and they should keep it like that
I won't argue with you, but I'm curious as to why you think that. My feeling is that the XP is currently artificially inflated because of a difference in how the simulated battles played out at some time in the past. The logic for determining damage in the simulated battles is very different than in standard combat, so my feeling is that they might have had a situation in the past where simulated XP was much lower than you would get playing the combat out manually. Whatever the case might have or might have not been in the past, the way it currently works does factor in the max HP of the target and deal simulated damage to him, so ultimately the amount of hit points dealt to the enemy are about the same, so simulated battles give far more exp than played out battles.
I personally think this is bad for the game, because it makes it more grindy in a different sense. Rather that it encourages you to autoresolve battles to level up your troops rather than actually playing the part of the game that they've obviously put the most effort into. The actual combat of the game in the battlefield is something that works quite well and is very nuanced. The map and campaign and simulated battles is very rudimentary and still needs a lot of work.
Plus autoresolving battles causes the player character to fail to gain any experience as well.
Now, if this is because you feel it takes too long to level up characters in the normal way, I might agree with you. But this is more to do with the way that XP is handled, especially for troops, even moreso for troops that have a hard time managing to land hits and kills.
Plus leveling troops through autoresolving battles is quite silly. Looters will level your troops up ridiculously fast. Plus because of another bug they can not kill you, but again only in simulations Specifically, the first item in the list of non-civilian equipment sets they use has a blunt weapon, and when the game auto resolves the combat, it only picks from the first equipmentset, and since blunt weapons will never kill a target, autoresolving will never lose any troops. However, if you do actual combat, it will fill it with random looters, so they'll have pickaxes and hatchets and pitchforks and whatnot as well, and those CAN kill your troops.
You can get your guys up to tier 6 after a few combats with looters without risking death from any of them.
My feeling is that playing out a battle should at least give a similar amount of experience to auto-resolving it, so that you're at least not penalized for playing out the battle. I can see an argument against giving out less exp for a simulated battle, because it might make players feel like they need to play out boring battles that they would rather skip. But giving more exp for simulations means that players might feel forced to skip battles they would rather play, and since the actual battle is currently the most cohesive part of the game, I think that's a shame.
I think it would be best for a happy medium, that maybe the XP bonus is increased for both simulated and played combats, or that the XP requirement is reduced for leveling to some extent. Plus, I think looters should be fixed so that they can threaten death through simulated combat, but at the same time simulated combat should be improved so that things like unit type advantage and speed get taken into consideration. So for instance, archers get additional strikes against infantry, but shielded units get a high chance to block arrows, infantry strike more frequently than cavalry, but cavalry get hit less frequently than infantry and archers, cavalry don't get blocked by shields, but take full damage from two handed weapons and polearms. Archers can't fire after being hit by cavalry. So if you want to go against forest bandits, you bring shielded infantry and cavalry. If you want to go against desert bandits you bring some pikemen and maybe knights for horse archers, etc. This is still not perfect, but simulates the battle a bit more realistically.
Right now, simulated combat ignores everything except troop level and the weapon damage type of the first equipment set to determine whether they can kill or just wound. Armor doesn't matter, speed doesn't matter, shields don't matter, bows don't matter. The only question is what is the level of the attacker and defender, and when they kill someone, are they using a blunt weapon.