Pick rates are not entirely explained by power level. If they were, you’d expect Brigitte to be the most-picked support based on the complaints all over the forums, wouldn’t you?
If pick rates are indicative of power level, then Reinhardt is the most overpowered hero in the game right now and needs nerfs yesterday! …but you see how silly the idea of ‘nerf Reinhardt’ sounds?
Thing is Brig winrate is dramatically high and we know Rein pickrate is dependant on external factors.
Aha! So pick rates are not necessarily something you can even blame on the hero who’s getting picked so much? Perhaps, then, if you don’t think Mercy’s high pick rate is because of her aesthetic appeal, you can see how the double sniper meta favors a healer who can both reach the snipers and resurrect sniped allies?
Mercy’s pick rate has been as much a product of the meta around her as was is a product of her own personal viability. Hanzo, specifically, forced teams to run Mercy/Zen just to counter and support that one hero because he was so ridiculously overpowered.
Nerfing Hanzo and Widowmaker without touching Mercy at all would have lowered Mercy’s pick rate.
I think Mercy’s pickrate was high because she had no weakness when it came to Utility, Survival, and Healing.
Which is why I told the devs to reduce her healing, because it made more sense to do that than to reduce Survival/Utility.
Mercy has a weakness: she doesn’t contribute significantly to the team’s damage output. Even with damage boost included, her damage output is far, far below any other support hero. Damage is a kind of non-healing utility, and it is a utility that other supports can better provide.
Mercy is not the optimal pick when your team needs more damage output.
Which is kinda like comparing who is the tallest dwarf.
Every healer has weak DPS besides Zen.
In games like this people don’t care about aesthetics, its what makes you win more games, that website is accurate
Given the difference in damage output, it’s more like comparing dwarves to ants.
Mercy does not need to do damage, she amplifies the damage of someone that does monstrous amounts of damage, add that to her damage and it looks a lot different.
Still mostly irrelevant to overall balance on healers.
I guess that’s why Ana was a throw pick at all ranks after they nerfed her damage to 60 per shot, huh? At least she was kind of okay-ish in the right hands once she got brought back to 70. Damage absolutely matters. One of Moira’s biggest contributions as a support is the ability to finish off low health enemies, just like one of Ana’s biggest contributions is her ability to 3-tap squishies at long range.
And try playing Lucio without using your primary fire. See where not doing damage as a support gets you.
Moira’s damage output average per-game is about 6 times higher than Mercy’s total damage contribution including damage boost. They aren’t even in the same league. Nobody is. Mercy is all alone down at the bottom of that competition. Completely removing Mercy’s right click from the game would be a terrible idea, but in terms of affecting Mercy’s power level it would be insignificant compared to the impact of the healing nerf she received.
That’s a fun piece of trivia, but like I said, I don’t think it matters much when compared to Utility, Healing, Survival.
If you really care about it, you could class it inside Utility and Survival.
Damage is part of a support’s non-healing utility. Healing itself is also utility, it just doesn’t often get called that because it’s generally thought of as a primary function of the hero… but when you consider Soldier: 76’s biotic field, thinking of healing as utility makes a lot of sense.
When considering how much healing a support should be able to do, all of their non-healing utility, including damage, needs to be taken into consideration. Because Mercy has far less non-healing utility than other supports, she needs to be the best healer in the game.
Just note that “best healer” and “best support” are two completely different concepts.
So, the other reason I picked the 50hps nerf was because it would be sure to hit higher ranks, without hitting lower ranks.
Mostly because lower ranks rely on Survival for teammates that don’t peel for them.
But also because the slower fights mean more time to heal, and less tanks which are harder to heal.
Well… as it turns out, the change hit lower ranks too and the difference between the effects on low rank and high rank Mercy players is only marginal. As I noted in my edit of the OP, Mercy was even the worst support in Bronze when I went to look last. Ana is of course fighting her pretty hard for that title.
It makes sense when you think about it, though… Mercy in high ranks heals more than Mercy in low ranks, but it has very little to do with mechanical issues the likes of which Ana must face, which is what people normally assume separates the SR divisions. Instead, it’s about survival, selecting the right target, and being in the right place at the right time to maintain the uptime of the healing beam as much as possible. Low rank players have less healing beam uptime, but the healing nerf reduces their effective healing by the same percentage as it reduces the effective healing of a much more skilled player.
In addition, Mercy’s value to the team is only as good as the player she chooses to pocket. In lower ranks, Mercy’s value is inherently limited by her own teammates and her win rate down there has never been good. Though it’s possible to climb the ranks with any hero if you’re really good enough, Mercy is definitely a sub-optimal pick for a player who wants to climb quickly and has the skill to do so. That has always been true. That’s even more true with her nerf, but it’s not going to stop people in low ranks from picking her because a good number of them don’t even care about winning. They just think she’s pretty. 
If you want a change made to affect Mercy differently at different skill tiers, then it’s going to need to be something better than a flat change to her healing rate. Instead, she needs changes to her ultimate and abilities. She needs changes that reallocate her power into resources that she has to manage and make hard decisions to use, just as she used to have with Resurrect as an ultimate.
Huh, looks like Ana passed Mercy in Diamond/M/GM on weekly values.
Yep, because Diamond/M/GM players are the first to now recognize that Mercy’s healing just doesn’t cut it anymore.
Win rate dropped first. Now people recognize they are losing when they pick Mercy. Now, pick rate will continue to fall.
And I’ll be here just to say I told you so. 
Didn’t I tell you the point was to hit higher ranks?
Mercy is still #1 on weeklies at Plat and below.
Pick rate alone doesn’t mean anything without context. Her pick rate is still higher in lower ranks because those ranks haven’t yet realized that they need to swap off of Mercy to win more. It has still fallen across the board, and I would not be surprised if she falls out of favor at all ranks before the numbers settle.
Pick rate is correlated with the appeal and perceived power level of a hero. It is not correlated directly with the actual power level. It’s going to take time for everyone to realize just how weak Mercy is now, but even when they do some people will still refuse to switch because the hero appeals to them in some other way, beyond increased chance of winning.
You did say that Mercy would get hit hardest in higher ranks, and I suppose that I would say that statement may be technically correct. I should acknowledge that. However, I don’t think that the difference between how hard the change hits high ranks and how hard it hits low ranks is as significant as you would like it to be, and I also question why you want Mercy hit in high ranks anyway. Her pick rate before this last patch was already lower in GM than anywhere else.
Not on PS4, where both Ana and Zen have dramatically lower win rates and pick rates in GM.