What are your thoughts and feelings when it comes to simulated results?
Example: You have an army and a castle. An event happens where enemies attack. Your army's power is then calculated against the enemy's power and the result ends in either your win or loss.
Interestingly enough... I've played games that have this. As a general rule... I tend to abhor it unless the results are controllable to some degree in the "short term". If the only way to influence these "simulations" is with a lot of time and slow build up... then I'm just not interested all that much. I like a little Agency.
The example I like to cite for this is Samurai Warriors 2 Empires. While I love the game (it's pretty fun and on higher difficulties, control of the battlefield is far more important than any individual battle on it you win... which for me, makes it feel more like war and less like a series of excessive beat-downs and chunky salsa). If you aren't sure what the game is... if you've ever seen "Dynasty Warriors"... it's that... except it's in Japan and takes place around Nobunaga Oda and that period of the unification of Japan. Empires has "TBS" strategy in it as well, where you must take the country you start in and conquer the rest of Japan. You spend turns taking actions, you get so many actions per month/year, and the computer players do the same.
The game itself has two ways to "simulate" battles. Sometimes, from the "world map" you can simply simulate a battle. It takes army size and the strength of your Leaders into consideration and weighs it against the enemy and what they have, then determines a winner. You'll never use this, because it seems largely random.
The second way it "simulates" battles is by any area of the map that is not "loaded in" (basically, anything not around the player and everything you can't see). It will simulate what your armies do, what your generals do, and how well they perform based on "tactic" cards (which are rock, paper scissors, in 3 tiers... a tier two "rock" card beats a tier 1 paper card, but if they're both tier one, the paper card wins... The tactics are speed, defense, and offense.), based on Generals you have with you, based on the "morale" of your army (morale seems to affect how quickly enemy armies and units are dispatched by yours, if their meter is larger by enough, they'll begin steamrolling you), and based on any stats/weapons your units and generals are using. Likewise, some maps have "default" ways they play out with guaranteed wins for specific sides of the conflict, unless you intervene in significant ways (so historical battles are usually guaranteed to end in specific ways, unless you figure out the triggers for ending those fights or for steamrolling the enemy despite its advantages).
Because of that second way of simulating, you could sometimes count on your armies to do their job and the most you'd have to manage is swapping out your "tactics" card (you had to earn the cards on the world map, and then you could only bring so many with you each battle) to gain victory. However, I discovered early on that if you knew what you were doing on the battlefield... relying on this simulation was worthless. I found ways to routinely win battles where I'd have little more than 1000 soldiers against armies in a fortified location with over 50,000 soldiers. The key to doing this was simply to abuse the CRAP out of the Simulation. Namely, balance between being where you need to be... and being places where the simulation can take over. Generals that could curb stomp me were things to avoid... I'd pop the correct tactics and let the simulation handle them (even with inferior troops, if you were cutting off supply lines or kiting leaders away from valuable locations, your soldiers would use the Simulation to defeat generals pretty regularly or take locations on the map without your help).
I got in the habit of fighting every single important battle and abusing how the battles turned out just to make the "Simulation" work in my favor, even when there's no way I should've been routinely winning battles as lopsided as 1000 against 50,000. Granted, some of these battles lasted close to the Time Limits of 30 or 45 minutes... But, it didn't matter to me, so long as I won. The most troops I could lose was 1000, and the most they could lose was 50,000.
Sheer math. The simulation often hurt the chances of the enemy once I figured out how it worked and how it could be easily exploited.
In fact, you'll see these same exploits in almost any kind of "simulation" type game there is. Once a player understands the "rules" of how something is calculated and what they can get away with... it's a pretty easy exploit.
For example, in Cities Skylines... the game itself puts a massive emphasis on having good roads. This is never truly necessary. Once I understood how the game was calculating travel times and distances my population were willing to travel for any one particular thing... It was easy enough to build "blocks" that catered to every need and required very little "road" travel time. I put a huge multi-lane highway straight up through the middle of my city, put a bunch of easily accessible "side streets" that went into fairly perfectly formed "blocks" of zoning... and despite the heavy traffic all the time, and all the complaints... I made money hand over fist. The traffic issues were simply getting into and out of the city. The regular "blocks" of Residential, Commercial, Industrial, and Entertainment venues had very little traffic due to the populace not really needing to go very far for anything. I split up my "services" on either side of the highway to prevent them from trying to cross the busy highway and left the game to its own devices.
