AMA responses Feb 21

#1 - Feb. 21, 2023, 8:32 p.m.
Blizzard Post

ok…gonna try to summarize the responses to AMA on here…might be slow going cause the AMA itself is dealing with issues…but anyway…here they are:

btw i cant reply to myself unless others respond first…so im going to shove them all up here unless others say something :stuck_out_tongue:

Is the quicker patch schedule (thank you!) relegating the Experiment card to history?

What’s the most insane change to a hero you’ve tried out internally?

Josh (Lead Balance Designer - Blizzard) —

The Experimental Card was much more useful in the past when the patches were further apart (months vs weeks). The turn around time to have it patched and reacted to between releases means it isn’t really feasible as a testing for upcoming live patches sort of thing, though hopefully we’ll try to find some fun uses for it again in the future outside of just patch previews.

The most polarizing changes are usually during the hero prototype or rework phase rather than anything for live balance. There have been a lot. The ones that immediately come to mind are Sombra Hack reducing max health for a minute, Wrecking Ball knocking down enemies he rolls over, and having an ability to tether and drag enemies with her mech (launching them tethered to her ult was hilarious and mean).

What is your opinion on Pharmercy? Is there any intention to make Pharah more viable without a pocket Mercy, and less annoying with a pocket Mercy?

Josh (Lead Balance Designer - Blizzard) —

this is an interesting one since I often play Pharah even without Mercy, so I’m probably biased there. Pharah is a hero that is tough to find buffs for (and she performs better on the stats side than most so it always becomes a question of if she even needs changes). She is quite situational as a pick but very capable of taking over the game in the right circumstances and is sort of a harder skill check for the enemy team than most heroes in terms of trying to counter her. I wouldn’t put that all on the Mercy combo, but it is very powerful.

We’ve considered potentially reworking Pharah to make her more interactive, but it would be a dicey one since to make big changes there, it would likely mean she has to lose the air time she is capable of. Her design started as the ‘rocket launcher’ hero but her identity quickly shifted to the ‘always flying’ hero. Losing that would likely alienate a large portion of her fans.

We’ve heard a few times recently that the hero design and balance teams are working toward reducing ‘pain points’ in the game (CC, one-shots, etc.). There has also been some criticism/observations that new heroes are much more flexible than older (more simple/“honest”) heroes. What have you learned from OW2 launch that will inform your hero design for new characters as well as reworks for older heroes?

Alec (Lead Hero Designer - Blizzard) —

We have been revisiting some of our hero design philosophies and seeing how they line up with OW2. One of the core pillars we have when developing a new hero is Counterplay in Mind. Heroes should have clear strengths and weaknesses as well as space for situational upside. Some of our recent heroes may have been doing too many things well on launch without those weaknesses to offset their strengths.

As we look into the future, it’s quite easy to look at the positive feedback surrounding added complexity in the hero kits and continue down that path. For us though, creating Overwatch heroes will always be about variety and making sure there are different playstyles with varying amounts of complexity. We tend to look at this over the course of the year, balancing those three hero releases so that there’s a healthy mix of playstyles.

Are there currently any updates you could give us on a Cassidy “rework”/updates to his kit? (It was stated on Twitter the team ultimately wants more utility for him) The buffs this recent patch helped him a ton in damage, but there are still complaints about his grenade being awful to play with and against. Are there possibilities for his flash bang to come back? Thanks!

Alec (Lead Hero Designer - Blizzard) —

This is something we are actively working on and have tinkered with iterations internally! The overall goal here is to bring back some lost synergy in Cassidy’s kit (ex/Fan the Hammer losing some of its ease of use). Magnetic Grenade is rather flat in what it does and while some abilities can be in this space Cassidy’s kit already has a lot of simplicity, making Magnetic Grenade a prime candidate for adjustments. We won’t be in full stun mode like Flash Bang, but are looking at ways to push Magnetic Grenade into an ability that either enables Cassidy to line up more shots or provide newfound utility.

