Hex Mask UI Icons Arrow Down Arrow Left Arrow Right Arrow Up Brackets Calendar Checkmark Chevron Down Chevron Left Chevron Right Chevron Up Clock Close Connect Copy crown download Ellipsis External URL Favorite (off) Favorite (on) Fullscreen Help Home Info Live Minus Mobile Menu Mobile Overflow Paginate Dot (filled) Paginate Dot (empty) Pin Placeholder Icon Play Video Circle Play Video Plus Search Team No-Team-Logo_Fallback_003 Ticket Trophy User Input Video Video Play Icon World X Game Mode Icons Hero Icons Map Icons Platform Icons Exclusive on Battle.net Social Icons Mobile Discord Facebook Instagram Reddit Twitch Twitter VK Youtube Weibo WeChat Workshop Icons WorkshopControl WorkshopShare WorkshopInfinite

Inside Overwatch: Balance Design and the Experimental Card

Inside Overwatch: Balance Design and the Experimental Card
“A perfectly balanced game doesn’t mean it will be fun. Design something fun and then balance it, it’s often more difficult the other way around.” – Josh Noh, Overwatch Game Designer

As any transcendent omnic monk will tell you, finding the path to balance is no simple task. Thankfully, the Overwatch team has a buddy-cop-duo of designers dedicated to finding the sweet spot where everyone feels powerful, but not punishing. Geoff Goodman, a 20+ year Blizzard veteran teams up every day with Josh Noh, a hotshot top 500 player, to help create and balance the living landscape of Overwatch’s Hero design. 

Recently, Geoff and Josh had some ride-along additions to their two-man team in the form of professional Overwatch League players, streamers, and content creators. As some of the most active members of the Overwatch community, these lucky few donned their designer jeans (...get it?) and were asked to takeover for Geoff and Josh, suggesting their own changes to the game. The goal: achieve the seemingly unachievable--perfect balance. After that, test it out in the Overwatch League Flash Ops Experimental tournament! Easy enough, right?  

We stole some time from the dynamic duo to find out more about what went into the experimental card process, hear some stories from development, and to get an apology for those who recently took up their mantle: 


Josh: I started at Blizzard in 2012 working as a QA Analyst for World of Warcraft looking at combat and raids content. In 2013 I moved over to Overwatch, because they needed someone to test hit detection and hero ability mechanics. My team leader at the time had previously been working on Project Titan (Editor’s note: this was the MMORPG that the Overwatch team was previously working on) and recommended me to the position since he knew I had a history of playing FPS and PvP games at a high level.

While working on the Heroes QA sub-team for Overwatch, I sent tons of feedback and analysis emails to the design team from my perspective as a GM/Top 500 player. Eventually they must have gotten sick of it and gave me a shot to make some heroes in 2018 as a game designer.

Geoff: I started in tech support, answering phones back in December of 2000. Eventually I got onto the World of Warcraft team somewhere in 2002, I think. I worked on dungeons and raids, as a spawner and encounter designer. WoW was a blast to work on, but I couldn’t resist an opportunity to work on a shooter once I found out one was being developed. I worked on Titan until it was cancelled, and then started working on Overwatch right after as the sole Hero Designer.

I’m not sure of any specific gaming flexes I have, other than the crazy breadth of games I play. I have about 600 games on my steam library alone, almost all of which I’ve actually played. (Editor’s note: wow.)

Josh: I don’t really have an explicit philosophy but I can share a couple general considerations that often come up. A perfectly balanced game doesn’t mean it will be fun. Design something fun and then balance it, it’s often more difficult the other way around. Perception of balance can be more impactful on the state of a game than the “true” numerical balance. This is especially true in competitive team games as social pressure is a significant factor. Be aware of your own biases when analyzing things. Put the gameplay and player experience first.

Geoff: Josh hit it perfectly. The only thing I’d add is that the fantasy of a design is important to reinforce, as well as the gameplay. This is especially true of heroes, where there is a ton of personality and character that the gameplay should also reflect. So Widowmaker feels slow and calculated, but Tracer feels fast and snappy, and Junkrat is just pure chaos.

Josh: I’ll definitely check out the tournament. I’m always interested in what team comps and strategies people will try based on balance changes. When I first saw the list of changes, I was surprised at how responsible the panel was being with the actual values, it was hard to get crazy changes past them apparently. In a lot of cases, they may have gone too small with things like 5% damage reduction, 15 more armor, or 5 increased damage on Junkrat mines, etc. Those kinds of values are very difficult to feel any difference in game, though occasionally they might make a difference.

Geoff: I’m going to watch for sure! I have some guesses as to what the comps will look like but I’m looking forward to seeing what comes out of it.

