The support patch came out on August 9th, and no part of it has generated as much discussion as the Mercy nerf. I’ve seen a lot of claims since then about Mercy’s state in the game that need clearing up. In order to address these claims from an informed position, I’ve looked at one month’s worth of data from Overbuff.
I separated the data into four weeks - August 9-15, 16-22, 23-29 and September 2-8 (Note: September 1 was the first day of season 12. Overbuff has no grandmaster Mercy data for that day at all. In order to keep the lengths of the periods the same, I opted to go for 2-8 as opposed to 1-7). In each week, and in each tier diamond - gm, I found the average pick rates and win rates for Mercy. I also calculated the long term average win rate for each tier among all characters (weighted by pick rate so it makes sense). I’m including the average win rate, because that’s around the win rate you have to maintain in order to stay in the relevant tier. For completeness sake, I’ve also presented similar data for all tiers.
The story in GM
Weekly win rate averages: 51.42, 51.33, 50.10, 49.31.
Weekly pick rate averages: 8.45, 5.84, 4.49, 2.69
Average win rate in GM: 54.5
The story in masters
Weekly win rate averages: 49.08, 49.01, 49.22, 49.03
Weekly pick rate averages: 8.56, 4.9, 4.92, 4.17
Average win rate in masters: 52.4
The story in diamond
Weekly win rate averages: 50.61, 50.29, 50.30, 49.83
Weekly pick rate averages: 8.31, 6.56, 5.62, 4.84
Average win rate in diamond: 51.5
The story in diamond
Weekly win rate averages: 50.61, 50.29, 50.30, 49.83
Weekly pick rate averages: 8.31, 6.56, 5.62, 4.84
Average win rate in diamond: 51.5
The story in all tiers
Weekly win rate averages: 49.05, 48.75, 48.81, 48.68
Weekly pick rate averages: 9.57, 8.01, 7.3, 6.7
Average win rate in all tiers: 50.5
Note: In diamond - gm, Mercy was the lowest win rate healer throughout the whole month. Usually, the distance to the next healer in masters-gm was about 3%. Mercy also spent the whole month in these tiers in the bottom 4 win rate characters period, almost always in the bottom 3, and usually bottom period in at least one of them.
Now, I’d like to address the claims I’ve been hearing based on this data. Broadly speaking, I separate them into 4 categories: Mercy is balanced, Mercy will be balanced, balance is meaningless, and Mercy should not be balanced.
Mercy is balanced claims: I’ve heard people say that Mercy is fine, that Mercy is in a good place, that she’s more balanced than she’s ever been, etc. These claims are 100% false. Mercy’s win rates are below the tier average throughout the SR spectrum. The problem gets progressively worse the higher up you go. Being 3% away from the tier average is incredibly bad, given that the entire spread of win rate averages of commonly played characters is about 6%. Statistically, if you are playing Mercy right now then you are deranking, and if you’re playing her in high ranks then you’re throwing.
Another version of this attempts to shield itself from reality by making the non falsifiable claim that Mercy is fine because she still has mobility, consistency, damage boost, and res. You can tell that a claim about Mercy is on its face wrong if you can make it verbatim regardless of what number her healing is changed to. But even more importantly than that, this kind of theoretical analysis is not very useful. The final arbiter of what is good in Overwatch is what actually happens in the game, and the actual numbers point out clear as day that Mercy is not fine. Just as a reminder - after the Ana nerfs pros thought she would always be fine because she still had anti heals, sleep, and high burst healing.
Mercy will be balanced claims: I’ve heard people say that people are still learning how to play post nerf Mercy, and that the numbers will get better once people figure out what maps/comps she’s good at. From the start, this claim is dubious. Mercy has been around since the game came out and, she’s not hard to figure out. Mercy should be good on maps with lots of verticality, and should be good in mobile comps that can exploit vericality, as well as comps centered on long range damage. I don’t think higher ranks players have any trouble at all figuring this out.
Looking at the data however, this claim doesn’t appear to be true. It’s been a month and the win rates show no indication of an upward trend. Mercy’s pick rates have already plummeted quite severely. People have already heavily restricted the situations they play Mercy in. They just haven’t found success. If this claim were true, you would have expected to see an upward trend when the pick rates decreased, especially in higher ranks where people make better strategic choices. It’s just not there. As a side note: the hero you are thinking of when you make this claim is Zenyatta. His pick rates have gone down, but his win rates have not.
In order to believe this claim, you need to believe an very unlikely constellation of ideas. This claim has grown a beard and needs a shave from Ockham’s razor. Its fig leaf of plausibility is shrinking by the day and it’s becoming indecent.
Balance is meaningless: These are claims of the form - win rates don’t matter/don’t mean much. Bulls have to eat quite a lot of food. They can’t digest it all, and so that food comes to an unpleasant and inevitable conclusion. A similar situation holds here. Win rate is the number one statistic that matters. The only variable that determines whether or not a character is good is whether or not they’re winning games. Pick rate is only secondary evidence towards this.
Now, that isn’t to say that one can just look at win rates in a vacuum. Some thought has to be applied to them. For instance, situations where a character is on both sides have to be taken into account. These situations will always cause a character’s win rates to trend closer to the tier average (For example, when a character is picked on both sides in every game then their win rates are the exact tier average). When a character is below the average win rate (as Mercy is), that means that non mirror matches are actually significantly worse than the win rate indicates.
Just because you have to think about something in order to speak about it in an informed way doesn’t make it meaningless. I suspect that many of you making this claim would not make it about other characters.
Mercy shouldn’t be balanced: These are claims of the form - Mercy is mechanically easier than Ana, so she shouldn’t do as well in higher tiers. Leaving off the fact that non of the other characters are balanced in this way, it’s a non sequitur. Mercy players don’t want to be better than Ana at higher ranks. They want to be viable there. When you are making this claim, you are really thinking about someone else. Moira is better than Ana at lower ranks, and worse than her in higher ranks. She’s viable at every rank. She’s also very easy to play (easier than Mercy in my opinion). Moira is the model Mercy players are looking for.
In any case, these are claims that can’t really addressed by numbers. They can only be addressed by love. If you’re making this silly claim, I love you anyways.
Data and claims aside, I have 3 requests to make:
Pro Mercy crowd: I’ve seen a lot of you retreat to the claim “Mercy is balanced now but boring”. When you do this, the other side just says that fun is subjective and dismisses you. Don’t be so hasty to retreat from reality. Mercy is not balanced. That needs to be fixed.
Anti Mercy crowd: Some of you are making honest well reasoned arguments. You have a natural and worthy position to take here - Blizzard should be cautious with regards to Mercy to prevent her from dominating the support roster. You guys should stick to points like this where you are in the right.
Blizzard: In trying to balance pick rates, you over nerfed Mercy. You know this. My guess is that you knew it all along. You will need to fix it eventually. The healing will need to go back to 60. Start figuring out how to do this without breaking the game. Valkyrie and res are both controversial messes. Do something to them, and make it happen. Also, despite my criticisms I love you guys too.
:
![:smiley:](https://d74zh8edaijub.cloudfront.net/en/overwatch/images/emoji/twitter/smiley.png?v=5" ":smiley:")
Most people don’t bother clicking links and arrows.