The Problem With Overwatch's 'Dive' Meta 

Image credit: Blizzard.

It's a new Overwatch match, a fresh start. You've got a good feeling about this one. But then you get within spitting distance of the point, and suddenly Winston crashes down on you, Tracer sucks up your health pool like her weapon's a giant milkshake straw, and Genji chops up what's left — all in the blink of an eye. It sucks! Welcome to the dive meta.

The highest echelons of Overwatch play have, for the past few months, been ruled by "dive" compositions. Basically, these teams select hyper-mobile characters like Winston, Tracer, and Genji (or, increasingly, Soldier 76) to focus offence on individual heroes and tear them to confetti in seconds. Often, another person playing as Zenyatta will put an Orb of Discord on the target, boosting the damage they take even more. Tanks like D.Va and Zarya and healers like Lucio round out the composition, shielding and healing the attackers, who are squishy on their own.

If a team communicates well, it's an incredibly effective strategy. Players who aren't anticipating a dive get absolutely melted, their ashes used to garnish Winston's peanut butter victory sandwich. You can understand, then, why some players find it so frustrating: losing to a dive comp isn't interesting or exciting. You're done before you even get a chance to put up a fight. Many players — whether they play in higher tiers of competitive mode or watch Overwatch esports — are sick of it.

So, Blizzard should do something about dive comps, right? Tweak some heroes, twist some balance nobs, delete Genji entirely? The issue is, however, far more complicated than it seems on its face. Let's break this down.

Diving into the deep end

The downsides of dive comps are a symptom of a sickness, not the sickness itself. In fact, taken on their own merits, they can be pretty fun. Where once tanks dominated Overwatch's meta, trudging forward and keeping fights relatively slow-paced, a dive comp battle is full of sprinting, leaping, and close shaves. I've seen some players say they prefer it to previous metas and hope it sticks around, and personally, I enjoy watching pro players duke it out with dive comps more than I did the other dominant metas.

Where dive comps start to become a problem, however, is in effective countermeasures. While a poorly organised dive comp will generally capsize against the rocks of a decent defence, a coordinated one can only really be countered by another dive comp. Previously dominant defensively-oriented comps like "triple tank" (where three tanks form the backbone, usually with Ana somewhere in the mix) lack the mobility to avoid a smart dive.

As a result, the current Overwatch meta — especially in esports — isn't really "dive." It's Dive vs Dive. Clash of the Dives. War For the Planet of the Dives.

Fans are clamoring for variety. Dive, many say, would be totally fine and even welcome if some kind of rock-paper-scissors dynamic enclosed it and at least a couple other equally viable team comps. So far, however, Overwatch's highest levels have been defined by singularly dominant compositions, rather than colourful chess matches where tactics shift to fit the needs of the moment.

A year ago, people were calling the 2/2/2 meta (two tanks, two healers, two DPS) "stale and boring to watch." Later in the year, fans called triple tank "the cancer meta." So dive has its own problems, but people also dislike it because, like previous dominant metas, it's overstayed its welcome.

Unfortunately, Overwatch's dominant metas have a way of snowballing. Top-level competitive players watch pros and decide which heroes are "on-meta." If other competitive players try to play off-meta heroes, they get chewed out. But the only way the meta can evolve (outside of a sudden, seismic balance change from Blizzard) is through experimentation.

Some players, however, refuse to go out on a limb because they don't want to risk losing precious rank points, and others either feel too socially pressured to switch heroes or realise they can't make anything meaningful happen without support from their team. So the meta stagnates. People will figure out a game-changing counter to dive comps at some point, but until then, high-level Overwatch remains a place where the hero pool suddenly becomes very shallow, despite how interesting and varied it could be.

Storm's coming

There's obviously a pattern here: a new comp arises and becomes dominant at the expense of other comps' viability. Top-level Overwatch is briefly exciting, but then it calcifies into this scaly scab of a thing for a few months.