Geoff: I had a lot of fun doing this, even though my job was pretty much to just sit back listen. It was nice to just be a fly on the wall for a change. I don’t remember anything like this ever being done anywhere at Blizzard actually. It's nice to have the experimental card to have an outlet for stuff like this! For specific tuning numbers, we did a secondary pass in Discord with all the panelists to make sure everyone was happy with the result. The one thing I feel bad about though, is not really thinking about the personal backlash the panelists might get. I’m so used to getting that kind of critical feedback, as its really just a part of the job, but for new people to come in and take on the daunting task of balancing/tweaking a game like Overwatch, there was inevitably going to be some of that headed their way. I hope all the panelists didn’t get too overwhelmed by reading that kind of feedback to their changes!

Josh: We playtested the changes internally just to check them out. I immediately tried out the double Helix Rockets and double Torbjorn Turrets since they sounded strong, but I wasn’t sure if they would be in a real game. Then I saw the tiniest Wrecking Ball zoom past me and HE WAS SO SMALL! It was a lot of fun.

Geoff: There is a bunch of super interesting stuff in here that I’m looking forward to seeing tested out. High on my list that I’m looking forward to are Baptiste immortality field 1 health minimum, Echo’s Duplicate change, and Widowmaker’s Venom Mine big buffs.

Josh: I always filled for my team and ended up not having a main. I try to play everything to varying degrees of success. It’s hard to pick just one WILD change with so many heroes. I’ve narrowed it down to Doomfist Rocket Punch/Tracer Blink can be aimed vertically, a new Zenyatta passive that lets him float freely into the air, or Mercy can Guardian Angel to enemies.

Geoff: The one that comes to mind right now is giving Widowmaker and/or Wrecking Ball multiple charges of grapple. This would certainly fit the bill of wildly unbalanced, but damn if it isn’t fun to grapple multiple times in a row like Spiderman.

Josh: Feedback discussions from all sources can be very valuable to see for a designer, especially when framed as detailing the symptoms: “here is this problem I’m facing and why I think it’s a problem”. Proposing potential solutions doesn’t hurt but usually isn’t very helpful since there are a ton of internal considerations and priorities that the player won’t be aware of.

Geoff: We gather as much feedback as we can from everywhere!

Geoff: I think I was able to implement all of them in about a day. After that we ran a playtest to see how it played and check for any bugs at the same time. It was interesting to do our normal feedback session after but not really go on to iterate on the changes as we normally would. For our own changes, it can really vary how long we iterate on them before trying them out live. In some cases, like the Moira fade experiment we ran a while back, it took a bit of iteration time, and we still weren’t sure about it, but it was in an interesting enough place to get feedback from the entire community. That’s why we decided to throw it onto the experimental card to see what people thought. For more simple changes its just about putting them in and running some playtests and seeing how they feel, which generally takes about a week.

Josh: Often when we’re making something new there aren’t real assets for it yet from the artists, but the designers will put in a placeholder effect in that they’ve made to convey the idea. For Hanzo’s Dragonstrike ultimate it was giant debug circles that looked like a caterpillar, with spinning googly-eyes stuck to the front so you know it’s the head. That cracked me up. Similar to that, when Lúcio’s Jazzy skin was first going in it was supposed to have different music than the base skin, but since it wasn’t ready yet there was a Scott Lawlor voice recording that said “Jazzy speed song, go go go.” Every single time you swapped to the speed song. It was amazing, I wish we added that as an unlockable voice line. (Editor’s note: Can’t we?!)

Geoff: There are so many of these it’s hard to even just pick one. One that comes to mind right now though, is Lúcio’s current ultimate was actually his third ultimate we tried for him. The first one was kind of a ‘time stop’, or more accurately a ‘time super-slow’ that would slow everyone on the map down to like 5% of normal speed, I think we called it Tempo Shift or something. It would also slow down projectiles that were mid-flight, but Lúcio himself would be unaffected by any of this and could skate around like normal, with the caveat that he couldn’t hurt any enemies. The idea was to use your boops to knock enemies around (and projectiles, at the time his boop could reflect projectiles), and heal your team, to set up a more advantages situation for your team once the slow ended. It was crazy and had some fun moments, but ultimately people just hated being slowed like that, even if it was an ultimate.

The next thing we tried was to allow him to fill other players ult meter some amount by using his own. We tried a couple different versions of this but it was both crazy powerful but also no one ever ‘felt’ it happening, which is basically the opposite of what you want when you’re designing something. Eventually we tried Sound Barrier as a way he could save his team from big burst damage.


A huge McCree tip of the hat to Geoff, Josh, and everyone who collaborated on the experimental card patch notes. If you haven’t already, you can read them here. However, as of the publish date of this article, they’re no longer playable in-game. 

59 Comments