The party's policy is to balance the castles in terms of strength. Not necessarily to achieve perfect balance, but so that it's possible to play normally with any castle against any other, perhaps with a slight disadvantage (<2000).
The second point of the party's policy is to have fewer lines in the changelog. To avoid having, as in HotA 100500, many small changes that have little impact. This especially applies to the standard SoD functionality. If a change isn't necessary, it's better not to make it. For example, the prices for the psionic creatures, bird creatures, and dwell bird creatures in SoD are clearly too low, so they were raised without much hesitation, because it's unlikely that they will need to be reverted.
The third point of the party's policy is to avoid making too many changes, introducing and reverting features, and raising or lowering numbers. The strategy here is to first introduce all the conceptual features that shape the balance, then test them in tournaments, and based on the results, slightly adjust the numbers (prices, starting quantities) up or down. Again, the requirements for standard functionality are much stricter.
Well, regarding the second point, not quite. If it's possible to achieve a balance shift either through a set of small changes or through a radically drastic change in gameplay, the preference is given to the former. Rather, the second point is only true to the extent that it follows from the third – we really don't want to be constantly changing things back and forth.
Now, regarding the Shakti issue: in HotA 1.2, it had SoD stacks, and there was a feeling that this dungeon, which was already one of the fastest-paced castles in the game, was now possibly too strong. But it wasn't possible to turn this feeling into a specific nerf. Then, Vik joined the team and, as expected, pointed out this issue and suggested already proven versions – the ones that shall not be named – and raised them to the average SoD level. Everyone agreed that the average SoD pace for the dungeon is quite satisfactory for HotA, especially since in SoD it's actually less, and people weren't complaining (and they don't always play full SoD with experience). Yes, HotA is not SoD, and the same new items can also affect the balance, and we will look at the feedback, but that's how the numbers were formed.
The fact that 23-24-26-27, etc. troglodytes come in stacks is absurd. If aesthetics are so important, then there should be only 20-25-30 troglodytes, according to the logic of the developers, which will remain a mystery to me until the end of time (regarding the aesthetic beauty of the numbers).
It's not so much about aesthetic beauty as it is about the integrity of the game mechanics. There are standard quantities in the stacks of the first level, with a small exception in the form of centaurs, and in order to introduce some kind of intermediate inconsistency, there needs to be a very good reason, which there isn't yet. If it turns out that, oh no, 20-30 is too few, 25-35 is too many, then a reason will appear.
In general, I don't see anything wrong with slightly increasing the stacks of Shakti if there is a general opinion that the dungeon is underpowered (including from players who specialize in it). There will, of course, be no democracy with voting, but all feedback will be taken into account. So yes, the opinion doesn't go into the void.
Cassiopeia was reduced from 120 to 100 with the argument "it's too strong!", and troglodytes are still a more useful unit than nymphs.
Um, I wouldn't rush to conclusions). There is a very thin competition between troglodytes and nymphs, and I can't immediately give preference to one of them. Greater melee combat ability and reliability versus greater mobility and damage... It strongly depends on the situation