Any info on Season 4’s new hero? We know they’re a support of course, but is there any other sneak peak we can get? :eyes:

Alec (Lead Hero Designer - Blizzard) —

Gonna have to wait a little longer, but I’ll echo something Aaron said earlier that the next hero will have really exciting ways of interacting with your team :eyes:. Can’t wait to share more soon!

Hi, and thank you for the opportunity.

I have two questions:

  • Is it still planned for Moira to get some kind of utility/soft rework or you are happy with the way she currently is? I remember reading or hearing somewhere around OW2 release that there are still plans for her to get utility?
  • Do you think that Mercy changes, to be more specific GA cooldown nerf was a good or bad change? You said multiple times that you don’t want core part of heroes kits and playstyle to be changed with buffs/nerfs so Mercy cooldown nerf was really a big one and it wasn’t received well by Mercy players mostly because they think that it made her feel “clunky” and is not synergizing well with the rest of her kit. Do you agree with this?

Once again, thanks!

Alec (Lead Hero Designer - Blizzard) —

I’ll answer the first question here, we’ll cover the Mercy questions in another reply :slight_smile:.

The Moira conversation can be tricky as she offers a sought-out playstyle for many Support players, but she still lacks the utility we want in the Support role. With the Necrotic Orb changes in the beta, it broke Moira’s flow and pushed her in a passive direction. So there’s some things we want to keep intact with any future changes: Moira still having this identity of being able to dish out damage (though maybe tuned down just a tad) and remain a comfort pick for many players as well as preserving her current flow. It’s a tough challenge so we will see where we end up, we’ll share more here as things develop.

Can you show unmirrored win rates for all heroes across all ranks? I always like seeing that data even though it is rarely shown.

Alec (Lead Hero Designer - Blizzard) —

One of the things we are currently working on is bringing the community into the fold more on the data we are seeing. We want to inform our players as best we can in order to have better discussions around balance changes and the reasons behind them.

That said, we don’t want to be in a place where we publish all current live data as that can have very unwanted meta effects. If players feel stuck playing meta heroes at times, having official live data could only further cement those sorts of feelings, even if players can have success on a wide range of heroes. Data only tells part of the story when considering the experience of playing with or against a hero.

So what y’all will continue to see are some data shared in livestreams, retrospectives that tell the story of hero performance over a season, and quick updates on the top performers in a particular meta. Hopefully these initiatives start to spark more fruitful conversations surrounding hero balance. We’ll also look at new ways to share general data around win rates in the future.

Why are Orisa and Junkrat so easy to play and yet get so much value out of their kits? And yet Reinhart and Wrecking Ball are hard to play with much less value…

Tess - (Assoc. Game Designer - Blizzard) —

Every character in the game is going to get varying amounts of value at varying time investments. Some characters, like Wrecking Ball, require a lot more investment in to start getting large amounts of value from. A player gets less value at the start because they’re being constantly exposed to new situations and aren’t going to confidently deal with them. As the player experiences more and learns from their mistakes, they start getting more and more value. Other characters get a lot of value at the start, but time investment doesn’t necessarily reward as much value increase as others. Not every character’s value increases over time is linear. Mercy gets a lot of value from new players because of the ease of using her Caduceus Staff, but the value for time investment will fall off fast unless players dedicate themselves to mastering her movement and positioning.

We think it’s important to have characters that have varying value curves for difficulty to serve different players. So many people who never played FPSes or even video games before got into Overwatch because of characters like Mercy. On the flipside, many players love devoting themselves to difficult or complex heroes, and we wouldn’t be able to create rewarding experiences for them without characters like Wrecking Ball.

What does the team think about heroes passives, and are you open to change them in the future?

