Usage Rate New Template.jpeg

What’s up guys and gals, CaptainPlanet here to present the Overwatch Hero Tier List and Meta Report: A Tale of Two Comps. The bulk of this week’s data came from the Overwatch Monthly Melee, and thus this “Pre-Orisa Patch” pro meta was heavily influenced by two teams. Of course I’m talking about Selfless and Rogue, who stuck to their guns throughout the entire melee, rarely even swapping their flex roles while dominating their competition. These two dive comp juggernauts met twice: once in the winners bracket semis and again after Selfless ran through the loser’s bracket to the grand finals, so both of them represent a large chunk of this weeks’ dataset. Later on in the report I will talk more about what made these the two hottest teams in NA Overwatch so successful, but first let’s take a look at the (slightly) new and improved tiers:

new tier template.jpeg


S Tier (>=95% Usage Rate): Lucio (96%)

Tier 1 (>80% Usage Rate): No one!

Tier 2 (>50% Usage Rate): Ana (76%), Tracer (68%), Reinhardt (56%), Soldier 76 (55%)

Tier 3 (>20% Usage Rate): Roadhog (45%), Genji (45%), Winston (44%), Zarya (30%), D.Va (23%)

Tier 4 (>5% Usage Rate): Pharah (16%), Zenyatta (13%), Mercy (10%), McCree (9%)

Tier 5 (<5% Usage Rate): Widowmaker (4.5%), Torbjorn (3%), Reaper (2%), Mei (2%), Sombra (1%), Symmetra (1%), Junkrat (0%), Bastion (0%), Hanzo (0%)

Tiers/Usage Discussion

Last report, I asked for feedback on how to better present my Tiers and while I received many amazing recommendations many of it boiled down to me not really generating enough information each week to analyze the lower usage heroes in terms of strength. I’m going to continue displaying the tiers in terms of usage time – however I did take a common suggestion to heart, and removed the “grading” scale in lieu of a numbering scale. I did retain the “S Tier” as a legacy tier though, since heroes that are used for >95% of match time are kind of ridiculous…right?

Anyway, outside of “Mr. S Tier” Lucio, this week’s tiers looked pretty balanced! In a world where the majority of non-Rogue team compositions had at least 2 tanks and 2 healers, it was exciting to see many different tanks, dps, and healers exceed Tier 5 status. Selfless ran a 2/2/2 composition, Rogue ran a 3 DPS, 1 tank comp, and other teams like LG Evil and Immortals still ran triple tank comps – diversity was surely on display the past few weeks in pro Overwatch! Looking to the Tier 5 heroes themselves, some actually had their own few minutes of fame. Buds popped off several times on Widowmaker, one of the only players to use her this weekend. NiCOgdh’s swap to Reaper on Temple of Anubis was vital in Rogue’s second point capture over Selfless. On the flip side, Sinatraa’s Sombra hacked Akm’s Pharah mid-Barrage, preventing Rogue from capturing Anubis in a separate match. Much of Overwatch’s diverse cast has been on display in pro tourneys as of late – but this may be fleeting as balance changes and Orisa have just been released. The effects on the meta that changes to Winston’s Barrier, Sombra’s Teleport, Ana’s damage, and more are yet to be seen, and I will not pretend to know exactly what’s going to happen. Instead, let’s take a closer look at the meta that was on display at the Overwatch Monthly Melee, by breaking down Rogue and Selfless’ similar, but different takes on team composition.

A Tale of Two Team Comps


The Lineup

To this day, no one is sure if KnoxXx can play any other hero other than Winston following Rogue’s move to Vegas, where KnoxXx presumably underwent “Lucio Conversion Therapy” to transform him into a Winston main. Jokes aside, the current iteration of Rogue plays one team comp, and they play it extremely well. KnoxXx on Winston, SoOn on Tracer, uNKOE on Ana/Zenyatta, winz on Lucio, NiCOgdh on Genji, and aKm on McCree/Soldier/Pharah. At the top levels of Overwatch, there has always been a battle of competing philosophies: “do we play the heroes we think are the strongest in the meta, or do we play the heroes that we’re best at”? Both philosphies have worked out in the past: Misfits famously won Dreamhack Winter playing a Widowmaker dive lineup into Fnatic’s triple tank roster, but EnVyUs arguably mastered and dominated with tri/quad tank lineups in the height of that meta.

