What should a good GM do – what should his or her practices be – in three steps?
1. It is laudable to over-prepare; it is lamentable to over-plan.
This could be an entire argument in and of itself, so please allow me to limit this to the detail I will provide in the following paragraphs. One of the traits a good – meaning functional – GM is to be prepared to run the game, by which I mean the system. This does not mandate that a potential GM needs to lock him- or herself away and study the mechanics as though a proper understanding would enable him or her to pass the Bar Exam. It does, however, mean that the GM should have a mastery of the basic rules and if not a knowledge of the more complicated concepts and mechanics, than any easy reference to them (say, like a GM Screen with various rules printed on the inside). Nothing does more to draw everyone out of the experience than constant rules referencing. Some rules referencing is unavoidable, if only to settle questions between players about how some situation is resolved, but knowing the rules helps the GM attain and maintain the position of authority at the table. After all, he or she is the ultimate arbiter of what happens in the game.
That is not the extent of preparation, but the beginning. It better serves a GM to prepare as much of the game world/universe as is practical before the actual playing begins. I note "as practical" because fully fleshing out the world/universe can lead to a restrictive outlook from the GM and a constricting environment for the players. Fully developing the game environment before the players have any level of input falls squarely in the camp of over-planning, and that will be addressed later. The prepared GM has more than an idea of the setting, but rather has a concept that can be well described and explained to the players in a consistent manner.
The game world/universe is not just an empty box into which concepts can be plugged, but a vibrant entity that has a life given to it by game play. As such, the game world/universe needs to be able to meet not just the vision of the GM, but also fit the expectations of the players, and accommodate both the backgrounds of the players characters and the consequences of the actions taken by said characters. Preparation allows the GM to exercise a more consistent mien and the freedom to be creative within the established framework, thus not bringing about unintended moments of parody or absurdity.
Over-planning, on the other hand, prefigures the actions the player characters will take. It established a plot-line which the players must navigate in order for the evolving story to make sense. In actuality, over-planning diminishes the evolution of the story; the only changes that end up happening are the ones that either allow for lost player characters to return to the regularly scheduled plot or – in extreme cases – to abandoning the ruined story in favor of a simpler one the PCs can't help but follow. Over-planning is not only restrictive – for all parties involved – it can also lead to resentment because the value of choice is trivialized when the GM adheres to the planned scenarios.
Being prepared means being able to handle the rules, the world, and the PCs' choices affecting what is happening. Over-planning means anticipating the actions of the PCs and plotting out plot lines and story arcs for the players to follow. One allows for interactive creativity, the other for creativity (maybe) forced upon the rest of the group.
By means of example, I would point to a brief Vampire: The Masquerade (3rd Ed.) chronicle I ran titled Dallas: By the Light of the Silvery Moon; I believe it lasted somewhere in the neighborhood of twenty sessions. I had a spiral notebook with 300 pages of notes, maps, and stats for the anticipated NPCs (and that was just for Dallas proper). I would reacquaint myself with the notes between sessions, updating them with what the PCs did, who they met or influenced, and locations visited. I would then review where the last session left off – and what may have been left unresolved from sessions prior to the last – and have a loose plan for the introduction of the night's session. The players would take the game in whatever direction they pleased, oftentimes scary and wholly unanticipated, but I could react to them in a responsible manner because I had prepared for as much of the game world as possible and knew the rules backwards and forwards.
The opposite of this would be an AD&D 2nd Ed. campaign in which I played where, regardless of anything the players wanted to do, the PCs were moved from location to location towards some eventual goal that – lo and behold – was instead accomplished by an NPC (so that the players couldn't screw it up). Yes, that is an extreme example of planning (and plotting) out a story and sticking to it, but many GMs still indulge in this to a larger degree than necessary. It ruins spontaneity (or at least reduces the impact of it to nothing) and kills the cooperative aspect of evolving the story the players and GM are experiencing.
2. Know your players; know their characters.
I am sure this is a subject often brought up and with many different points of view, but there is such a person as the problem player. The first step a GM should take, when taking inventory of his or her players, is identifying whether any of them are of this breed. The problem player is oftentimes, but not solely, one whose primary goal at the table is to foster some sort of discord among the characters. This is a problem that cannot be easily solved once it has reared its ugly head, in part because the problem player has exposed him- or herself as a person who has little regard for the other players' enjoyment of their characters.
