Mercy’s rework is a shining example of the exact opposite way to go about a rework. Resurrect was never designed to be an E ability, especially if it still has the restrictions an ultimate does. She can’t keep up with the other healer’s healing but hey, at least she’s consistent, am I right? Rez can’t even be used more than 8 times in a long match, because of how risky it is to pull, damage boost is only super useful on snipers because armor, Valkyrie is the least emgaging experiemce I’ve ever had in a game, including playing artillery in World of Tanks. She’s got nothing but survivability on every other healer, yet even that’s teammate dependent.
6 Likes
Not necessarily tempo rez but either mercy 2.1 or 2.2 had the slow rez and then 2 instant rezzes per ult which was my favorite.
Just out of curiosity, how many of you who are for reverting Mercy and ,rightfullly in my opinion, think the rework should not have happened, complained about Roadhog and/or Scatter?
I found scatter annoying but I never complained about it. As for Roadhog, meh, I didn’t even think he was a problem before.
Blizz, bring back mass Resurrect!
That’s fine,. I just have a feeling a lot of the people for a revert are partly responsible for what happened to Roadhog and they all helped set a precedent
I always thought the rework shouldn’t have happened. But my complaints about scatter is that it was inconsistent both for shooter and defender.
It made no sense that the move was most accurate by firing into the ground. At no point did I ever think he needed a leap, faster arrows, or a more murder fan the hammer.
My complaint with Roadhog is him melting tanks. I like that he doesn’t but I think it’s dumb they turned his combo into a 50/50 RNG for squishies. It made his mechanic iffy and let dive be free.
With the Winston shield buff squishies had that and other mechanics to get out of hook situations. Turning his gun into I hope everything hits ruins his enforcer/bruiser roll as a tank imo
My complaint with Roadhog is him melting tanks. I like that he doesn’t but I think it’s dumb they turned his combo into a 50/50 RNG for squishies. It made his mechanic iffy and let dive be free.
What? DIve was based on the tanks so what you seem to be implying makes no sense because the part that you didn’t like, but was statistically balanced, like Mass Rez, is what they removed and gave way to Mass Res.
There was no balance reason to change Roadhog, none whatsoever. Look at the reasoning:
Overwatch Forums
So you know what you helped do when you complained about Hog? You set a precedent. “Not ok with our playerbase”. Sound familiar?
The estimation is based off of actual math. It might not be 100,000 but it’s on that order. When the post says things like 98% confidence in a 0.8%, you would be justified in believing that it’s more of a 96% confidence in a 0.9% spread, but it’s not far off from that.
You don’t have to look very far for this. Go to Overbuff, and look at win rate averages before and after the introduction of private profiles. They fluctuated for a few days before people actually turned on public profiles, and then stabilized on the previous averages.
The math says that it doesn’t really matter. What matters is how many games Overbuff actually samples.
2 Likes
A) You are correct here. There is some selection bias in that Overbuff only displays data for people who look themselves up. It’s unclear how correlated this is to win rates, but if anything this should suugest that Overbuff win rates are slightly higher than in game win rates (people who look themselves up would tend to care more about performance). But even disregarding this point, the claim - among the 100,000 or so daily games recorded by Overbuff, the Mercy nerf has resulted in substantial decreases to win rates and pick rates is incredibly unlikely to not be a reflection of an actual trend in the actual game.
B) The post contains a math based estimation of how big a sample Overbuff is taking. It’s an estimation, but the central limit theorem says that it should be a pretty good one.
It doesn’t actually matter what percentage of profiles are private for this to work. All that matters is that private profiles are not correlated with win rates. You can tell this by comparing win rates before and after the introduction of private profiles.
To see the counter intuitive fact that the percentage of private profiles doesn’t matter, consider the following example. A coin is flipped 100 million times. You’re kind of distracted, so you only record the results of one million coin flips at random. In those one million flips, the coin landed on heads 48% of the time. You have a very high level of confidence that the coin is weighted, and that 48% is extremely close to the amount of times that the coin actually comes out heads. It might not be exactly 48%. It might be 48.1%, or 47.9%, but the probability that it’s 52% is incredibly small. It doesn’t matter if the coin was flipped 100 million times or 100 billion times. If you recorded a million results at random, you have enough confidence to make the claim.
The post does have some assumptions, but they’re not particularly big or strange. The numbers presented should be very close to what we see in reality.
Blizzard’s data is obviously the best data, but the Overbuff data is what we have and it’s pretty good. It’s good enough to base reasonable rational conclusions off of.
4 Likes
Dive jumped once Winston shield got buffed and Dva could dunk on damage. And Ana getting nerfed.
The game changed a lot. Dva got changed, triple tanked still ruled then Ana and Roadhog though the order I don’t remember it was two years ago.
But I do remember most of the dive heroes didn’t work back then cause they couldn’t do
enough damage or burst through bionade’s healing.