Luckily for Rogue, their hero interests coincided perfectly with what most pros believed was the strongest lineup philosophy in this final week before Orisa’s release: dive comps. There are lots of flavors of dive comps, but Rogue’s preferred heroes all fit within the genre. SoOn is arguably the greatest Tracer in Overwatch, challenged only by Koreans and his Selfless counterpart: Sinatraa. NiCOgdh brought the skilled Genji that Rogue has so obviously been missing since TviQ’s departure; he had several multi-kills in last weekend’s Overwatch Monthly Melee that he had no business actually pulling off. aKm and hitscan are a match made in heaven, and it helps that Soldier 76 is the strongest, most consistent DPS hero in the current patch. uNKOE can frag out like a DPS on Zenyatta or hit clutch Sleep Darts on Ana, heroes that he’s stood out on regardless of the meta. Post-reprogramming, KnoxXx seemed to have taken a page out of Miro’s book and was pushing the limits of how aggressive Winston can be played. And winz plays a serviceable Lucio after months of being relegated to Zarya play.

The Strategy

Selfless may have had more press recently, but both they and Rogue have nearly perfected two of the most aggressive playstyles in Overwatch. Courtesy of both of their team compositions having a Tracer, Rogue and Selfless have a high degree of flexibility in how they can create openings for initial pick-offs. Rogue prefers a more traditional approach because they have KnoxXx – they abide by the mantra of “just send the monkey in!” KnoxXx will jump in with SoOn and NiCO close behind, drawing attention and fire. Distraction created, SoOn can latch on to the nearest support and usually full-clip them down – or NiCO will finish them off with a quick dash. While the enemy is preoccupied with these three terrors, aKm can usually have his way with the enemy on his hitscan hero of choice, all the while uNKOE keeps KnoxXx alive on Ana or increases the team’s fragging power with Zenyatta.

What’s even more interesting is that Rogue executes this strategy on Defense as well. Since KnoxXx can’t play (or at least, hasn’t yet played) Reinhardt, the principle of “send the monkey in!” applies on both sides of the payload. When the enemy team collapses on the payload, Rogue will engage much in the same way they would on offense: a Winston-lead pile-driver of high mobility heroes with hitscan and support backup. Rogue also highly values contesting the cart at all costs on payload maps, and will rush back from spawn to contest using their heroes’ movement abilities – almost all of their hero pool has them after all. They even do this when they’re at a player disadvantage: sending players in one by one to try to at least take someone out with them. Why is this? Certainly “trickling in” on defense can’t be a good idea right?



Rogue has stumbled upon a defensive concept in Overwatch which works for certain points of certain maps. It’s a counter-intuitive strategy that actually helped them win fights that on paper they had no business winning. But let’s back up first. Begin by imagining King’s Row first point. The attackers have a tiny distance from their spawn to the point, and the defenders have an extremely lengthy run back from spawn. Therefore, trading kills on offense is advantageous, since the spawn distances are unequal. Many players might stop there, thinking that this is true for the entire offense, but it’s not. Some points – often the second point of payload maps – have a spawn distance tipping point where the defense spawn becomes closer to the point or payload than the offense. So the idea gets flipped: it’s actually better for defense to trade kills the closer the payload gets to a checkpoint. Rogue has realized this, and will even take fights when the payload is <1m from the checkpoint, because for every 1:1 trade they make and longer they continue the fight, the more likely they are to win the entire engagement and burn off the offense’s time bank.

It’s smart for them to take these close-to-the-point fights for other reasons as well. On the second point of many maps (Watchpoint, Eichenwalde, Route 66, Dorado to name a few), there’s a delay in the cart moving forward, because a door has to open. This “dead time” doesn’t require the defenders to contest the point, and they can use it to re-group and set up. Also, by contesting and dying right before the cart reaches its checkpoint, they can ensure that they are not split-spawned by the enemy team as the spawns update to the final checkpoint. Combining this stalling, spawn-abusing defensive strategy with their offensive uber-dive, Rogue has created a beast that not even Selfless could tame.


The Lineup

In a lot of ways, Selfless resembles Rogue in their team composition and reasoning behind it. Like Rogue, Selfless has subscribed to the theory that developing a unique team comp, built to their own players’ strengths, is the proper way path to success in Overwatch. Dafran and Sinatraa are almost interchangeable in their hero pools to aKm and SoOn respectively, except that Sinatraa was seen playing Genji, Sombra, and Zarya occasionally instead of sticking to Tracer. Like KnoxXx and Winston, I’m not sure whether or not Kresnik can play any other hero than Reinhardt. In fact, according to the data this week, he did not swap outside of a few panic Mei swaps on Temple of Anubis. Keeping to that trend, Michael3D and DhaK essentially only played Ana and Lucio respectively as well. Finally, while I know Emongg has a deeper hero pool than just Roadhog, his Hooks were as vital to Selfless’ success as NiCOgdh’s Genji antics – he’d often pop off hook after hook, kill after kill in the midst of prolonged melees, slowly turning the tides to Selfless’s favor.

