Overwatch 2 director confirms what we already knew: the ‘meta’ isn’t all that important
Overwatch 2 game director Aaron Keller has been doing something pretty neat lately: talking about Overwatch 2. The de facto voice of Blizzard on matters of Overwatch recently made it a goal to communicate with players more often about the current state of the game, and the man is on a roll. So far with his “Director’s Take” series, Keller has written about ranked woes (opens in new tab), the future of Arcade mode, and the failed experiment that is map pools (opens in new tab).
It’s refreshing to see so much transparency from Overwatch’s top brass on topics that matter most to daily players, and today’s blog concerning win rates and hero popularity (opens in new tab) is easily his most illuminating yet.
Right out of the gate, Keller shares some interesting observations about Overwatch 2’s ranked meta in the first weeks of season 3. For instance, “most supports are viable in nearly every skill tier,” something that didn’t feel very true back in season 1. He also confirms one development that players have noticed looking at semi-reliable stat aggregation sites like Overbuff (opens in new tab): Brigitte is having a moment right now.
“Brigitte has really popped this season and has the highest win rate for nearly all skill tiers except for Top 500, where Zen takes the lead, with both averaging out to a nearly 55% win rate,” Keller said. “On the other end of the spectrum are Kiriko and Moira at around 45%.”
While you’d think that those who are playing a competitive shooter, particularly its most competitive mode, would gravitate toward heroes with the highest win rates, Keller says that’s not the case (names bolded by me):
“When we look at who is actually being played, the top supports are Ana, Kiriko, and Mercy for nearly all skill tiers until Silver and Bronze, where Moira becomes picked quite a bit.
“This goes to show, that heroes with the highest win rates aren’t always the heroes that are played the most.”
This goes to show, that heroes with the highest win rates aren’t always the heroes that are played the most.
Aaron Keller
I find that fascinating. We know that in a competitive environment like ranked Overwatch, it can seem like sticking to the consensus-adopted meta of the month is all that matters. We also know that anecdotally it can feel like teammates are choosing heroes based on their preference, not necessarily what’s best for the team composition, so it’s interesting to see the official numbers back that up. Keller notes it’s “counterintuitive” that Kiriko is so popular despite winning less often and, on the flip side, Briggite wins constantly with a much lower presence.
There are several possible takeaways from this. I tend to think this suggests that, even at high levels of play, Overwatch players don’t like to feel boxed in by a small pool of “best” hero picks. People like to play their favorite heroes, and in fact, they’re unlikely to swap (opens in new tab) to a hero they don’t like.
As Keller mentions, the win/pick rate disparities could relate to other factors that aren’t being directly compared here, like overall team composition or map differences. Either way, Keller says the team “tends to think that highly skilled players are continuing to pick certain heroes for good reasons,” and is gathering additional metrics to study this. “If it’s fruitful, expect to hear more about it.”
In any case, I think the numbers indicate that there’s a significant chunk of the Overwatch 2 community (a probable majority, even) for whom the meta isn’t important. And if most players don’t care about the meta, how much should Blizzard care?
Maybe it comes down to keeping things balanced enough that the “best” heroes are only the best by a few percentage points. We certainly heard about it when supports weren’t getting enough love (opens in new tab) and Sojourn was one-shotting everyone into oblivion. It’s safe to say that players care about every hero being viable so that their favorites never have to fall out of fashion, and based on this data, that may be enough.
