Thanks for offering a different spin. Let me add to it.
Talked about this before, but indeed the perspective is a bit off.
Challenge is easy to understand. We have an obstacle and either numbers and or Mechanics, make it difficult to overcome it.
Fun is much more difficult to define. From my PoV Fun in ARPG’s with build and playstyle variety comes from overcoming the obstacle your way.
Now the greyzone is, how high the challenge should be and how much impact your playstyle can have.
I don’t think Moon has in any way stated, they want to become the toughest ARPG, with the most limited playstyles on the market. Have they? On the contrary, they have on multiple occasions communicated, that they want to introduce more playstyles into the game. They want variety. So one would assume, they want the playstyles that are available to be somewhat balanced.
Thus, the question to ask is: “How to balance?”
Balancing by difficulty is basically impossible and very restrictive, thus imo the easiest and most effective metrics to balance by are efficiency and popularity, NOT difficulty.
Identify what the most efficient, statistically relevant playstyle is (ignore select outliers piloted by hardcore gamers, until they become statistically relevant, unless truly broken).
If its too efficient, nerf it. If it is not, buff all other less efficient playstyles up to that level.
Also interact with the community and especially ask, if the most efficient playstyles are fun. Even if efficient, they might have to be readjusted.
Identify what the least popular playstyles are, figure out, why they aren’t as popular by interacting with the community, then adjust them. If they feel good to play, are FUN, but just lack efficiency, buff them.
In an ideal world, the devs would also communicate, what their goals are in terms of efficiency. “A casual player with X playstyle should be able to finish the crucible in on average X minutes”; “A hardcore player with a perfect setup should be able to finish the crucible in on average X minutes”. Calculating in losses along the way.
Currently we are restricted in terms of playstyles, we have playstyles that clearly outshine others in both ease of play AND efficiency if piloted well, while also having no clearly communicated benchmarks, neither about difficulty nor efficiency. Which is why this argument is a little bit to heated from my PoV.