My recommendation – and one that it took me a good long while to adopt – is to view each potential player as an applicant. Are they well suited for play with others? Some players come right out and let the GM know they enjoy playing characters who are disruptive. This has happened to me four times in the last three months (they were looking to get into any type of home campaign), and in each instance I let them know the following: I would not ask them to join an established group I had (they seem to want to disrupt a functioning dynamic), I would not want to start a new group for them (as they have expressed that they aren't interested in building a good group dynamic), and that I find that approach to playing very immature. This is not a process that is simply limited to when a player enters the group, but one that should be revisited whenever a conflict seems to arise without legitimate cause. Players should likewise be consistently evaluating the GM to see if he or she is delivering the kind of experience the group – and individual players – want and deserve.
Once the group is established, it may seem more than a little difficult to consider removing a player from the group. It should be, but it should also not be the first resort to conflict. If the GM has taken the time to become acquainted with the players, it should be easy to have discussions with the players about the problems that arise during play (it is my position that the GM should not take on the responsibility of managing real-world relationships of the players; such a proposition would require the role of GM to pay extremely well). Still, problem players are like a cancer for a game group, and it they cannot be treated with a discussion of what led to some early conflicts, they should be excised from the group. Likewise, players who are unhappy with the role their GM has taken on – oftentimes in regards to the level of responsiveness to the players – should first have a discussion about their disappointment with the GM, and, if the problems continue, leave the GM (which may also mean leaving or ending the group).
Knowing the players has more benefits than being able to identify problem players and taking early action to resolve conflicts at the table. It allows the GM to have a better grasp on what these players are looking for in the gaming experience. This is not an absolute – I played in a Dungeons & Dragons 3E campaign at NIU where the GM and three of the players all favored combat over roleplaying sequences, but only one player was disappointed when we did back-to-back sessions without a fight (and the GM was the happiest person of all!) – but it is a great place to start to get an idea of what kind of experience the players expect.
Knowing the player characters is a little easier, but in order to do it correctly one must have some kind of understanding of the player. I think many GMs are against the notion of the broken character. I am against the disproportionately powerful character, but that can often be regulated by some simple instructions on character creation at the beginning of the campaign; for example, not allowing mega-heroes in Heroes Unlimited. On the other hand, I find it a little high minded for a GM – especially when it's me – to tell a player he can't make the kind of character he or she wants to play. The only question is how to accommodate the type of character into the world – and this is where preparation again comes into play – which is easier if the GM does not have an overly restrictive view of what belongs in the game world/universe.
Knowing the characters means being aware of their motivations and their capabilities. This enables the GM to interject (personally or emotionally) meaningful moments for the player characters into the game. It enables the grander story to blend in various elements that would otherwise be disparate appeals to individual character desires that become known only in retrospect. The more obvious aspect of knowing the player characters is that challenges and encounters can be properly proportioned to be – or appear to be – risky without being over- or underwhelming. Knowing the characters makes preparation much easier, and can reduce the desire to over-plan.
3. Remember that it is your game, too.
Being a GM is not just about being the world and its reactions to what the player characters do. Ultimately, the GM is another player at the table, and one who deserves as much of a chance to enjoy the game as everyone else. Just as the GM shouldn't reduce the amount of influence the other players have in the story, he or she shouldn't feel that they can't add their voice and contributions to the story and setting. Table top roleplaying is a cooperative experience, and while the GM is weighted with the authority of being the final arbiter, he or she has to find a way to be an equal in terms of having a good time. Luckily, many GMs find pleasure in how a good story unfolds, and how players take unexpected routes and find novel solutions to the resolution of the challenges in the game.
As a GM, one of the things I like to do is create tiny rewards for those moments where the players surprise and impress me. These oftentimes have little more in-game (mechanical) effect than what the player characters may have acquired on their own, but I make every attempt to tie them to the moment, the character, and the overall story. I have had a fair amount of success doing this, but I also know that there are players for whom it does not do much.
I apologize for not being able to fully articulate how to exercise a level of participation in the game that does not rise to the level of ownership nor sink to a level of being subservient to the players. I do know that it is very important to find this level of balance. This may seem to go against the notion of the GM being the one who gets to choose who is fit to sit at the table – that is an inconsistency I cannot easily resolve, either. Maybe the easiest solution to this is to make sure that enjoyment can be found in various ways, and that there is more of a need to allow to exercise caution when exerting the powers available to the position of GM.
I hope that meets some of the criteria of the challenge.