Tess - (Assoc. Game Designer - Blizzard) —

Hero passives are super interesting because they often can define heroes and actively change the way the kit is played. Since many passives have fewer restrictions than active abilities (due to the lack of cooldowns), certain passives become the defining feature of the kit. Lucio’s Wall Ride is a great example of a passive that completely changes how a character would play. If he lacked wall riding, his style of supporting via harassment and displacement would be significantly hindered, and he would likely focus a lot more on getting value from Crossfade. It’s also just super fun to have unrestricted strong movement. Other characters’s passives might be less defining compared to Lucio’s, but they still can push the player towards playing in specific ways. Sombra’s Opportunist pushes players towards eliminating critically injured enemies, but it doesn’t define her gameplay loop like Lucio’s Wall Ride does.

When it comes to changing hero passives, I think everyone’s open towards anything that will improve the game, but there definitely are difficulties when it comes to changing abilities that players have grown attached to. We have more flexibility in adjusting role passives, which we’ve done a few times since our beta last year.

Many have noticed that games in OW2 are often very one-sided. Most of the community seem convinced its caused by poor matchmaking, but is it? How much of it is actually a byproduct of the faster and more punishing gameplay in OW2?

Are you looking at ways to make matches feel more fair outside of adjusting the matchmaking?

**Gavin - (Systems Designer - Blizzard) — **

It’s certainly a complex issue, but we do suspect that some aspects of OW2 can lead to the game feeling more one-sided than OW1 did in some cases.

We’ve been looking at all sorts of issues that affect how one-sided our games feel. We know that leavers cause matchmaking to feel bad in Quickplay because our priority is to get a player into the match as quickly as possible, and this doesn’t always mean the player that is backfilled is as good a fit for that match as the one that left. We think we can improve our process here in the future.

Addressing more punishing gameplay specifically, we know that getting back-spawned or spawn killed if your spawn room just shut down (due to an objective capture) are bigger swings since we have fewer heroes in a match. We’re working on a solution for these issues for Season 4.

We know that there isn’t a silver bullet for solving the issue of one-sided matches, but we’re attacking the problem from several angles simultaneously!

have you considered reworking the Coin Line challenge? it can be extremely tedious to win 5 arcade games, especially with Mystery Heroes being removed from the pool, and frustrating to play

**Gavin - (Systems Designer - Blizzard) — **

Coin Line is being replaced with “Win 5 games in Mystery Heroes, Free-for-All Deathmatch, or any Arcade mode” in the mid-season patch along with other Weekly Challenges changes!

Any word on fixing prior season’s ranks disappearing? Community level tournament need that player information in order to qualify their players for their tiers of competition. Anything you can do to help is appreciated.

**Gavin - (Systems Designer - Blizzard) — **

This should be fixed… right now! See our patch notes at Overwatch 2 - Patch Notes for details.

Last time we got an official rank distribution was in 2018 I think? Is there any shot we get a new distribution or confirmation that it’s the same in ow2?

**Gavin - (Systems Designer - Blizzard) — **

Due to the SR Decay system we launched with in OW2 our SR distribution curve was noticeably shifted left (meaning that there were more lower ranked players than higher ranked players). Most players did not play enough games for their SR and their MMR to converge.

In Season 3 we made several changes that shifted the curve back to the right (meaning that we moved back toward our intended distribution of ranks) to begin correcting for SR Decay’s removal at the start of Season 4. Fortunately, our MMR distribution curve (MMR being the value we use to make matches) has remained the same, meaning that our ability to predict outcomes and make the best matches we can remain unchanged.

What did you mean by "[Generally speaking we have plenty of players in all of the roles at almost all ranks]("?

Since there’s certainly some sort of scarcity going on, which the matchmaker is heavily compensating for.

Is it something along the lines of:

  • Scenario A. An abundance of Gold/Silver Support players, but not enough for Plat and above.
  • Scenario B. Enough players per role, but substantially different playtime per role.

**Gavin - (Systems Designer - Blizzard) — **

Feels like the community isn’t really aware of what the “root cause(s)” of these matchmaking issues are.

