A game for loosers
Making a multiplayer modification is more of an art then anything else. While it can be visually stunning, it's that which is unseen to the player that decides the true value of the mod. This invisible factor of a mod, the game mechanics, has yet to be turned into a true science. Some games seem to be lucky and have what it takes, causing the mods popularity to soar. Others mods can have some unseen flaw that somehow causes the mod to fade into obscurity despite the best efforts of the developer. Much like not seeing the trees through the forest, a game developer can concentrate on modifying so many different aspects of a mod in order to fix it that the most important factor can be overlooked.
Making a game where the winning players have fun is an easy task. The great difficulty lies in making a game where loosing does not make you turn off the game in disgust or frustration. For singleplayer or coop games this is much less of a problem, but for multiplayer games this is a great dilemma. One side always has to loose, meaning that at any one time half of the players playing the mod are on the loosing side. It is these loosing players that will decide the popularity of your mod. The winning side has nothing to complain about, they're winning and having a blast. The loosing side on the other hand can be miserable, and once they think they have almost no hope in winning they may switch to another server, or even worst switch to a different mod.
The most popular modification of all time is almost BUILT for the looser. Counter-strike has many different mechanics that make playing the loosing side much more bearable. Every several minutes everyone respawns and the round begins again. Character development exists in the game, but only allows limited advancement. Death is never a total loss, while ones equipment may be lost they still retain their in-game cash. Also even if your team looses, if you killed several of the opposition you gain more cash. These four factors are extremely important to the success of counter-strike, all of which try to prevent one team from perpetually steamrolling the other side.
So what can we learn from mods like counter-strike? Making sure the loosing side has hope. Making sure the loosing side still believes they can win the next round. The loosing side does not even have to be loosing, they must simply believe that they will loose. When a player leaves the loosing side because they have lost hope, whatever situation the loosing side was in just got worse because now their a player short.
Having multiple rounds or frequent restarting of character positions can continually give the loosing side a chance at the bat. While multiple rounds per map are by no means necessary, without the multiple rounds the developers will have the problem of one side camping the high ground of a map. Players that are killed re-spawn and then rarely wait for teammates before proceeding back into the firefight, providing a trickle of players which are much easier to kill then fighting the whole team at once. One method to help avoid needing short rounds is radar to allow teammates to help find each other and provide an actual challenge against the winning side.
The worst blows to moral are from long term character development. If the other side starts to get better and better equipment and it looks like the trend will continue to the end of the round, then what hope does the loosing side have? As fun as getting new weapons and abilities are, the makers of a mod must realize that if they reward the winning side too often then the loosing side may become completely overwhelmed. Thus itís always important to have limited character development, to make sure that a starting character has at least a chance of killing a fully developed character.
One of the four things counter-strike did right was rewarding cash to those who killed opponents. Even if the player is on the loosing side, this can make them feel like they are actually winners. As long as the game rewards aren't entirely linked with the success of the team as whole, players can have a personal victory in spite of the team loosing overall. The worst part of being on the loosing side is dying. A lot. Games where a killed player looses all of their characters development are games where the death penalty may be too strong, especially if it took the player a lot of work to develop their character.
The most important thing the developer of a multiplayer mod can do is acknowledge that their mods success rests almost entirely in the hands of the loosing side. Keeping this in mind while creating the game mechanics can decide if the mod is a true masterpiece, or yet another piece of work to fall into obscurity.