This raises a question: what role does Blizzard play in all of this? Did it design a game where the strategic possibility space is too narrow for more than one kingpin comp to thrive at a time? And is it failing to address that problem now, despite ample awareness that it exists? A vocal contingent of players seem to think the answer to those questions is, "Yes, duh. Why hasn't Blizzard pressed the magic button that fixes it yet?"

In an attempt to assuage those concerns, Overwatch director Jeff Kaplan recently published a lengthy post on the game's forums. The crux of his argument was that the top-level meta only represents a particular group of players — "the top 3rd of all players," he said, and even then, they play catch up with pros for "weeks if not months."

Kaplan doesn't feel like it's Blizzard's place to meddle with the meta. "I do not agree with the philosophy that we should just make balance changes solely to shift people off the meta," he wrote, noting that he feels like, right now, the game is balanced. "The game team should be constantly evaluating balance and making changes that are actually needed because a hero is unbalanced. But making changes to a hero because their pick rate is too high or too low is not my idea of responsible game balance."

He added that he prefers to see a game's meta shift either because the game is unbalanced and Blizzard makes a change to balance it, or because players sussed out a new strategy. He called the latter "the best-case scenario," but admitted that it can take a lot of time.

Kaplan also isolated an idea some players have proposed: a MOBA-inspired pick-and-ban system that allows teams to forbid their opponents from selecting certain heroes. This would add another strategic layer to the game and require teams to vary up their hero selections. Seems like a great solution, but Overwatch isn't a MOBA — at the moment, it also has significantly fewer heroes than LoL or DOTA 2 — and Kaplan doesn't think a pick-and-ban system fits the game his team made.

"I prefer to think that Overwatch allows you to be creative, which is different than forces you to be creative," he explained. "I don't want to watch the best Genji player in the world play Zarya — I want to see him/her play Genji. And also, seeing how many of you 'main' heroes because you love them, I don't want the game — or your opponent — telling you you're not allowed to play that hero."

Kaplan closed out his post by saying that he understands the desire for more variety at Overwatch's highest levels of competition, and he hopes multiple viable comps will blossom over time. He added that there will probably be a new meta in a few months, and it will come with its own complications. "If you're the type of person who feels like the meta should shift every 2 weeks, then you'll probably be sick of that meta and wishing it was back in the good ol' dive comp days," said Kaplan. "I just caution against wanting change for the sake of change."

Soon Enough

Already, there are signs that Overwatch's meta is about to evolve. Sombra, Pharah, and Mercy have started making more appearances in recent pro games, and a change to Reaper's "life steal" ability has made him significantly more survivable. The latter development is especially big, because Reaper is a nightmare for Winston, who often functions as the vanguard of dive comps, leaping in and sowing chaos among enemy ranks. If Winston's getting countered at every turn, what happens next?

Then, of course, there's Doomfist, who's on Overwatch's public test realm right now and will probably punch his way into the live game in a week or two. At first, people had him pegged as another asset to the dive meta, but his combo attacks double as mobility skills, meaning that he can't actually deal much damage if he wants to leap away from enemies and survive encounters.

In Doomfist's first pro appearance, a special PTR exhibition match between Meta Athena and new Korean team Ardeont, Meta Athena tried out a handful of non-dive compositions that attempted to account for his vulnerability and offensive potential. They had some early success against Ardeont's dive comp, though they did not win the match.

While dive comps could evolve to account for Reaper, Doomfist, and other wildcards, it seems likely that a new meta will arise sooner rather than later. The question, then, is whether or not that new meta will allow for multiple, varied comps to rule the roost. History says the answer is "probably not." That's a problem on multiple levels, especially with Blizzard's much-ballyhooed, now officially Bob Kraft-endorsed Overwatch League on the way. A sport with little strategic variety and months-long bouts of stagnation is a hard sell to the mainstream audiences Blizzard and its extremely wealthy partners are trying to court. Don't get me wrong: matches can be fun to watch, but it doesn't take long for deja vu to set in.

Perhaps, though, that's just the nature of the game Blizzard has designed. In his post, Kaplan pointed out that there are many games with metas that are set or evolve slowly, pointing to everything from Team Fortress 2 (back in the day) to baseball as examples. "This doesn't mean the game isn't balanced or fun or fun to watch," he said. But Overwatch is not Team Fortress 2 or baseball, and it's following a very different path from both of those games.

