Understood, and taken in good faith.
I agree that it is a trivial rule to drop, but it perplexes me that it is still there. If you want to approximate steady state, it is easier to just not add noise by default, no?
There are nine distinct morale values, all of which modify slightly different facets of the domain, and then there’s a table of 12 modifiers to the roll. Me, I’d be pretty happy with domains having two morale states (“tolerates ruler” and “open revolt”), with a d20 roll triggering a revolt on a 1+ (or 5+ or 9+ if the ruler has done something egregious lately) and revolts persisting until either egregious ruler behavior has been addressed / peasant demands are met, or they are put down by force. That’s about the level of complexity I’m in the market for.
At the end of the day, my players only want to deal with one layer of henchmen. As a consequence, PC domains are practically limited to one layer deep (if they weren’t already by other factors). Likewise, I have a limited amount of prep-time and interest for NPC realms, which is best served by paying attention to the count/duke layers of the chain (who make reasonable patrons or villains; not too high that the players are irrelevant, not so low that the players can kill them trivially). I don’t care if there are marquis or whatever below them or not, and my players sure aren’t willing to manage a multi-layered domain structure, so any actual rules for low-tier (or very-high tier) vassal rulers are wasted space as far as I’m concernd. It’s not that I can’t ignore them - it’s just one more thing I have to houserule around, particularly given shrinking maximum personal domain size. Houserules are expensive; in a complex system, the number of unexpected possible interactions between parts grows superlinearly with the number of parts.
Lacking a printer, our options are pretty much spreadsheets, wiki pages, or text files. Or proper databases, I guess. Updating hex populations over time, and then generating income for each month given varying population and morale, is a very natural spreadsheet operation, and doing it manually per-hex (possibly 1.5mi hex) sounds like an awful lot of unnecessary work.
There’s a bit of a difference between the expense of designing a class, which is undertaken but once, and the recurring expense of a complex domain system, which requires work every month of game time (which might be twice a session or more). If anything, that post’s conclusion on fighting value 1 is a simplification and generalization of the current rules. I have very little trouble with math in principle; it is repetitive math, compulsory math, and math that comes up during play, which are best eliminated. Automation is a band-aid, a crutch. If my players were hot-to-trot and wanted to design Traveller starships with Fire, Fusion, and Steel and run domains with population detailed down to 1.5-mile hexes, I’d be grateful for having complex options. But they don’t, which is why I’m doing all the math once and building standardized domains, so just like with standardized Traveller deckplans, there’s an option on a table that works closely enough and has all the actionable stats in one place. So what if the population is off by 5%? The players don’t have an accurate census in-world, and any discrepancy in income can be handwaved as either particularly aggressive or lax tax collection. I’m not willing to worry about it. As far as my players are concerned, a domain is a thing that gives you gold and XP every month, and helps offset the cost of the mercenary army you wanted. Broadly, the point of the game for us is killing things and taking their stuff (because these give you XP). Everything of interest is one of: threat, weapon, loot, simultaneously weapon and loot, or Not Sure Yet. The value in a domain is measured in how much better it makes you at killing things, and how much stuff it gives you, and we measure its cost in paperwork against its value in those terms.
Simulation is good in that a good simulation avoids breaking suspension of disbelief, but if simulation were my end, I would be programming, not sitting around a table with dice and friends. Having a properly-balanced game is good in that overcoming true challenges is deeply rewarding, but if that sort of satisfaction were my end, I would be playing RTSs. Narrativism is good, in that a well-executed story can tap symbols and trigger emotional response, strengthening the game-ritual, but if telling stories were my aim, I’d be writing novels rather than playing. Simulation is the groundwork on which the other two can build towards emotionally satisfying conclusions without being interrupted by confusion and WTF. But, like most infrastructure, it seems best to me if it is simple, robust, and hidden.