I've noticed a problem.
It's not so much a case of Story vs. Game, because neither story nor game really approach the dilemma I've recently been seeing with the rule mechanisms I've been developing for Walkabout.
The last combat example really highlighted it.
The example showed the first sequence of a combat (a sequence two combatants engaged in about ten combat actions each). It showed some of the good sides of the system and some of the bad sides.
Let's say roughly half of the actions were "attacks" and half were "other actions". In a traditional RPG, this exchange would take a minimum of about 30 die rolls. Each "attack" would consist of a die roll for each attempt to hit, followed by a die roll for damage...there might be a chance to dodge a blow, absorb the worst of the injury (armour or saving throws, or both). Each "other action" would consist of another die roll, probably against some static difficulty to see if it passed or failed, or maybe in a contested roll against the opponent.
This can be really slow and tedious in some games. Especially in those games where there is a high chance to miss your opponent in combat.
The Walkabout example shows that each sequence of combat takes a little bit more effort to set up; but once the tokens have been drawn and allocated, it allows for a quick and free flowing narrative as the effect of the traits are alternately described by the players involved. The core setup of the exchange might take three times longer, but each action in the exchange takes a third of the time ("attack roll + defense roll + damage roll + description" reduced to "look at token + description"). It only takes about four actions in the conflict (two per combatant) before it starts becoming a quicker system that is more carefully integrated into the characters abilities and the situation at hand.
I also like the interplay that develops as each player narrates back and forth throughout the conflict. One player might narrate taking the upper hand, while the other plays dirty and gets in a quick hit. Every action has a chance to be countered, but if you counter the last action then you miss out on the opportunity to take the offensive. Do you choose to play it safe and avoid injury, or do you choose to forego the consequences for a bit of extra damage on your opponent?
We also see in the example that a better prepared combatant is far more likely to end up on top after a combat sequence (I tried to show a fairly balanced drawing of tokens for each combatant). It's still possible for a experienced and prepare combatant to draw all white tokens and be beaten by a less worthy opponent who draws all black tokens; but this kind of scenario is unlikely.
As a system to replicate scenes of dramatic swordplay and derring do, this also works well...step left, step right, see an opening, thrust, parry, step left again...at the end of the exchange, the combatant with more tokens gets to narrate how their final action decimates their opponent with a flurry of uncountered blows, or perhaps the combatant combines their successes for a single vicious strikes that runs their opponent through.
Guns and dramatic mid-conflict attacks.
Guns don't play a big role in Walkabout, but the game is post apocalyptic. Guns appear frequently in this type of setting so they need to be modeled more effectively in the rules. We also need a system where a dramatic swing with a heavy sword can literally tear a villain in half.
This is taken into consideration with a few of the FUBAR rules that will be ported across to Walkabout. Lesser NPCs only require a single success to be removed from the story (this includes the average person on the street, the typical bodyguard...anyone who is added into the scene for colour rather than for dramatic story development).
It's in the big fights with named antagonists that something just doesn't seem right.
I think we need cascade effects.
By this, I mean that one trait is activated by another in a rapid succession before the opponent gets a chance to respond. This means that players get the chance to inflict massive damage when their firearms attack is followed up immediately by the bonus from their gun's "Lethal" trait, and then followed by their "Weapon" relationship. A single strike can inflict a triple degree of penalty, removing someone from the scene immediately (unless they can respond with a defensive or protective trait on their next action, which might trigger a cascade effect of it's own to eliminate more of the damage). With enough interlinked traits, a good cascade effect might remove a character from play permanently with a single strike.
Of course the problem with this is that named antagonists should also gain access to cascade effects. If the odds are in a protagonist's favour, and they are capable of demolishing an opponent with a dramatic cascade of devastation...then there needs to be the same risk in reverse when the odds are in an antagonist's favour.
One of the things I like about the notion of cascade effects is that it brings more prominent traits into the story more quickly. It doesn't have to be restricted to combat situations. One character might initiate a cascade effect which starts by uncovering a minor situational clue about the mystery at hand, and this leads through a cascade that unravels to a minor short term clue, a minor long term clue and then a major long term clue.
It speeds up the mechanisms at work, allowing them to drop out of the picture so players can focus back on the story.
I think cascade effects need to be reigned in at some level, and I haven't worked out the best way to do this. More thought required.