That's the problem with a "simulation" of that nature. They almost always have an exploit. You either need to count on players finding it and being okay with them using that exploit... Or, you need to spend a ton of time making sure it can't be exploited and works properly.
You attempt to learn magic from an ancient scroll. It then runs a calculation of your aptitude with magic vs the difficulty of the spell where you can fail to learn the spell (with added possibility of the scroll disintegrating and being lost forever since paper doesn't exactly keep its integrity over the millennia.)
Do you feel these types of things are unfair to the player or do you think it adds further complexity to the default systems in place I.E. Use a spell book, get a spell.
With this example, I don't mind it. A little RNG. I've played a lot of Rogue-likes that take this approach. Often if something I know is "difficult" to learn, I'll simply hold onto it until I'm confident I can learn it as a 100% chance. Yep, even if you don't display the "chance". One of my favorites was a game called "ADOM" (Ancient Domains of Mystery). It had tomes where you could read them and learn the spell "Wish". You could fail quite a lot at learning it. To the point that the book turns into ashes without you learning anything (every failed attempt makes a roll on what happens... sometimes the book is fine... sometimes it turns to dust... sometimes you get a random cast of the spell... and... other things). I simply went out of my way to maximize my chances of learning the spell. I held onto it until my Literacy was maxed out. I held onto it until I had enough stat points into Magic to comprehend it properly. I held onto it until I had enough food and privacy to sit down and read the tome for over 24 hours straight without interruption (if you were interrupted, you had to start all over... being hungry could interrupt you. Getting attacked could interrupt you. Your Deity talking to you could interrupt you.).
Though, you're going to have to expect a lot of savescumming with such a system. Something I do to games that aren't Rogue-likes. Oh, I have a 1% chance to pick this lock? Well, let me reload until I nail that 1% chance.
If the player really wants something, they'll reload to get it.
Now, what I did for the "simulated chance" is in my "Thievery" and "Pillaging" systems. Basically, your stats for particular characters determine "how good" of loot you get. With "Thievery", there are 5 loot items. Even if your stats are extremely low, you get something. The player never knows the threshold for what items appear with which stats. With "Pillage", there are 3 loot items. Again, the player never knows the threshold for these items. The player is given a small "buff" roll to their attempt in hopes that it "pushes them over to the next tier", but it's not guaranteed and is based on the difficulty of the chest. There are 5 tiers of difficulty for these chests. The "easiest" will make a random roll from 1-10 points. The "hardest" will make a random roll from 0-2. The player never knows what they rolled. They aren't even aware they're getting this slight bonus. There is no changed dialogue for an indicator of "how good" of items you got from the chest. As far as you know, as the player, this chest always contains exactly 5 Hyper Potions. You may never know that you just barely passed over rolling the Sword of Infinite Destruction.
Which brings me to the other thing I've done with this "simulation" system. Higher numbers aren't always best. Sometimes, the "median" number will hold the best loot. This means that an underpowered player may get a boon, while a super powerful player might get something more in line with keeping them at their current power level.
But, again, they'll never know. There is no "Failure" state. They might deduce that the chests have more than one reward based on the flavor text for its difficulty... They might deduce that there has to be some kind of determining factor for what they get when they 100% succeed on every Thievery or Pillaging check. But, how many rewards? Which is the best one?
If your "Simulation" has no "Fail State" that the player is really aware of, it's a lot more difficult for that player to decide when to "reload" and try again. I mean, they opened this chest and found 6 Elixirs! That's awesome! Surely that was the best prize! But, maybe it wasn't. Maybe there was an "Immune to every Element in the game" accessory in there they could've gotten too. But, they didn't know it was there. They have no way of knowing what was there. A bunch of reloads might even give the same result. Or, only a second option if they "roll" good enough (or bad enough). They likely won't know there were ever more than two rewards for any of these chests.
I think that's the key to doing this sort of simulation. No "Fail" state. The player is simply given something else and never told that they could've gotten something much better.
