So, basically frame tests of character as true judgment calls, exactly like what I said. A choice has to be made two or more options of moral weight.
I'm talking about tests of temptation or will, things like "Can my character be goaded into a fight?" or "Will I resist a bribe?" or "Can I be charmed by a skilled manipulator"? Places where the right choice as a player are obvious, but a character in a story might make substandard decisions.
I want players to have agency over judgement calls, where there are two or more options uncertain options that need to have a choice be made between them. But fighting temptation, or exerting willpower, isn't a choice between two uncertain options (generally, I'm sure there are use cases where it can be framed that way), it's a test of character makeup and concept. It's a challenge with stakes, and those should generally be resolved via mechanics.
These posts prompted some thoughts.
A couple of them also relate to this post:
All that we have heard of this situation is that the character has a conflict --- get information vital to his quest vs. avoid triggering his lust, and that he decided (however he decided) to go into the brothel. That sounds like "a choice, a hard one" to me.
How is it decided (i) what the quest is, and (ii) where the information is? More generally, how does the game generate this choice, about whether or not to expose one's PC to the risk of temptation?
Different ways of doing that are more or less railroad-y.
And following on, it might be interesting to drill down a bit more into "the right choice
as a player is obvious". Does that mean
the most expedient choice?
The choice that is most likely to progress the "quest"? (Which also brings us back to the first question, of where the quest came from.)
Thinking back to the example of my Steel test when Aedhros tried to stab the innkeeper: the "quest" - that is, the goal of killing the innkeeper - was decided by me, playing Aedhros. Given that goal, stabbing is the right choice. But It's not as if Aedhros hesitating
sets me back as a player - it's not comparable to, say, losing all the treasure down a chasm in a classic D&D dungeon. So the function of the Steel test, at least in that context, is not to put to the test my ability to complete my goal.
It does test "character makeup and concept" - how ruthless is Aedhros?
A further thought, a bit different from those above:
what counts as a "judgement call"? This is heavily shaped by the system. In my Prince Valiant play, for example, it was a judgement call (by Sir Morgath's player) to have Sir Morgath climb down the castle wall on a rope and rescue Lady Lorette. The player realised there was a risk that Sir Morgath might be smitten by the lady, but he made the decision anyway. (Why? In part because daring rescues are the sort of thing you do when playing Prince Valiant; in part because it was what he though Sir Morgath would do; in part because he knew that it would earn Sir Morgath Fame, which is the "XP" of Prince Valiant, similar to Glory in Pendragon.")
In Classic Traveller, by way of contrast, there is no risk that a daring rescue might result in this sort of consequence. That already tells us something about the different experiences the games aspire to: Prince Valiant is romantic and a bit pulp-y; Classic Traveller takes itself fairly seriously and is more focused on logistical and administrative operations than on Star Wars-esque daring rescues.
But in Classic Traveller, if you choose for your PC to enter a fight then you are choosing to risk Morale checks. And those moral checks are influenced by a commander's Leader and Tactics skills. So Classic Traveller, as part of its seriousness, foregrounds small unit cohesion and resolve in a way that Prince Valiant doesn't. (In Prince Valiant's mass combat rules, an individual can be defeated, or can triumph, independently of what happens to the army as a whole.)
These differences are important (from the point of view of RPGing): they are part of what means that different RPGs give rise to different play experiences.