I like the idea, but RTA and all its derivatives kill the strategic part of the game. After all, Heroes is primarily a strategy game.
When I was thinking about builds that would lead to the maximum number of horses (knights), I felt like a Korean Starcraft player, looking for the perfect timings for building certain buildings, in order to maximize the chance of successfully implementing a build (which also needs to be chosen from many alternatives). And you are talking about killing the strategic part of the game. The meaning of strategy lies in the efficient management of resources, and this is literally implemented through construction. In my opinion, construction is a much more interesting part of the game than even hero development, because development does not operate with the player's resources, while construction directly works with them.
For example, playing as Dungeon, you can go for a magic build, through beasts/witches/dragons, while discarding minotaurs OR assassins (you won't have time to build other units), or you can go for a maximum combat build, on assassins, minotaurs, lizards, and hydras under a Castle (you won't have time to build other units). And these are just 2 opposite options out of a hundred. Isn't that strategy?
You can play for early economy (building a capitol on the first week), sacrificing tempo for building creatures, you can ignore the economy on the first week (without a town hall), and instead quickly develop it on the second week, or you can not even build a capitol or even a magistrate, some builds are quite capable of playing on a low economy.
In general, I had an idea - to limit the maximum level of heroes from 40 to 20 - that would really make the ultimate hero decisive.
Specifically, on the arena, levels 12-15 are planned (starting at 12, the rest for money), rarely 16 and higher.
Note that only 15-16 "days" are given for construction, which means that a final battle on the 3rd week is simulated, and this is with full city development (not a rush). In this time, you can develop any thematic city build (by building 60% of all buildings). If you prolong the game longer, it is actually the endgame.
besides, I really don't like the fact that some network games boil down to just one battle (which players sometimes even agree on - like "are you ready? then I attack") and that's it... that is, it doesn't even look like a cooperative game - they leveled up in their starting location, ate on mobs, played one battle and that's it... that is, all the interaction between players is one battle, but where is the strategy here? - capturing territories, holding them, creating several armies fighting in several directions, and performing different tasks (defensive or offensive, for example), the correct placement of forces on the world map, luring into a trap with some bait hero or a seemingly abandoned valuable mine, etc.
Specifically, on the arena, such interaction is also implemented, at the construction stage, albeit only through weekly reconnaissance of the enemy garrison, but this already gives a lot of information about what strategy the opponent is going for, what creatures he is building, and what he is not, where he has upgrades, whether he is building a fast economy, or many troops, or going into magic, or going into early technologies (angels or horses on the 1st week).
And once again, I remind you that the arena is a transitional stage to a full-fledged map, just as duels were a transitional stage to the arena. The sequence of the WGE project development is as follows:
WGT (Wargame Tactics) - duels, troops, magic
WGA (Wargame Arena) - cities, heroes, artifacts
WGS (Wargame Standard) - full-fledged map
WGC (Wargame Classic) - a game with already ready armies and already developed cities (actually the endgame, a snapshot of the game, starting from the 3rd week)
I cannot jump to WGS, bypassing WGA, because it is in WGA that the balancing of heroes, artifacts, and cities will take place - that basis on which the Standard will subsequently be based.