Thanks @Papa_Marsh, I admit there are quite a few drawbacks to the idea I put forward there.
I agree that I would be accepting that hero A is more or less useful than hero B and that’s a bummer because it means, for campaign and non-PvP, there is little reason to upgrade hero B vs hero A. Strike against the idea on that grounds. For PvP, however, I don’t think this particular issue applies. Specifically, because the power values would now reflect the actual value of the hero, hero B would actually be more viable than previously. If you have a 5th slot open, you could put the ‘better’ of two heroes (hero A) as you currently would, when they have the same power value, or you could put the weaker of the two because it has a far lower hit to your overall team power and you might be able to use that hero better than the average player.
As for team composition, I think I may have been unclear. The power value of a hero does not change based on the team they are on. At its simplest version, the proposal is simply a monthly calibration of hero value based on how often a hero is used (regardless of team comp).
Again, I admit the solution is probably not terribly ideal as is but it was an attempt to shift the large amount of manual work required in seeing which heros and skills need buffs and, crucially, whether they need a buff at each star level or only at the low/med/high end. Even if that work were to be done, the reality is that, in a vacuum, it’s near impossible to tell whether a hero/skill is under/over powered and by how much, which is why it would be such a perfect candidate for a market-based solution since Efficient Market Hypothesis provides such a strong mechanism for identifying value.