The way balance between factions is studied needs to be changed – a tournament should be created, something like "Castle vs. Necropolis on Jebel," where in each round, the colors are specified, for example, in the 1st round, red Castle vs. blue Necropolis, in the 2nd round, the opposite, and so on. Opponents switch colors (and, accordingly, factions) after each restart. This is to compare the factions themselves, and not the skill of their players. Based on the results of several dozen games and player surveys, it will be possible to draw more or less adequate conclusions.

It won't be Castle vs. Necropolis on Jebel, but Castle vs. Negro Gosh on Jebel. And that's a huge difference. Even if we assume that Necropolis = Negro Gosh, it still won't be a comparison of factions. Often, Gosh is played as a main hero for other factions (especially where there is no terrain penalty). And also, in the original template, the Necropolis villages and the Necropolis center were cut. This would, of course, be a plus for the Necropolis, but not a significant one.
When comparing these cities on Jebel, I would immediately pay attention to the 7th-level troops. How easy it is to get Angels or Dragons on day 1-2, and what dividends this will bring. How easily both can be built, and what an Angel is against a Bone Thrower, and an Archangel against a Ghost. What is the overall useful speed of the troops in movement points. And let's compare the offensive power of both cities, for example, against an Orcish army of Efreet-Sultans, which are often found on the Global Map... where it turns out that skeletons without Gosh are far from what you'd like. I've played a lot... you have to try very hard to lose with Castle, even with more or less equal starting conditions, and even if there's a slight advantage for the opponent. I thought that Citadel was a more or less real competitor, especially due to the combination of easy and fast construction of the powerful Gray Dragons and the infamous Crag :)