I’ll respectfully agree to disagree, as this could very well be subjective. More things to do in my opinion is a great measure of how engaging a hero is. Mercy 1.0 was a pocket healer who played passively. Mercy now can dive into enemy back line to save the over extending flanker, bunny hop to get further behind them keeping herself safe and forcing the enemy team to choose whether to sacrifice their healers who are about to get destroyed by a pocketed flanker or their tank line who are going to immediately get rushed at the same time since the enemies healers are distracted.
I’ll admit she’s a lot more forgiving now, but in my opinion that doesn’t make her less engaging.
Lucio ults in anticipation of, or during, a teamfight. Mercy ulted when an ally, or several allies, were dead. The most common use of Resurrect was on two players at once, and typically, at least two allies die every teamfight.
What is that math?
No, if we were to ignore usage delay, Mercy would have been using Resurrect once every minute and 5-10 seconds. 10/9.2 =/= .5
It is the case.
It is also the case that Resurrect was used more often than Valkyrie currently is, which you clearly had no clue about upon entering this discussion, thus demonstrating your knowledge level on this subject.
Those two assertions are not mutually exclusive. They are both true.
Offensive and defensive assists do not provide ultimate charge. Eliminations do not provide ultimate charge.
The only stats that provide ultimate charge are damage, healing, and damage amplification. Get that through your head, please.
You just said it wasn’t cause that math was bad… while it was your math, Now you’re saying it was the case. Like, do you even know what you’re arguing about anymore? At this point I could say the sky was blue and you would argue against it.
The statement above applies here as well. You’re actually arguing Moth Mercy ress’ the same amount on average and more/less than she used to res on average pre-rework.
Yet you made up a number of eliminations to take away from the ult charge of hero’s in your math. I am not the one who came up with your broken system. I never once claimed they did out of context of your system.
At this point, I’m done. This is getting ridiculous.
Excellent, I’m happy to discuss the merits/flaws of the model. I’ll readily admit it’s not perfect, but I think it’s better than you might give it credit for (though I would be happy to hear improvements).
It’s important to have a numerical framework when making numerical balance changes to heroes. Especially large numerical changes. Right now, I think the devs tend to chase their own tails and are really only looking at pick rates (which favor Mercy) without any real consideration of anything else.
I agree with you that the main flaw of the model (which I allude to in the OP) is that certain things are not taken into account (speed boost, anti heal, nano, the bursty nature of Ana/Moira/Brig heals, discord, CC etc.) As a result of this, it undersells Ana and Lucio by quite a bit - they’re better than the model predicts. It might also undersell Zen and Brig, but they’re already on top.
Mercy however, is encapsulated really well by the model. Her healing is not bursty outside of res, and everything else is accounted for. The model shows her squarely in the middle, not in need of buffs or nerfs. If she does need a nerf, it should be to the only thing that the model doesn’t handle well - the burstiness of res. I’d be perfectly happy with a nerf to res (I’m not a member of the revert faction).
Having said that, I think it’s a good model - it encapsulates as much as I could encapsulate regarding the heroes. Can you think of a good way to improve it? I’d be really happy with that.
One thing a good model should do is make predictions. In my way of viewing it, win rates are important prediction. Better heroes should win more often. I know you have some objections to win rate (I’ll quote them), but they don’t actually turn out to be problems as I explain below.
I think you might be looking at the wrong list. The model is the second list. That one has Brigitte way on top. In terms of winrates, the model predicts Brig, Zen, and Moira to have high winrates (Brig more than Zen), Mercy and Lucio to have middling winrates (Mercy more than Lucio), and Ana to have low win rates.
Mercy’s win rates are currently unreliable for the reason you mentioned. BUT her pick rate wasn’t nearly this high from late January till early May (it hovered from 6-8 for most of that period). And during that time her win rates were middle of the pack (below Zen and Moira, as the model
predicted, and below Brigitte once she was introduced).
Now, regarding Moira. Moira is the only healer that doesn’t fit well. It’s overselling her. You had the following objection:
The Overbuff stats for GM Moiras are 6,914 damage and 12,389 healing giving a total of 19,303 (higher than the all elos contribution for obvious reasons). GM Moiras are definitely not off alone dpsing, and the healing numbers are high. In fact, due to the fact that Moira’s healing potential is so much higher than her damage potential I’d say that any Moira with a really high number was a healer and not a dps.
