So it is nice if I can envision something outside that framework, giving me something new to consider.
Earlier today, between breakfast and lamenting the necessity of cover letters and their actual value in obtaining a meaningful position, I found myself thinking of random generators from a different perspective. I'd like to present that perspective now.
The intrinsic problem with any game mechanic based on a die roll is in their limitation. Typically, dice are used either to indicate pass/fail (such as the 'to hit' die) or to offer a range of possibilities (as in any random monster generator). Board games such as Monopoly or the game of Life use the second method by streaming the options in a linear loop or ribbon - the multiple possible results of rolling a '7' are different if you're on Park Place than they are if you're on Boardwalk. The existence of cards or other game mechanics will then result in certain parts of the board being more valuable than other parts. The table below, for instance, gives the average number of rolls needed in order to recoup the cost necessary to buy the property:
|Nothing beats New York|
It is in this manner that certain board games made the best possible use of dice before the existence of role-playing games.
The pass-fail result is a harder mechanic to build into a strategy. If I need a 17 to hit a given monster, then all I can really do is keep trying to hit every round and wait for the odds to pay off. In later games, we saw an attempt to change this by allowing players to increase their odds through gaining a variety of skills - but fundamentally the static result remains the same: there are only two options:
Suppose we want to decide something less concrete than whether or not a character can jump a ten foot gap without falling - say, in picking the commander of a group of soldiers under our command. Imagine that we assign all commanders two fundamental traits, loyalty and ability . . . and then when the battle occurs, we have that commander roll pass/fail on both those traits. This would give us a slightly more elaborate set of results:
That is a bit better. Offers some nuance. We can assign at least three different results for the battle. Where both rolls pass, terrific success. Where both rolls fail, a rout. And when one roll passes and the other fails, the battle (or whatever situation is to be decided) ends in a stalemate. Moreover, we can split the stalemate result into two forms - one due to the commander not being loyal enough and the other due to the commander not being able enough. If a plan is in involved, there are meaningful differences between being loyal and being able.
More importantly, note how we've reduced the chance of a total failure from 50% to 25%. That is significant - and will become more significant as we move forward.
Suppose we introduce a third trait in our commanders: personality. Loyalty only describes the commander's attitude towards us - personality describes the soldiers attitude towards the leader we've selected.
Now we have three separate rolls to make - which requires that I produce something like a 3-D chart to keep track:
Best I can do, I'm afraid. This sort of thing can quickly get confusing, so I will try to explain. The upper and lower four square arrangements are reproductions of the loyalty/ability table above. In the upper square, the third roll, the one for personality, passes universally. In the lower square, the third roll fails universally.
Now we have nuance. There is only a 1 in 8 chance of all three rolls pass; but also only a 1 in 8 chance of all rolls failing. There is a 3 in 8 chance that two rolls will pass and a 3 in 8 chance that two rolls will fail. Even without distinguishing which roll passes and which roll fails, we have four levels of result: a total success, a moderate success, a moderate failure and a total failure. We have 8 different categories that we can use to designate specific circumstances surrounding what abilities pass and what abilities fail - which in turn can never be entirely equal in their importance.
For example, suppose a commander fails at loyalty and personality, but succeeds in ability. This could suggest that while the commander does not care for the plan or the soldiers care for the commander, the commander's sheer capacity to manage the situation stopped the battle from becoming a complete rout. A commander like this should perhaps not be in charge of men but of the whole war, where he or she can take part in making the plan.
Suppose, instead, the commander fails at personality and ability but nevertheless remained loyal - the commander lost the battle and sacrificed the men, but yet managed to save his or her self. Is loyalty in this case really a useful trait? You'll have to decide.
Unfortunately, with three rolls we have no potential stalemate. Thus I am pressed into producing a 4-D diagram. I apologize if this gets confusing:
I think I will disdain from this point from going through the specific possibilities. By now the reader should have the idea and be able to count. What's important about the above is that none of the rolls need to be a 50/50 chance of pass failure. Obviously, you would want the best possible set of traits imaginable, basing them on intelligence (ability), wisdom (loyalty), charisma (personality) and perhaps a completely outside the framework determination for sanity (which would not be wholly evident, would it?). Balancing these different attributes is what makes for a similar nuance that a board game offers to dice-as-selection-tools.
Since a character's ability stats are themselves based on a selection die, and since we can enhance any characteristic by selecting aspects of it still further, there's no end to the nuance that can be achieved. We can get further interesting patterns by changing for what length of time the above rolling applies: A single round? An hour? A battle? The 'to hit' die system works because we expect the players to be able to roll several times to get an overall average of success. By increasing the number of times we need to roll to determine the result, we produce a more reliable mean from which to judge success (consider the table above regarding Monopoly properties).
This is all old hat. I meant to expand the reader's perception of the decisions being made by making them more visible. I hope that it helped.