Neither A or B. Our wider matches aren’t because of any type of scarcity, they’re because the criteria we use to reject candidates are currently a bit too simple. We’re adding additional tuning knobs to correct this issue soon.

Hi Grey!!! :wink:

Also please please please make sr visible again. Matchmaking is miserable at the moment and it feels like I’m getting rolled every game, I’d love to know if I suck or if it’s the matchmaking

**Gavin - (Systems Designer - Blizzard) — **

In Season 4 we’re removing SR decay, meaning that in most cases your SR and your MMR will be the same so this should directly help with this issue. We’re also working on a way to display the percentage of players in your division that you are ranked higher than, so that you have a more accurate picture of your progress toward ranking up.

Currently, many low lobby players will see a “Masters Challenger” titled player and freak out, despite that title being earned in another role/mode. Has the team been thinking of ways to reduce this confusion, either by showing ranks or changing the title’s names to reflect the role/mode they were earned in (ex. Diamond Tank, Open Queue Master, etc.)

**Gavin - (Systems Designer - Blizzard) — **

We agree that confusion caused by these titles is problematic and we’re working on a solution for Season 4!

It feels like it’s super easy to progress in ranks at the moment.

4/6 players on my team have currently gone to GM in the past week, and none of us deserve it

**Gavin - (Systems Designer - Blizzard) — **

There are two issues at play here. The first is that we have reduced SR decay this Season and increased the rate that SR decay is removed when you win games in preparation for removing this system entirely for Season 4. This means that players are reaching their true ranks more quickly than they ever have in OW2. OW2 really does have a different metagame and emphasizes a different skillset than OW1 did, so it is inevitable that some players will find themselves higher ranked in OW2 than they did in OW1.

The second is that our first two seasons had a small amount of unintentional SR deflation, so we’ve mostly corrected that issue in Season 3. At the highest ranks (like GM where you and your friends are) we’ve learned we had an inaccuracy in our formula that resulted in an overcorrection.

Junkrat has been a menace in pub tier metal ranks for years stretching into OW1, but it seems he’s just become a gigantic pain in the a$$ at every single level of play. The mine change helped with some of his more salt inducing one shots, but are there any further changes the team is considering?

**Alec (Lead Hero Designer - Blizzard) — **

Junkrat has moved quite a bit since his latest nerf two weeks ago (down 3-4% at both the highest ranks and the metal ranks). We understand that there are further pain points in his kit though. One change we’d like to make but it will take some additional tech is: Junkrat’s grenades have a smaller hitbox after bouncing off a wall. We believe that type of change would alleviate a lot of the pain points when facing Junkrat and still keep his gameplay rewarding. No timeline though as it involves multiple teams.

what are you doing about match leavers? They’re constantly ruining metal rank games. Can we have a harsher punishment?

**Gavin - (Systems Designer - Blizzard) — **

We agree that we can do more to disincentivize leaving games, especially at the end of competitive seasons. We’re considering making ranked penalties partially carry over from season to season and we’d like to make sure that leaving games is never optimal for progressing challenges regardless of whether you’re in ranked or unranked.

I feel like the community sometimes loses patience waiting for balance changes because they do not fully understand what you are doing behind the scenes to balance the game (and/or create new heroes), and therefore how long it needs to take. What are the different roles in the team, the methods and timeline and the typical work yall do in order to balance the game?

**Alec (Lead Hero Designer - Blizzard) — **

I love these types of questions! So it depends on the types of changes we want to make. Let’s use two changes as an example here: Orisa’s recent Fortify nerf (125 to 75) and something like Brig’s upcoming ultimate rework (I’m going to try to answer this without revealing too much about the change).

Say we wanted to just do a seemingly simple change like the Orisa Fortify nerf. It’d start with the balance team making a decision then informing other groups (for this usually our Producer/QA/Live Ops) that a change is coming and will require a hotfix. Then a designer would go into Orisa’s scripts, make the change, and submit. At that point QA and Live Ops are notified, QA tests the change to make sure it’s all good and Live Ops will then get the build process going. We usually see that hotfix occur the next day, so overall a small change like this takes about two days (critical changes can happen faster but hotfixes have a chance to introduce a small amount of instability).