Blizzard's simultaneously pushing it to be one of the world's biggest esports and a game that's fun for the majority of its audience, who are not even remotely close to being esports pros. It's a complicated balancing act. As for where Overwatch will end up, that's anybody's guess.


    This largely due to slow movement speeds coupled with character abilities that create a massive difference in combat capability, resulting in every fight being pure fight or flight.

    There is no middle ground in Overwatch, which is why most fights are decided very quickly.

    I have the answer to the Dive meta: boost Torbjorn.

    Let him have a dual ultimate. The first stays as is. The second lets him put down a second turret.

    In addition, give his turrets shields.

    That'll show the squishy divers.

      Lol Dva is a big part of dive these days and utterly ruins Torb turrets. Even winnie can easily manage a non ult'd turret using his bubble well.

      Personally i like dive and hope it sticks around!

        Yeah I know, that's why I think Torb's turret needs shields. It's not durable enough. The few extra seconds that would take D.Va or Winston to burn through shields as well as the turret could make a big difference, especially if there were a second turret (or Torb shooting them with his rivet gun).

          Yeah cause the Torb/Bastion, Mercy, Rein combo isn't cheese enough as is, now you want basically Alpha Bastion (Look it up) for Torb? It would break the game completely.

            Yeah, I know what alpha Bastion was like.

            However, my suggestion would create more of a Rock-Paper-Scissors set-up, where if you are facing a Torb/Mercy/Rein etc combo then you switch out to get some long-range power like Widow or Hanzo. Then the other team would need to switch to Dive comp to counter this, etc.

            I think there needs to be (as was originally intended) reason to switch heroes during matches. The only way to do this is to make certain heroes MUCH stronger against certain set-ups and other heroes MUCH stronger against those heroes. At the moment, it's a twitch game more than a tactical game. There are plenty of twitch games out there (CoD, CS etc) so Overwatch could really differentiate itself by making dynamic team comp a more crucial factor in winning matches.

      There actually was a bug where you could have two Torb turrets up at once and you could upgrade both.

      Not sure if that's still a thing. Certainly not going to try replicating it as I don't want to risk getting banned for intentionally exploiting a bug.

    "If a team communicates well"

    Therein lies the problem (for the non-professional players at least).

    So many people are anxious to always go this team setup in comp but the problem is you need to be a good team to pull it off successfully and there'll always be that one teammate that isn't communicating with the rest or another teammate whose just not that good at the dive comp character they ended up picking.

    I always say leave the meta team setups to the professionals. It really only applies to them anyway since its impossible for a team of randoms to do as well as the pros do.

    Last edited 13/07/17 1:29 pm

      Meanwhile my housemate argues that people should use the pro meta picks even when solo queuing ranked because otherwise, "You're putting yourself at a disadvantage."

      Naturally I called bullshit on it, simply because the pro setups only work that well with their level of skill and communication. Which, as you said, is something that simply doesn't exist in the ranked environment for average players.

        Unless you're playing at a very high level you're probably best off just getting as good as you possibly can at 2-3 characters (I'd say only one but somebody else might pick first, so you need a couple, ideally in different categories) and playing the best strategy you can as a team with everyone using a character they're actually good with.

        The number of games I've lost because players are harassed into switching to characters they aren't comfortable with because 'we need x to counter their y and z that you're playing now is countered by w anyway'... People will do this even when your team is currently winning for some reason that completely baffles me.

        I like the game but I really, really hate the community.

        Last edited 13/07/17 3:12 pm

          "The number of games I've lost because players are harassed into switching to characters they aren't comfortable with"

          In my experience, that situation pales in significance compared to the number of games I've lost because people refused to switch off a character just because they were more comfortable with them. If Pharah is eviscerating the team, and you refuse to get off Junkrat, you can be damn sure I'll get salty about it. I don't know why people have the right to ruin the fun of other players because they're unwilling to broaden their skillbase a little.

