HW also has the other oddity that all skills are equally useful unless the GM deems otherwise. So if someone comes at you with a sword, you can pit your orate skill against their fighting skill which is cinematically satisfying but rather undermines the use of the fighting skill and its physical impact. This puts a big burden on the GM in justifying the value of skills to each situation and can be quite awkward sometimes, and probably not what you want in a narrative game. But that's by the by.
So, I'm not entirely sure where these ramblings are going. Time to take a step back.
It's conflict that makes a game interesting (cf Games don't kill people do they - Greg Costikyan). It's emotional investment in that conflict that makes the game personal and memorable, in other words worthwhile or fun. There are various kinds of emotional investment in games. The theory (cf GNS, Robin's Laws of Gaming etc) generally seeks to identify to which part of the game the investment is attached, be it the winning, the narrative, the shared imaginary space, character development, emotional bonding with the character, escapism, friendship with the other players, the mystery, moral choices etc.
I guess education can be added to the list above and whilst I have run a couple of educational games, that's not what most of my gaming is about and not what I want to consider here. No, I'm talking about gaming for fun, not profit or education.
Games need some method of conflict resolution that doesn't step on anyone's emotional investment. You'd hope it would actually increase or at least maintain it. So I suppose what I'm looking at is matching up the focus of emotional investment to a method of conflict resolution.
I need to cover something about the structure of conflict resolution.
In old school gaming, the structure is a series of encounters which have a well defined purpose. The purpose is set by the GM and the series of encounters forms the plot. Typical conflicts to be resolved are "Let us pass", "Give us what we need", "Find the clue", "Kill the bad guy".
In new school games it's sometimes about what would be good for the narrative (either GM or player decision), sometimes about pushing the characters buttons, i.e aggresively framed scenes that put the character in a dilemma where they have to make some kind of moral choice that has real meaning to the PC. Not just about whether they should defeat the evil one, but what they are prepared to lose in order to defeat the evil one, and the consequences of this choice.