The model actually fit Moira much better between January and May. I think the reason is that at higher elos Moira is currently only really used in one strategy (heavy multitank) and her fortunes rise and fall with the success of that strategy. She is (or the community perceives her to be) a bit too specialized.
Stop switching back and forth between the hypothetical section and the section that has averages. They aren’t the same thing.
And, nowhere in that post did I say that Mercy’s ultimate normally charges from 0 to full every 31 seconds. It’s pretty obvious that I was referring to the optimal scenarios for comparison purposes.
There was no untrue statement in that section you quoted.
“meaning that if Mercy is healing uninterrupted for 30.660377… (decimals!) seconds, her ultimate will go from 0% to 100%.”
That is completely true.
Now, will Mercy be healing for 31 seconds straight? No. The post even said that. You might have seen that if you weren’t selectively reading:
“Sounds powerful? Bear in mind these are under perfect circumstances for the Mercy, in which her team is constantly soaking damage, yet not enough for the enemy team to get a real advantage. The Mercy probably is not being targeted at all, as that would distract her from healing, and she may need to draw her pistol if she is attacked. Essentially, the teams would need to be in a poke battle with no ultimates being expended, no Biotic Grenades, no stuns on the Mercy, no picks, etc. Basically a standstill at the choke point.
For 31 seconds.”
It was math, and the post openly states that it does not represent how the game typically plays… as I already pointed out.
I have remained consistent. You are trying to put words in my mouth because you aren’t able to follow the painfully obvious trail of breadcrumbs.
That was not what I was referring to. This is what I was referring to:
When Mercy used Resurrect more often than Lucio used Sound Barrier. Both my player experience and statistics point to that.
They have the same number or revived players/game.
Current Mercy activates the ability more often than Mercy 1.x.
The assertions are perfectly fine coexisting; Current Mercy uses Resurrect 6.2 times/game on average, and revives 6.2 players. Mercy 1.x uses Resurrect 3 times/game on average (a random number to illustrate the fact that this number is flexible), and still revives 6.3 players/game. Resurrect on Mercy 1.x was able to revive more than one player for every use, remember?
Those statements are not mutually exclusive.
That number was not made up; it was taken from Overbuff. Because again, sources of ultimate charge do not provide ultimate charge to the player when said playing is using their ultimates. Thus, the damage from those eliminations would not count towards their ultimate gain, and needed to be removed from the equation.
I honestly find it kind of sad that I needed to spell that out.
Oops, you’re right I was. It’s actually interesting to see how well that list meshes with win rates. I do have to admit it makes sense, as Brigitte, Moira and Zen all have great heal and/or damage output. They bring measurable value with two of the most important stats in the game. I personally wouldn’t know how to improve on that other then the obvious of finding a way to calculate what isn’t included, which is easier said then done i could imagine.
I very clearly stated I was talking about your hypothetical, but continue beating on that strawman.
To which i wasn’t speaking on, as we were still talking about hypotheticals. At least, I was.
I mean, it was pretty obvious.
Do I even have to respond to the rest of this?
I mean, do you want me to link the previous post i replied to you again?
Your statistics are explicitly about the theoretical amount of times a hero can ult, to which you very clearly cover yourself with
Your statistics do not show this. And your personal experience has been more than proven bias at this point.
Is this your final answer? ‘cause you’re saying Mercy had mass res that often and came out to the same amount of res per game as she currently does. I’m inclined to believe res is relatively close to what she used to do. I am however fairly suspect of this whole ordeal mercy had mass res more times than lucio on average and still only had 6 res’ on average if it was used on 2-3 people most times. That puts it a 2-3 ults in a game, but you know.
This didn’t really help your case at all.
The made up number was 200, you took away from each dps hero’s damage who’s stats were tracked by overbuff under their ultimate kills. If you went that far you might as well have added assists into the equation. Also what about tanks. What about the damage sponge’s like roadhog. It was faulty, like i’ve been saying from the very beginning.
Then why even bother bringing it up? Yeah, I know that she didn’t actually use Resurrect every 31 seconds. That was stated in the post.
That quote was from the hypotheticals.
I mean, it’s literally right below the section you quoted:
"Mercy gains 4 charge points for every 5 healing done. With her healing rate of 60 HPS, she gains 53 charge points/second while healing (48 from healing, 5 from passive ultimate generation that every hero has). The requirement to use Resurrect is 1625, meaning that if Mercy is healing uninterrupted for 30.660377… (decimals!) seconds, her ultimate will go from 0% to 100%.
