is that A) online games are not movies or TV, and B) conflict does not have to be an immediate life or death choice to create drama. Designers tend to ignore this, because it is so much easier to just set up PvP and let ‘er rip. If persistent world designers wrote TV shows, you’d see sitcoms in which the parents discover their child disobeyed their orders to not hang out with the local bad boys (the conflict) and, instead of spending the show deciding how to deal with this thorny issue (the drama), they just gank the kid and then go steal a new one from the neighbors.

The fact is, this Myth is just the designers rationalizing their own inability to come up with something better than "Ganking for cash and prizes" with the development and management resources they have available. If you stop and think about it, more drama is created if the conflict merely suggests that life and death might be on the line later if the conflict is not resolved by other means. And far more drama is created by taking the player just to the edge of death, then jerking him back from the precipice.

The main problem as I see it is that designers seem to want to be directors, not designers. Unfortunately, what works really well in TV and the movies (or for six people around a tabletop) doesn’t work very well at all in an environment that includes tens of thousands of heroes looking for the tools to create their own legend and story. Participants want to have an effect on the world; observers just want a ripping good yarn. It requires far more work, organization and execution skills to engage a participant than it does an observer.

Development teams just don’t plan for the resources to manage that kind of world, because, dammit, it’s their world! I’m designing a moving, heart-gripping story; get with the program! So what we get is the same old thing, over and over. Of course, the serial numbers and model name are filed off, polished and presented as "new and different!" Just as this year’s crop of sitcoms is touted as new and different, even though they are the same old jokes we’ve been hearing on TV for the last thirty years.

And of course, in the grand tradition of online games, we’ll have to wait for most of these extravaganzas to crash and burn before we start to get some real work into creating a new persistent world that allows players to create their legends so successfully, it is a runaway hit. I’m not sure exactly why it is that publishers and developers feel they can ignore history and empirical evidence when designing a persistent world; it is just so.

It’s like every physicist in the world ignoring the work of Planck and Einstein every time they start a new research project that involves cosmology. They may get to the same place eventually, it just takes long and costs more.

And isn’t that kind of stupid?

Recent Discussions on Biting the Hand: