Uncertainty is not randomness
For more than a month I have been inactive because I have been busy with too many activities and I had not the opportunity to stop, think and spend time on things that are useful for the mind... Then I realized that my absence was undoubtedly too long, and fortunately I came across two very interesting posts that gave me a lot to think about and allowed me to come to conclusions that, as is almost always the case, are taken for granted by those who think them (and here lies the difference between a good communicator and a clumsy person like me in this respect...).
I start with the first post: very simple, very clear and without any doubt logically perfect! I refer to 'A True Test of Skill' by Dwiz.
I will take this as a starting point to talk about the logical process I applied to the mechanics of the VI·VIII·X KUP RPG. You can read all over the net about the mathematical logic of RPG mechanics, dealing with issues of variability, volatility and so on. As far as I am concerned, I have defined my game mechanics according to a logic that the author of the above post expresses clearly and perfectly: an action must have a random component (resulting from the roll of a die), but it must not be preponderant!
I will make the comparison with D&D, as it is from that game that I started to try to make 'something better' with mine. In D&D, the result of a check is 1d20 with a bonus, which is (approximately) the PC's level and stat modifier. In order to have less weight for the random part of the die than the defined part, you need either several levels of experience or superhuman stats. A third level character with an excellent stat could have a +5 on the die roll... it doesn't take a mathematical genius to realise that the weight of the fixed part is less than half of the average die roll.
In order to reward more the part related to the player's choices (and give less weight to causality), the basic mechanic I set up in VI·VIII·X is diametrically opposed: the 'fixed' part is the sum of two scores (from 1 to 10) of the PC's stats, the known and the unknown, while the part determined by chance represents a third score (from -2 to +10) based on the die roll. The GM may add other modifiers to this portion to compensate for the negative term and to give more weight to the PC's strategy. Ultimately, it is very difficult to have cases where the random part has as much weight as that defined by the PC's characteristics. The rule of thumb in my approach is 2/3 of fixed part against 1/3 of random part: the strategy in this case wins against luck! …this, needless to say, should not be meant as less uncertainty compared to D&D! Uncertainty is a different concept than absence of randomness…
One criticism of such a mathematical approach is that there is not enough room for a low-level PC to emerge victorious from a confrontation with, say, a dragon. This is absolutely true! And it is true precisely because a low-level PC has no chance of making it... this does not mean that the game cannot be fun or challenging for the players: in the rules, I have deliberately included in the reward system the fact of surrendering or fleeing in the face of an 'impossible' confrontation! I'm more than convinced that many people don't like this mode because it takes away the epic aspect of the game... in VI·VIII·X, a PC like Frodo would never have succeeded in destroying the One Ring... in fact, that's not exactly true: Frodo often avoided forces greater than himself because he knew he wouldn't be able to defeat them!
[I added some further thoughts in this add-on]