Sounds powerful? Bear in mind these are under perfect circumstances for the Mercy, in which her team is constantly soaking damage, yet not enough for the enemy team to get a real advantage. The Mercy probably is not being targeted at all, as that would distract her from healing, and she may need to draw her pistol if she is attacked. Essentially, the teams would need to be in a poke battle with no ultimates being expended, no Biotic Grenades, no stuns on the Mercy, no picks, etc. Basically a standstill at the choke point.
For 31 seconds."
I’m curious as to why you went to the hypotheticals within thay post it the first place. They aren’t relevant to how fast Mercy charged her ultimate relative to Valkyrie.
I don’t know, do you want me to link the previous post I replied to you again?
Those numbers represent how quickly those ultimates are charged. Being that Resurrect has a higher number and is given the same opportunities of use as Sound Barrier, Resurrect will be used more often.
How so? I’m intrigued to see how you might follow up on that claim to give it actual substance, rather than trying to disregard statistics and other players’ experiences with only your own personal experience.
Or maybe you’ll just pass over this section.
Okay, wow, that is some miserable reading comprehension.
Go look at all of those numbers in that post. Those do not represent how often those abilities were actually used. Those numbers represent how quickly players charge them on average. Obviously, a lot of that charge will be nullified by players holding onto their ultimates for a period of time. Those numbers tell us on average, how fast each ultimate would charge if the player used it as soon as it was available. They do not tell us exactly how many ultimates a player gets every game.
And again, being that Resurrect has a higher number and is given the same opportunities of use as Sound Barrier, Resurrect will be used more often.
You’d be correct. That’s what the averages say.
The numbers suggest otherwise.
Explain. Nothing in that paragraph is false or contradictory to itself or anything I have said thus far.
Because 200 is the amount of health most heroes have. Some heroes have more, very few have less, and there is a very high chance that those eliminations are only partial kills completed by someone else. As a compromise, I made a blanket change to the numbers to factor in ultimate kills.
Assists don’t provide ultimate charge. How many times does it need to be said?
You’re the one who can’t go a post without putting words in my mouth.
Well, she survived… as trash tier to a point Blizz wisely buffed her healing to 60 hp/s (yep, this is the third thing of Mercy that has been reverted for bad). Now she will survive for sure but just because she is bad design. Ez mode valk that gives all your work for free but can’t save anyone from anything (“Heroes die!” she should say, as her healing now is paired with Lucio’s amp it up) and busted and unbalanceable E rez. Both are the problem and both must go forever and never return. Rez must be Mercy’s ult because it can’t be entirely removed, that important is for her.
It’s ridiculous how desperately Blizz is trying to do the impossible for this design to work, when it was doomed to fail since the beginning because it’s based just on bad designs removed for good long ago.
The data does show she has the best healing in the game, even as a mostly single-target healer.
This is a nonsequitor, as it has nothing to do with the stats in question or the claim that was made. You’re just changing the devs words from “Mercy has the best healing” to “Mercy is the best support.”
Niether the stats, nor the devs themselves, claim that he is the worst SUPPORT, but the data absolutely shows that he has the lowest overall healing output. Stop shifting the goalposts.
This is all irrelevant when the topic is total healing output.
Then you can’t say her value would place well below Ana if not all her value has been accounted for. I have no idea how you got the value for Mercy’s Rez either since that’s another intangible. Why didn’t you at least use the same method to give value to all the other intangibles?
The thread title is “Mercy doesn’t heal too much (with stats)” yet you only proved that she does the best healing in the game (at best), or that she does indeed heal too much (at worst). Everything else was just shifting the goalposts away from healing, even though that was what the thread title stated you would be addressing. Note that this is all a critique of the thread, not a statement on the nerf itself.
But… what if we had the mass res as an ultimate again, without invulnerability, but instead a small damage reduction (say 30-40% damage reduction like Hog’s healing). And Valkyrie could be the 90hp/s just lasting 5 seconds and maybe with a 20 second cooldown? Possibly even longer, or lower the healing done on that single target slightly.
They did not say that the Devs said Zenyatta was the worst healer. We all know that he is a great healer in the way he doesn’t just heal (unlike Mercy). He does damage, and can cause an enemy to take more damage, and has a very powerful ultimate when used in the right hands.
