Most of you who know me know my approach to RPGs is a bit like tissue paper. I do not run the same game system twice unless pressured specifically to do so. My philosophy is that a game should complete in three to six sessions and that each game should use its own system, engineered to support whatever it is you're trying to accomplish with your game.
Many people are very satisfied running their favorite system. "What I'm trying to accomplish with this game is to give my friends a good time! So I use a system I know we all enjoy!"
I say: you're really missing out! Oddball systems offer very different dynamics, and the fact that these are short games (even six sessions isn't really very long) means that even if they're terribly unbalanced, it's not going to be un-fun. Of course, not every oddball system is worth playing, and what you want to do with your game will never precisely fit any given system.
That's why I always roll my own.
So, for those of you who might want to try your hand at something deeper than picking a monster out of the manual, I'm planning on writing a set of articles on how to make your own systems. To find all of them, do a keyword search for "make your own" (link at bottom of essay).
Turn order is a huge part of how your game feels. Here are some basic systems:
Free-for-all. Games default to free-for-all, and most games will be free-for-all even if they have some specific method, because most systems are sloppy enough that players won't be really aggressive about wanting to go first. The less advantage there is to going first, the more likely free-for-all is.
Free-for-all does everything bidding does, except crappily. There's nothing really wrong with that: bidding is often overkill, and free-for-all is basically zero overhead. Regardless of how you design your game, there will be a lot of free-for-all chatting at basically every moment of the game.
However, I suggest that if free-for-all serves all your game's needs, you're probably using a really boring system. All your players should be panting to be the next player to go.
Round Robin. More common in card games than tabletop RPGs, round robin simply goes around the table in a predictable order over and over. This is terrible, because not everyone has equal opinions at all times. I find that round-robinning usually means that half the players say "I pass" and the other half say, "I can't do what I wanted to any more!"
Round robin is a basic choice for more competitive games, or recursive games where the action of the last player matters hugely. However, in these situations much of the game lies in who you sit next to, so there should always be a game mechanic for switching seats and controlling turn order!
It is also not uncommon for the GM (if there is one) to interleave himself instead of simply getting to go once. Usually, giving the players five turns in a row is just begging for unbalance.
Staggered Round. This is a kind of round robin progression which has a kind of built in "fairness". Usually a staggered round means each time you go around the table, you start such that the person who went first last time goes last this time. This isn't the only option: another fun option is the "ping-pong" method where you go around the table, "bounce off" the GM, and go around in the opposite direction. This creates a kind of tit for tat mechanic which keeps players sitting "downwind" safer. There are other kinds of mechanics for this sort of thing, too.
Again, however, where you sit is tremendously important, so changing seats should be part of the gameplay. However, controlling turn order is less important, so it'll generally be more transparent as to who will go in what order. While, you know, still being something vaguely resembling fair. So even though the rules seem more complex on the surface, staggered round is usually better for newbies than the round robin, because the round robin is either boring (if you can't change turn order) or surprising (if you can).
Initiative. Many RPGs force players to go in a specific order that has nothing to do with where they sit. This is usually an initiative system: the faster their character, the sooner in the turn order they get to go. It also allows the GMs to interject NPC turns in a meaningful way rather than an arbitrary one.
The downside is that it is not a very juicy mechanic. It is dusty dry, either largely luck or almost entirely predetermined by character creation. There is not a moment-to-moment feeling of staking your resources on turn order.
"What the hell is he talking about?"
Try a different turn order type, you'll see what I'm talking about. Initiative is simply the dryest form of controlling player action, short of round robin with no way to control turn order. Nobody really gets excited about initiative. At least, nobody after they've played any system using any other method.
Simul. Some games try for simultaneous play. This is good for games which are potentially very competitive, but for cooperative games it makes teamwork needlessly cloudy and difficult. Also, while some people take their turns absurdly slow, those same people take ages to try to figure out what to do simul. Simply put, players love to build on what other players have done, and this muddies that equation.
Plus, unless you have at least eight players, you won't be saving any time going simul. Trust me, I dedicated a good chunk of a summer to it. Using a time clock to force timely moves will alienate your players faster than you can say "".
Bidding. Bidding is when players spend resources to determine turn order. For example, if everyone has ten chits and gets two more chits a round. Everyone bids chits and play goes from highest bid to lowest bid. Bids are either simul with a lot of tabletalk or in some other turn order with minimal tabletalk. (Competitive games are, of course, simul with no tabletalk.)
This lets players decide how much they want to go first (or last, if that's better). And, of course, it doesn't have to be chits they can only use for initiative: you can have a kind of "action point" system. They spend action points, they get to go first but have less action points to spend on actually doing stuff...
If bidding is going to slow you down too much, you can "front load" it - bid once for an entire sequence instead of rebidding each round. Also, you may want to read about the controlled random method if you like bidding but it seems too slow.
In order for bidding to be a viable system, turn order has to really matter.
Wounded Puppies. Some systems determine who goes in what order based on how well they did last round. Some systems make the winners of the last round have a better place in the turn order, but I would suggest the opposite: negative feedback loops are almost always better than positive feedback loops.
How you measure "how well they did" will determine the entire flavor of your system. If people who get injured more get to go first, it will feel like a very adrenal, high-tension game. If people who are closer to the enemy get to go first, it will tilt the entire play style of ranged characters and melee characters. It doesn't even have to be last round: it can be determined by a day's activities or somesuch.
This system really says something. If you want to give your game a message, this is a really easy way to do it. Just don't get preachy.
Controlled-Random. This method gives the players some control over their initiative but introduces a heavy random element. Usually, it's a simplified bidding system where what you can bid isn't so easily divisible.
For example, you might have a hand of three cards. You can play one as your initiative. (Maybe the others are used for hit and damage, or maybe initiative in the following rounds.) While there is a big random element, you have a level of control. Especially if you can trade cards.
There are many methods to do controlled random. That's a post in and of itself. But the basic idea creates a very strong, tense situation and creates a very good atmosphere. I highly recommend it.
Reverse Bidding. Reverse bidding is another handy shortcut. Basically, you have a list of numbers from 1 - 20. The players all get to choose any number they please, with the better numbers going first. Then the GM gets to choose a number or numbers.
The catch here is that every turn, each gets a token put on it. What that token represents depends on what the game is - maybe it's an XP boost, maybe it's a story token, whatever. Whenever a player picks a number, they get the tokens. (The GM does not take the tokens unless he only gets one turn, for balance purposes).
This is a specific example of a general idea. The idea being that you let players choose to act inefficiently in turn order in exchange for an advantage in something else. This is a very strong general-purpose method that can be applied to almost any facet of the game, even though this example is specifically for turn order.
Complex Phases. Not really a "type", this is really a mixing of types. The idea here is that your turn order varies dramatically depending on the phase. Maybe you bid on the movement phase, go round robin in the action phase, and go wounded puppy in the recovery phase. Maybe you can be a huge bastard and actually make the types have turns that players somehow choose at the beginning of each round, so they have to decide whether they want to do round robin for the first phase or the fifth phase.
Complex phases create a very complex interconnection between the players and the way actions reverberate through the game. While unsuitable for beginners, it can provide endless fascination for more advanced players.
Because a turn will typically have at least two (usually three) complex phases in it, these are not suitable for "fast" turns like rounds of combat. Instead, these are more suitable for huge chunks of battle, or political maneuvering, etc.
GM Fiat. Don't you do it. Just don't you do it.
How many of these methods have you used in an RPG? Have you used any I didn't think of? Did you really read that whole freaking thing?
It's your turn now.