            Why does the Junkrat, specifically have to change? He's not going to do much to shut down the Pharah, but there are five other characters on the enemy team that he probably can kill, and five other players on your team who can switch to counter the Pharah if need be.

            Maybe they can also play Mercy and Zarya pretty well but don't know any hitscans and will spend the rest of the match doing nothing but missing if you order them to play one. You can say they're 'broadening their skillbase' if you want but you're still going to lose because they're ineffectual as that character and they're going to have an even worse time as a result of being bullied into playing a character they don't want to - now you're the one ruining their fun.

            There's a big difference between saying "Can anyone play a hitscan to counter their pharah?" and "You there, person that I have (largely arbitrarily) decided is the problem, switch to this specific character that you're terrible at, for all I know."

            That's a terrible strategy, trying to force a player to switch to something they have no skill or expertise with. Junkrat may suck against Pharah but the player may be effective against the other five enemies; if you force them to change to a class they're not good at, they're going to suck against all six enemies and your team is going to be much weaker as a result.

            And worse, if they do switch to a counter hero for the problem you're encountering and suck at it, that hero is now in use so nobody else (who might actually be competent at the class) can switch to it to do a better job until they switch back.

          I try to be (at minimum) mediocre with all characters. I will then pick a character that fulfills two requirements (a) it's one that I want to have fun with; and (b) it will complement the existing team picks, if not ideally, fairly well.

          Yeah I get the odd comment like being a shit healer or what have you, but it's par for the course. At least I can play a healer when needed and not suck worse than the majority of people :-)

          I've always said you should be good with at least one hero from each class to maximise your effectiveness in competitive.

          Saying you're a healer main or a tank main is all well and good, but sometimes a teammate will end up stuck playing a role they're just not good at.

          Some of the best wins I've had in comp is when the entire team changes things up with people going different roles.

        Picking meta in solo queue is just insane, pick something the suits your comp or hurts the enemy's. Nothing worse then watching someone pick Lucio because he is Meta and watch him try to provide any value to the widowmaker and torb on his team

    I'd really like to see Blizzard making changes to make team compositions more varied and creative. I don't care about seeing them balance against a specific meta, but I think the game needs a fairly far-reaching shakeup balance-wise. Not with the aim of preventing metas from developing, but with the aim of ending up in a scenario where there are maybe 3 popular and successful team compositions in the meta at any given time instead of only one.

    Frankly the fact that this is not the case makes me question the assertion that the game is well balanced currently and ever has been. There are examples of good games with basically totally stagnant metas, it's true, but those games all EITHER allow for a much wider variety of strategies and playstyles under the umbrella of their meta than Overwatch has or does OR have a much higher skill ceiling on the core mechanics, making the game feel deeper in that regard.

    This is not necessarily an easy thing to fix, but I think that if they can't find a way to adjust the game so that more strategies are viable at any given time it's going to seriously limit how long the game stays as popular as it is now.

    In the meantime it'd be good if they could patch the mid-tier competitive community to not turn into psychotic rage monsters every time somebody picks a hero they don't think is as good as some other hero.

    The rest of Kaplan's post should probably be posted in the article, because he does highlight that the vast majority of games (competitive included) aren't dive meta based.

    The meta problem I feel is actually a map problem - maps are a constant flow of open spaces that benefit attackers and chokepoints that benefit defenders. The dive comp is a way to force the argument as quickly as possible on chokepoints.

    This isn't help by high mobility characters combined with Overwatch's continuing fucking awful hitreg reducing the ability of defenders to kill attackers, hence lending towards dive v dive.

    I..... don't know if Overwatch will ever be even a remotely balanced game tbh.

    I've never found whatever is in popular meta to be a problem in comp. What is a problem is Blizzard's annoying habit of making major gameplay changes that royally screw up everything just before a new season begins and then spends all season "fixing it" (aka Kaplan saying "It's working as intended" till someone else on staff has the balls to tell him it's fucked, then they change it), thus making comp into more of a PTR situation than an actual competition.

    Competitive mystery heroes?

Join the discussion!

Trending Stories Right Now