Balancing Hero Shooters Is A Science That Requires Live Test Subjects

It won’t surprise you to hear that game development is hard. And in the ever growing games-as-a-service market, that development sometimes doesn’t have an easily defined endpoint. You can see this in hero shooters–the larger ones continuously get new characters every season, forcing the developers to reevaluate the in-game meta and rebalance existing characters over and over again.

But what about the new characters? Prior to releasing new playable characters, do developers try to balance them based on what the team perceives that character’s role in the meta will be? Or is there some other consideration that governs their approach? Once again, it won’t surprise you to learn that different studios have different strategies for how they attempt to balance their new hero characters prior to players getting their hands on them. The one common thing among them, however, is that they all agree it’s a big challenge.

Respawn, for example, aims to introduce new playable legends in Apex Legends as slightly too strong. It’s a trend you can see across the battle royale game’s Year 3 legends. Though Fuse did release in too weak a state, the same can’t be said for Valkyrie, Seer, and Ash. Seer especially launched in a devastatingly overpowered state, and Respawn had to quickly nerf him within weeks of the start of Season 10: Emergence. Neither Valkyrie nor Ash launched in quite as unbalanced a state, but both characters did (and continue to) draw ire from the community for their powerful passive abilities.

“You never really know how strong a character is before you release them to the public,” Apex Legends designer Devin Weise told me over email. “We do our best to gather internal feedback, in addition to sessions where we bring in external players (like pros and content creators). We use all of the feedback to iterate on the character and gain an understanding of what tuning knobs we have and what effect is when we adjust those knobs. That gives an idea of how strong they are but there really is no substitute for getting the character in the hands of millions of players in a live environment.”

He continued: “With that said, we aim towards the stronger side of things for release. After the initial excitement of a character launch, when they are picked on almost every squad, a lot of players tend to go back to their previous favorites (or check out newly-buffed characters). If the new character is too weak, we end up with less data and less experience to guide our balancing decisions. I think the key is making sure the character is good at specific things and not everything. Too much versatility makes older characters feel obsolete and muddies our understanding of the character’s identity within the roster.”

Ubisoft Montreal goes for a slightly different approach when it comes to Rainbow Six Siege, however. Like Respawn, getting feedback from a small pool of professional players and content creators ahead of a public release is still part of the developer’s strategy. But instead of aiming for new operators to be slightly too strong at first to gather more data from the public, Ubisoft Montreal tries to get new characters to feel as balanced as possible before the public gets their hands on them, and even then, the character spends weeks on a public test server first to ensure no additional tweaks need to be made.

“Our context is a bit different due to our asymmetric gameplay,” Rainbow Six Siege associate game director Aurelie Debant said to me via email. Siege does represent a different kind of hero shooter. While most games in the genre–like Apex Legends, Valorant, Overwatch, and Battlefield 2042–have their characters available to pick regardless of what team you’re playing on, Siege divides its characters into two separate camps: attackers and defenders. With that constraint in mind, Ubisoft Montreal needs to keep in mind how a new Siege operator will approach each map based on their role. A defender will always have allies who can help them be even more entrenched and protected for example, while an attacker will always have allies that can help them push forward even more aggressively.

“We pay a lot of attention to player frustrations: frustration when playing as the operator, and frustration when playing against the operator,” Debant said. “Fairness is a major component: in terms of the player, I must feel that I have the tools to act on the situations I encounter and when I fail, I need to identify what I could have done better. We must ensure all operators are fair, that none are underpowered and thus pointless to use and none are overpowered and thus dominate everything.”

She added: “This is the core to our philosophy. From the conception phase, we must ensure that we have enough balancing levers for each operator developed, including trade-offs, soft counters, and combos: Which operators could they combine with, and how? Which operator could they counter, and how? Which operators could they be countered by, and how? And we like to favor soft counters instead of hard counters, because soft counters create more gameplay opportunities and imply more skill than a hard counter.”

Riot Games doesn’t even consider balance when first making a new character for Valorant. That aspect of making a new character is important, but worrying about balance comes after the team has clarified how the new agent is supposed to change the core loop of the game. This can create the need to spend a bit more time balancing a new agent in the weeks leading up to their release, as seen with Chamber–his release was delayed, to the point that Valorant Episode 3, Act 3 started without him.

“When we first start designing new agents, we don’t even think about balance,” Valorant co-lead designer Max Grossman told me over email. “We spend months focusing on developing a unique core loop for the agent, usually centered around a single character defining ability (Jett’s Dash, Killjoy’s Turret, Reyna’s Dismiss). The entire reason we even release new agents is to make players think about the game differently, whether they’re playing as the agent, or against them.”

He continued: “We want to make sure that each agent delivers on a unique fantasy that adds new tactical challenges to the game. Once we do that, we have to ensure that every ability has reasonable counterplay to ensure the game stays healthy. After those elements are solidified, we focus on the nitty gritty details, tuning and tweaking abilities down to the millisecond.”

