Really? Well then, tell me how YOU'D figure out how proportionally less-powerful you would make the cleric compared to the other classes, based on the assumption that they will possibly keep another player of X class in the fight for Y number of rounds compared to a party without one... when the party is 3 fighters / 1 cleric, 2 fighters / 1 cleric / 1 rogue, 2 clerics / 2 rogues / 1 wizard, 3 wizards / 1 cleric, 6 clerics, etc. etc. etc.?
I reject the premise that the cleric needs to be "less powerful."
Like I posted upthread, if every class has some way to avoid damage that would be vaguely equivalent to cleric healing, then the class distribution doesn't matter, since they all have about the same number of rounds between each full recharge. Cleric healing is different (and in some ways better -- you can use it on a wounded ally, for instance, instead of just using it on yourself like a fighter's parry or a rogue's dodge), but it's not necessary.
As for adding more PC's, 4e's maths give us away to address that: add an extra "monster" per extra PC between the recharges. It's gonna eat up more table time, so it gets unwieldy at a certain point, but 5e's streamlined battles will make that point high.
Oh! And don't forget to add in the fact that some of those parties will also change how many of those PCs will be in melee (and thus soaking damage) and others will be at range (and thus not). So it's not just 3 fighters 1 cleric you need to account for... it's 3 fighters melee / 1 cleric ranged, or 2 fighters melee / 1 cleric melee / 1 fighter ranged, or 1 cleric melee / 3 fighters ranged, etc. etc. etc. Because those variables dictate how many rounds of combat there will be before someone falls unconscious too.
This is all largely ancillary to the maths. The party, total, has a total amount of HP, distributed amongst its members, unevenly, and puts out a certain amount of total damage, distributed unevenly among its members, too. Ranged might create a situation with slightly lower damage taken for slightly lower damage output. Different HP totals create a situation where when the party's damage output drops until an unconscious character can be brought back, so such a low-HP character need only be balanced on the idea that they will be putting out extra until they drop. This is largely a question of how much swing you want. A game designed without much swing, like 4e, won't have major differences in HP totals between the characters, meaning that the differences in when one goes unconscious are minor, and so the differences in output is minor. A game designed with a lot of swing, like 1e, will have more dramatic HP discrepancies, and thus more dramatic effect discrepancies, but can still reasonably fit within the framework. Too much swing is probably not an ideal situation (you wanna trade being super effective half the time for having to sit out half the time?), but it's not "imbalanced."
And of course... let's not forget that there might be more than one monster. And those monsters might gang up on one PC, or might split up onto two. Or split up so that every member of the party is attacked by a single monster all at the same time. Cause that'll change if/when/how long before someone falls unconscious and thus lose the party man-rounds of doing/soaking damage.
Again, this is ancillary. It's all being depleted from the party's total pool of resources. If someone spends less time taking action, their actions merely need to be made more effective.
And what about how far apart these PCs are when they get attacked? How many man-rounds might you lose because the cleric is A numbers of spaces away from one PC who drops to 0, but B number of spaces from another PC that drops to 0? And we all know exactly how every player of every cleric is going to decide who to help first if two PCs drop unconscious at the same time, right? No variables there, I'm sure.
The action economy is how you distribute your actions. Fewer actions should create greater impact with each action.
So go ahead. let's do all that and come up with a concrete number that says a cleric needs to be 83% as powerful as a fighter (in both damage done, and damage soaked) to account for all the damage they will prevent and/or add by not letting his fellow PCs fall unconscious. That way we know that the cleric is ENTIRELY BALANCED regardless of their number in the party (from a party of all clerics down to zero clerics) so that the amount of combat encounters a table can go through before resting is EXACTLY THE SAME.
That way there is zero difference between having a cleric and not having a cleric and they can all do 5 encounters before resting (or whatever number WotC wants to use).
Should be easy.
I feel like you may not be understanding me. The goal is not for there to be zero difference or perfect homogeneity. The goal is for clerics (or healers more generally) not to be necessary to perform "optimally."
And it's not that hard in concept (though breaking down specific maths becomes more of a detailed exercise). Every bit of variation you included can still ultimately be reduced to a ballpark average of X points over Y rounds before you need to rest. Whatever X and Y are shouldn't change with the addition or removal of a single class, and whatever internal variation X and Y have is ultimately ancillary to their effect over an entire period between getting your own points back.
Hope that's been elucidating.
)... that is meant as mostly humor, lol 