So chill on the outrage of what you think I am talking about.
Perhaps your missing the context of what I am saying. Roadhog at launch was a nightmare to squishies and Winston along with a back hook.
Now they changed a lot of things with him. Don’t remember the order but once his gun got buffed he became a must pick cause he could murder most of the cast.
My complaint about that back then was why would I ever pick Reaper when I got much better Reaper with health and hook?
And while I appreciate you think my post has that much power to control the devs it doesn’t. Cause my longest running most ignored complaint is lack of Guild/Clan system.
And Blizzard balance tends to be all over the place and wrong time. As I said a year ago it would have been more interesting if we had melty Zen, immortal bionade, and Tempo Rez along with Soldier Max dmg, super tank Dva and other versions of optimal heroes to see what truly sticks and remains broke .
But Roadhog’s gun could kill a tank or get them done to a sliver. Why would you play any other off tank or dps when he did that?
Its a failure on every concievable level.
1 Like
I think there’s a kernel of a good idea in the rework. Personally, I was pretty fine with it last month. I think that it could be changed into a fun, stable, and viable form for her with some work. It’s not there now.
I know how dive happenned. Secondly this:
Now they changed a lot of things with him. Don’t remember the order but once his gun got buffed he became a must pick cause he could murder most of the cast.
Is a lie along the lines of ëvery Mercy would play hide and seek. Statistically thius was not true. His gun was not buffed. After Hook 2.0 targets were pulled closer but his gun was not buffed
My complaint about that back then was why would I ever pick Reaper when I got much better Reaper with health and hook?
That’s a Reaper problem and not a Roadhog one. Guess what, Reaper is still trash.
And while I appreciate you think my post has that much power to control the devs it doesn’t. Cause my longest running most ignored complaint is lack of Guild/Clan system.
I didn’t say your post had that much influence. What I am saying is people like you who complained about Hog when statistically there was nothing wrong with him established a precedent for the same thing that happened to Mercy
But Roadhog’s gun could kill a tank or get them done to a sliver. Why would you play any other off tank or dps when he did that?
Considering that Dive was still meta even when he had his damage, you would play DVa for the best peel in the game, DM at 4 seconds at the time, along with mobility to contest high ground, to take engagements as you want etc. You know, the reason dive was so strong anyway.
If Roadhog was statistically a must pick, which he literally wasn’t then you would have a point. But he wasn’t. This is instead what happened:
Roadhog always goes flanking: Mercy is always hiding
Old Hog was OP: Mercy 1.0 was OP
This is exactly why even if I can agree with a revert, I cannot stand the people who post it, because I’m sure a lot of them advocated for nerfing Roadhog based on the same exact subjective BS that reworked Mercy.
When you claim he’s a must pick, when he statistically was not, you are doing the exact same thing as claiming Mercies were always hiding and getting 5 man resses. Do you see that?
1 Like
Just to stir the pot a little bit, I’ve actually been rolling this thought around in my head for awhile. I think something that might need to be considered in this situation further is what kind of players are choosing to have private profiles.
Not to make too big of an assumption here, but Mercy, as we know, is a controversial hero. There’s a significant amount of flack given against Mercy mains, especially over the past year. More than any other hero, Mercy and her mains get called out a lot for one-tricking or having a low hero pool. Thus, I think it has to be considered that a significant amount of Mercy mains have voluntarily made their profiles private, even after having used Overbuff. Overbuff might be a half-baked thought that they forgot they signed up for ages ago, and now to them, it’s better to keep snoops out of their profile in fear of being stereotyped.
But I also don’t know how Overbuff really functions behind the curtain. Do you know if this has any effect on it, or if it’s something else to be considered when approaching Overbuff? Perhaps I’m totally off the mark here and this doesn’t matter, but it’s been scratching the back of my head.
You make an important point. It’s not a priori unreasonable that private profiles should have something to do with Mercy win rates.
Luckily, we can check to see whether or not this is true (and I did prior to making my original post). Beyond the first few days (where not many people had their profiles set to public and the data was noisy), Mercy’s win rates didn’t see any real changes due to the introduction of private profiles. The introduction of private profiles didn’t seem to make much of a difference here.
3 Likes
We needed a 3rd rework the moment the cast time nerf was pushed through, but that doesn’t mean that 1.0 was salvageable.
Yes a rework WAS needed, however not particularly the one we got.
Wait lmao. Who took time and effort to flag this lol
4 Likes
Forums believe she’s a troll because they don’t like her videos.
Which is silly, cause you can disagree with someone without being hostile or flagging. /shrug.
1 Like
I hate your signature.
xox
oxx
xoo
~Sincerely yours,
A lover of bad, unfair and unfun balance.
Boosted trashcan 
P.S. Got carried away playing X’s & O’s. Looks like X’s won!