It’s in those details, as well as the soft counters and hard counters for a character, where the most important aspect of balancing a character comes into play. There are two major considerations to a character’s place in the in-game meta: a character’s implied power and their actual power. A character can actually be well-balanced–even overpowered–but if their abilities don’t feel strong, then players will complain that they feel underpowered.

“This is a challenge we face almost daily–the way an agent feels to play against is definitely not correlated to their actual power,” Grossman said. “Two agents that deal with this phenomenon in opposite ways are Breach and Sage. Breach has one of the lower winrates in Valorant, but players perpetually feel like he is overpowered. Sage, on the other hand, has one of the highest, but the community generally feels she is balanced (and potentially even underpowered).”

Grossman added that he feels this is an issue with most hero shooters–a character will feel overpowered when they have abilities that don’t afford opponents many opportunities to react to and counter. “This feeling is amplified when you are on the receiving end of very dangerous abilities with devastating effects, like Breach’s Flashpoint,” Grossman said. “Flashpoint has a really fast windup that is difficult to react to, and once you get hit, you are completely blind. On the other hand, Sage’s Slow Orb takes time to travel and expand after hitting the ground, giving players plenty of time to get out of the zone. Even if you’re standing in the zone, you can still fight back and use your abilities effectively.”

He continued: “Now here is where it gets tricky; for the sake of discussing an ability’s ‘strength,’ whenever you use an ability, the more it increases your chance to win the round, the stronger it is. With this definition (and our internal data), Sage’s Slow Orb ends up being a ‘stronger’ ability than Breach’s Flashpoint. Breach’s Flashpoint is still a powerful ability and it is actually more likely to lead to a kill than Sage’s Slow Orb, but it is much more difficult to actually capitalize on the effects.”

It’s a balancing act that all the studios I spoke to said will likely never be 100% right. Playtesting and player feedback can point the developers in the right direction, but it can then take months of time to adjust a character to a place where it feels right. And even once that’s achieved, a different character down the line can mess up that in-game meta and force more reworks.

Apex Legends has struggled in that space for a while with one of its oldest characters: Wattson, who has been in the battle royale game since Season 2: Battle Charge. She launched in a fairly strong state, quickly rising to be a regular pick for professional players alongside Pathfinder and Wraith. Despite this, casual players soon started complaining that Wattson felt underpowered. Which, from a certain standpoint, is true. As the game’s only true defender, Wattson does not possess abilities geared towards dealing damage. She’s a deterrent, a character who can construct a bunker that’s difficult to breach. So despite being very strong, Wattson didn’t actually feel all that strong to play in comparison to other legends who have more easily identifiable contributions to a team’s success like healing, teleportation, poisonous gas, or an airstrike.

“I think in an ideal world, a player would feel like their character is overpowered while the opponents feel like the character is fair,” Weise said. “But power-level is only a part of the discussion when designing and balancing characters. Abilities should be satisfying to use, the character should have an interesting play pattern, and they should bring something unique to the squad. Making sure characters are fun to use is a focus point before and after we release a character.”

There were numbers to back up Wattson’s place in the meta too as she has consistently had one of the highest win rates of any legend in the game. In June 2021, then-Respawn lead game designer Daniel Kleinout told players that her win rate was even outperforming Wraith. Wattson didn’t need a buff, but it felt like she did (it likely didn’t help that Wattson also suffered from substantial bugs in her ability kit for a few seasons, making her seem even more underpowered). So Respawn obliged the playerbase and decided to rebalance the character, implementing changes to Wattson at the start of Season 11: Escape. These adjustments didn’t make her much stronger, but it now feels significantly more satisfying to use her abilities and that in turn changes the perception of how powerful she is.

“While [Wattson] isn’t picked very often compared to most of the roster, she performs extremely well in all skill brackets (including in Arenas),” Weise said. “So, the past couple seasons of meta changes were more focused on other characters who were less effective in addition to being infrequently picked. That gave us time to work on changes for Wattson that would improve the experience of playing her instead of just bumping up numbers to make her more enticing.”

Apex Legends and Valorant are in the first weeks of their new seasons, and Rainbow Six Siege is weeks away from heading into its next season. Apex Legends Season 11 saw the release of Ash, a hit-and-run specialist with abilities geared towards hunting, and Valorant Episode 3, Act 3 added Chamber, a sentinel character who can summon a powerful sniper rifle. Rainbow Six Siege High Calibre will add Thorn, a defender with explosive traps and a unique SMG.

Remember: Enjoying games doesn’t preclude you from remaining informed about the culture of where the developers behind these games work. In 2018, Riot was accused of having a boy’s club “bro culture” environment, and then in January 2021, allegations were made against Riot CEO Nicolas Laurent that he had made unwanted sexual advances and harassed employees (an internal investigation of the company found nothing wrong with Laurent’s behavior). Ubisoft is similarly in murky water, having been accused of fostering a “frat house” workplace–allegations that still haven’t been adequately addressed by higher-ups at the company according to hundreds of past and present Ubisoft employees.