Now for the Brig Rally changes, these are changes that do involve other teams. So we may start off with a similar origin point, design wants to make a change. This change could involve various art disciplines (such as vfx, modeling, etc.) or sound, so now it’s important to see where we can fit on their calendar. Design mocks up the change, shows it to art/sound so we can get an estimate on how big the ask is and then hopefully slots the change in for a future season. Since our other disciplines are working on so many things, such a change may not happen the next season or even the season after – it’s dependent on how big your ask is, how much lead time design is giving the affected teams, how much bandwidth those teams have, and ultimately how high of a priority the change is. That process can take anywhere from a month to 6 months, so we have to be forward looking and work together constantly in order to deliver new things to the community in a timely manner.

I sometimes wonder how insane you people find the communities balancing ideas. And, do you balance with upcoming heroes in mind?

**Josh (Lead Balance Designer - Blizzard) — **

I love seeing the community engagement since it means players are passionate about the game. There are many great ideas that generate discussion for us or at least help inform us of the current community sentiments. The downside is that it can be disappointing for them when they don’t see their suggestions directly take effect in game and wonder why we aren’t using their perfectly calculated solutions.

Many players don’t consider the game balance ecosystem as a whole and focus almost entirely on their own experience with a particular hero and how to make that one hero stronger or feel better to play, which doesn’t always align with our goals as designers or what the stats imply about the current balance. There are always tons of other limiting factors like tech/schedule/implementation considerations players aren’t aware of so you really can’t fault anyone there.

For the second part of the question, we do think about upcoming heroes or changes and how they relate to the current roster. The recent examples being we didn’t make changes to Roadhog who was underperforming because we knew Kiriko cleanse could potentially make him more viable (it did) and that Junkerqueen ultimate was going to be really strong without the cleanse as well.

Hey guys, throughout my time playing Overwatch and watching people discuss balance I’ve noticed there is a theme of certain heroes being somewhat beloved by the community. I’m referring to heroes like Reinhardt, Winston, Tracer, Lucio and Ana who every one seems to be happy with being meta in ranked and open play at all times. Meanwhile there are heroes like Roadhog, Doomfist, and Moira that people absolutely despise ever being meta.

When discussing balance internally are these community sentiments about “evergreen” heroes ever talked about? Is there ever an attempt to make sure they are always toward the more powerful end of the hero spectrum? And how do you approach discussing and changing those less popular heroes? Is there more care involved?

Finally, why do you think some of these heroes are evergreen and others are not?

**Josh (Lead Balance Designer - Blizzard) — **

Great question! We’re certainly aware of the heroes that the community is generally more receptive to being meta though we don’t specifically plan around trying to make them the most powerful.

They are usually already popular heroes that will be played regardless of their underlying strength at any time. Many of these heroes have an advantage of having been available since the original game launch and are more accepted because of it.

As a fun exercise, imagine some of the launch heroes being released today and what the community reception to them would be just seeing the design/abilities on paper.

Our primary goal with hero balance is to try and get every hero to be viable at the top end of the competitive ladder (Masters-Top500, within a range of statistical viability since it’s a measurable goal). We also try to keep a handle on any significant outliers at lower skill tiers as well as large community pain points.

How are you feeling about snipers at the moment? Sometimes just feels oppressive playing against a good widow and that with one tank it’s not as simple as getting a dva or Winston to dive on them. Are players effectively able to counter snipers without simply having a better sniper now in 5v5?

**Josh (Lead Balance Designer - Blizzard) — **

I was looking at Widowmaker stats just last week, mainly for map balance discussions, but it was interesting to see that with the current map pool there were 4-5 maps where Widowmaker was really effective, 4-5 maps where she is quite bad or rarely picked (presumably because she’s not good there) and the rest were that she was playable, but not particularly great.

