While playtesting Spirit of the Far Future we encounter many cases where a GM is not actually needed for the fraction of the game we want to test — say when testing the combat system. In these cases we find that there are gaps in the FATE system that need to be filled. Following are ideas for filling those gaps that will certainly work when playing FATE as a framework for tactical gaming and may work for GMless role-playing.

Obstacles

There are three primary obstacles to getting rid of the GM in FATE:

• The GM is an infinite supply of arbitrated fate points. If the GM is not present then fate points dry up pretty fast through lack of compels.
• The GM supplies and maintains the context for the game by designing the initial scenario and propelling play towards the story points he wants to hit. He also draws teh maps.
• The GM paces the game, pushing players past boring bits and focusing them on interesting bits and generally trying to get the session to a satisfying place by the end of the night.

Solutions

Fate points

The solution here is to use a zero-sum fate point system. This means that whenever a fate point is spent by someone, someone else gets it, so a fixed number of fate points are in play and that number never changes. Specifically:

• When a player invokes one of his own aspects, the fate point is paid to anyone (at the invoker's discretion) to whom the invoke is detrimental. The underlying assumption here is that if no one cares whether or not the roll succeeds (that is, it's not obvious that anyone should get the point) then the roll is by definition unnecessary and the action should automatically succeed (say yes or roll rule).

• Follow the same rule as above when paying a fate point to invoke a stunt.
• Anyone can compel anyone elses aspects at any time and exchange points as though they were the GM. The effects of a compel must be normalised, however, as they are normally a highly discretionary effect, so see below.
• Any player can compel his own aspects for a normalised effect (see below), again paying or receiving from the player who is affected, though in this case advantageously.
• The effect of a compel is either loss of a single turn (if compelled before any dice hit the table) or -2 on a roll (if compelled after the dice hit the table). There must be a narrative associated with this effect if we're role-playing but that's less necessary if we're just tactically gaming. There should be a genuine story though even if it's not narrated.
• Players should be encouraged to haggle compels. Offering two fate if the opponent is not going for one is certainly reasonable. So is turning down two but offering to accept one. In a zero sum economy the players should be encouraged to treat fate points as their own resource with few rules for paying off other players.

Free tags and compels

If an aspect is subject to a free tag and has not yet been tagged, anyone can choose to compel it instead for free. Mechanically this means that the person compelling need not supply a fate point if the compel is accepted but must be paid a point if it's turned down. This generates more realistic results to bad consequences like "shattered kneecap" as the opponent can at least force a missed turn.

Using a compel in this way removes the aspect's "free tag" status — it is now just another aspect.

Context (from Universalis largely)

Before the game proper, start each player with a pool of fate points well beyond what yuo need for the rest of the game. 20 sounds good to me. Going around the table, each player can state tenets of the game at a cost of 1 fate point. Any player can offer a fate point to deny a proposed tenet. In fact as long as the proposal has more fate points on the table for it than against it, it becomes a tenet.

Set aside the fate points spent to establish context (I'm thinking they may somehow be useful later). What you have left is what you have going in to the tactical game, so players are forced to trade off contextual control for individual capability within the context,

My instinct is that players should be buying something fairly concrete, so keep track of who bought each tenet — stealing from Joshua Newman's Shock:, this person owns the tenet. If there is any question about how the tenet affects play during the game, the owner has final say on its effects. This reinforces the trade off between narrative authority and character autonomy.

Tenet ownership is used to:

• set scenes
• establish environmental aspects before playing a scene
• draw any map on which a conflict relating to the tenet takes place
• mediate any conflicts relating to the interpretation of the tenet

Disputes between tenet owners (arising from interpretations of a tenet that affect another tenet) should be resolved with a fate fight — each tenet owner in the conflict may draw a fate point from the context pool (see I told you I might use these later) and may bid their interpretation with their fate points. Anyone may bid on any side. All fate points in the fate fight go into the context pool.

Pacing

Play should go around the table in some established order (let's say clockwise). Each player should narrate as they see fit and if a conflict arises it is set up and played out as normal. That regulates pacing, but doesn't get to the numb of the GM input on pacing, which is to speed it up when it's boring and slow it down when it's fun.

So, if an extended conflict (anything that's going to take a series of opposed rolls to conduct — usually combat — rather than just a single roll) starts in your narration (whether you announce a conflict or someone else objects and requires a conflict to resolve it), you get a fate point from the context pool.