Partially because of the fact that she only does healing people overlook the fact that all the other supports are able to do damage as well. Mercy has to pull out her pistol, risk toxicity, and risk her teammates dying because she decided to pull her pistol out. Plus, her ultimate’s chained healing isn’t lowered in any way. It stays 60hp/s on every hero that is targetted or being chained healed (which obviously needs fixed).
??
This person is trying to explain how Mercy is actually not overpowered healing-wise. Nerfing her healing will just make her even more of a res-bot.
This person is literally giving you multiple reasons as to how nerfing Mercy’s healing isn’t going to help or anything really. It doesn’t even matter at this point what the topic is. This person took the time to do math and research and you are ignoring the fact that Mercy is in fact not the most powerful healer if you take into account damage
Because, it would take too much time to guess how much Ana would heal her target. We would need to wait for an average to come out.
Mercy’s healing output is higher then others because she doesn’t do damage, all the other healers are able to do damage without risking everyone’s lives. It technically is a statement on the nerf, the person considered damage because supports bring something ‘supportive’ to their team (Ana’s grenade and sleep, Zenyatta’s damage and Discord orb, Brig’s damage output and stun, and Lucio’s speed boost, and damage while being able to heal.)
Sorry if this is confusing in any way, I really need sleep xD
(I’ve been camping for the Time-Lost Proto-Drake on WoW :c)
Thanks for the feedback. I don’t think you’re correct here. I’m not shifting the goalposts, I’m addressing the exact claim I made. I’ll try to explain why.
My claim is not that Mercy doesn’t heal more than the other healers. She obviously does (though at high ranks the difference between her and Moira is very small). The question I’m addressing is whether or not Geoff is correct in claiming that this is too much healing. What is too much? In this case, enough to justify a 17% nerf to her healing (it’s actually bigger than that - she should get less Valkyries a game and Valkyries add to her healing, so it might end up closer to the 20% range). This is not a total healing discussion, it’s a quantitative analysis of whether these total healing amounts merit the too much label.
If you look at the list of healing averages, you’ll see Mercy at the top and Zenyatta at the bottom. You might conclude from that list that Mercy heals for too much and Zenyatta heals for too little. Nobody makes the second claim. His low healing is compensated by how much damage he brings to the table. It’s not a non sequitur, it’s an illustrative analogy to show that judging a healer’s healing amounts requires you to look at their whole kit. Zenyatta doesn’t heal for too little, his low healing is part of a bigger picture.
What I set out to do is to take everything a character does to swing hp in a fight (at least the things with stats), and put it all together in order to get a big picture of total hp swing. I feel like this sort of numerical scheme is completely necessary in order to make large numerical decisions regarding any character. Otherwise any changes made are just arbitrary.
The picture you get in the end is illustrative. Zenyatta is no longer on the bottom, he’s close to the top. We can safely conclude that his low healing isn’t too little. His contributions to the fight lie elsewhere.
In fact, my list correctly predicts the win rates of all the healers except for Moira (and it did a good job with her between January and May).
Now, the list isn’t perfect. I mentioned that there were certain things that are hard to evaluate numerically (like the benefit of Lucio’s speed boost) or that I don’t have stats for (like how much healing Ana’s grenade prevents). This means that certain characters (mainly Lucio and Ana) are actually better than the numbers I gathered show.
But one character who can be almost entirely encapsulated into a number is Mercy. I say almost entirely because of res. Res in not intangible - the average hp swing for res is average number of resurrections * average hp of a character brought back. I have the average number of resurrections, but not the average hp of a returned character. The average hp of a non Mercy character in the game is 290 hp. I chose to value it at 250 because in my experience it’s much harder to res a tank than a dps. if you value it at 290 hp her number will go up from 14,873 to 15,101 but her place in the list will not change.
Mercy is square in the middle of the list. She certainly doesn’t need a buff. Res on E (which was the cause of this whole mess) might even need a nerf. But the claim that she heals for too much is without base. She does heal for more than the others, but that’s the vast majority of her contribution to a fight.
A 17-20% nerf on her healing brings her number to somewhere between 12,400 to 12,800. It’s lower than Ana, and that’s before Ana’s buff and before accounting for the fact that everything about Mercy is encapsulated in her number but Ana has a whole lot of large things going for her (sleep dart, anti heal, nano boost) that were not accounted for. It’s a monstrously large nerf to take a healer from the middle of the pack to way way on the bottom. And the given justification is not really grounded in the numbers.