For anyone wondering her worst maps were 2 of the Push maps and 3 Control maps. Her best maps were 3 Payload and 2 Hybrid maps.

Snipers feel quite counterable to me at the moment but it’s certainly something we keep a close eye on to make sure they don’t get out of hand. I think the instances where snipers feel oppressive to play against are really memorable as it’s typical of human nature to remember the bad more than the good. At least for me, it’s rare that I remember when the enemy Widowmaker was ineffective and swapped off versus when she was dominating.

How often do you consider introducing new abilities to a hero’s kit like you did with’s micro missiles or hanzo’s jump?

And what heroes do you believe are in need of this?

**Alec (Lead Hero Designer - Blizzard) — **

When looking at adding to a hero or reworking them, we first ask: can we achieve our desired gameplay changes in their current kit and abilities? The GA meter for Mercy in OW2 is a good example of adding something new within her present abilities. If we can’t solve for desired gameplay within what we have, we’ll next ask if there is room on the hero for something brand new. We are currently asking those questions for Sombra and Roadhog. If we end up doing new abilities for either of them, it’ll take some time and I’d expect small holdover changes like the ones in our Season 3 patch.

What is your philosophy around using healthpools to balance non-tank heroes? More specifically, do you forsee any other heroes (existing or future) being balanced using healthpools <200 HP?

**Tess - (Assoc. Game Designer - Blizzard) — **

Varying health amounts is a great tool to utilize when designing or balancing a hero. Characters like Reaper and Mei, who thrive in close-range combat, can utilize their larger amounts of Health to help mitigate the risk of approaching close-range fights. Varying health amounts can also help provide flavor to a character. Tracer has a low amount of health because she’s quick and nimble, yet Torbjörn has more Health because he’s heavily armored. The last balance patch lowered Widowmaker’s health to 175, which shows we’re considering health amounts when it comes to balancing the heroes overall.

Hello. I’ve been wondering how Mercy’s pick and winrate has been since her changes? How is she finding herself in the hero roster after these changes?
Any changes planned for GA cooldown?
Any other changes planned for her?

Thanks for always pouring your heart in the work.

**Josh (Lead Balance Designer - Blizzard) — **

Mercy stayed about the same as before, the changes were mostly neutral in terms of power. Her win rate is still around 51-52%. There was a small drop in pick rate from 9 to 8%. Her ‘damage taken’ stats increased by like 4 or 5% but her average deaths stayed the same – likely we can attribute this to the new passive at work and taking more damage due to the longer GA cooldown.

We don’t have further changes planned for Mercy at the moment. We feel it would be good to let this one settle and see how players adjust to the changes since it can take some time, and then evaluate if more adjustments are necessary. The meta is also quite different between the two patches so it’s even more surprising that Mercy wasn’t really impacted overall.

Despite the stats not changing much there have been both positive and negative effects on the player experience. There has been positive sentiment around players feeling like they can more fairly try to counter Mercy in ways that were difficult with her extreme mobility before as well as Mercy players having more agency in their ability to save low health allies. However, at the same time the reduction in mobility is going to feel bad for Mercy players needing to adjust to the new timing.

Forum Avatar
Community Manager
#115 - Feb. 22, 2023, 12:50 a.m.
Blizzard Post

Thanks for putting this together, Baja, gonna pin this for a little while so everyone can catch up. Also expect one or two late answers in the Reddit Thread coming in so keep checking!

Forum Avatar
Community Manager
#118 - Feb. 22, 2023, 12:56 a.m.
Blizzard Post

FYI, we had allotted 11:00 - 13:00 for this (meaning we blocked time out on the team’s schedules to dedicate to the AuAA). Since we still had technical issues by the end of that time, the team is posting responses as they can between meetings throughout the day, including this new update from Josh about Mercy.