Just like Rogue, Selfless is definitely benefiting from having transcendent Soldier 76 and Tracer players on their roster in Dafran and Sinatraa. However, the similarities end there. dhaK has been one of the premier Lucio players in Overwatch since early beta – he was DSPStankying around maps before DSPStanky ever left the console realm. Hell, ask and pro player what “dhaKing” means and they’ll know – he was one of the first to popularize the high-altitude, point-stalling antics that Lucio players employ today. While dhaK can probably run circles around winz, uNKOE is a probably a better Ana than Michael3D, and definitely a better Zenyatta. Selfless doesn’t have a Genji main like NiCO, but they do have Emongg who fills out the “third fragger” role quite well. Despite Kresnik playing Reinhardt so passively, the way they play their lineup resembles a full dive comp: they leave Kresnik behind to push the payload and push into the enemy with their entire team.

The Strategy

Rogue has perfected the traditional dive comp, but Selfless has a whole different dive in mind – spawn camping. Spawn camping, when it works, makes a lot of sense because forcing the enemy team to fight off of the point allows Selfless to push the cart uncontested, or prevent the enemy team from pushing the cart themselves. Selfless has taken it to the extreme and actually chains wins at these spawn-fights, which drastically decreases the enemy teams’ ability to mount a counter-assault on the payload. The better Selfless forces these off-the-point fights, the more pressure they place on their opponents: sometimes forcing them to use ultimates or swap heroes just to escape spawn. At that point, even if Selfless loses the fight, they’ve won the war, since those ultimates are not being used to contest or push the payload. It’s a win-win-win situation, as long as Selfless can keep fighting.

To better understand the strategy behind Selfless’ rise to prominence, I reached out to LegitRC, Selfless’ coach, who told me that Selfless does not necessarily go into every fight hoping to push to spawn. In fact, Selfless generally enters fights without any respect to the other team’s setup – instead trying to get into positions where they can help each other when needed. He clarified by saying they “always focus on how to best assist each other”, but he was guarded about what he meant by this. From the matches I watched this weekend, it seems that this means every Selfless player will have a virtual “battle-buddy”: a player or players who they can rely on to either finish off a kill that they started, or save them from a flank they weren’t expecting. This battle buddy could change at any second based on positioning or timing, but every Selfless player is ready and aware to provide support for their teammates at a moment’s notice.

To give an example of how this system is used, LegitRC described uNKOE as a player that was nearly impossible to get to because he’s so good at dueling flankers on Ana and Zenyatta. Selfless’ battle buddy system, however, allows them to force transient 2v1s at any given time. It works by giving the players the freedom to play within the flow of the game, and react as a unit to each situation. Sinatraa pokes and prods from behind with Tracer, and Dafran and Emongg provide front-side pressure with Soldier and the threat of Roadhog hooks. The openings are small, but as soon as they appear Selfless looks to turn any advantage into a 2v1 or 3v1 situation. Sinatraa may not be able to out-duel uNKOE on his own, but he can usually get him low enough for Dafran’s Soldier 76 to finish him off from long range. Similarly, there was a play on Hollywood where Michael3D was unable to save himself from SoOn’s Tracer, but he was able to survive just long enough for Dafran to come to the rescue by nailing SoOn with a direct Helix.

But how does this all lead to spawn camping? Selfless’s fights are a shifting sea of attempted 2-3v1’s, and when a kill actually occurs then the whole thing snowballs. A pick leads to 6v5, which makes the 2v1 mini-fights more likely, which leads to more picks, which just makes things worse for the enemy. It’s one of the most dynamic attacks in Overwatch, powered by their individual skill and what LegitRC described as strong comms lead by dhaK as shotcaller. As the chain reaction of kills grows, so does Selfless’ aggression. LegitRC said they don’t care for what the other team is doing, they just want to put themselves into the best position to get kills. What better position than chasing after a retreating team? It just so happens – as described above – that spawn camping has a lot of secondary benefits but ultimately Selfless is only worried about one thing: taking, and winning as many fights as possible. Lucky for them, spawn camping is oftentimes the answer to doing so.

Final Thoughts and Shoutouts

Shoutout as always to the tournament organizers for putting on the tournaments from which these reports pull there data. This week I had data from the PIT qualifiers, the Carbon Entertainment Group Stages, and the Overwatch Monthly Melee which all required admins, casters, Twitch mods, players, managers, and many many more to bring entertainment to your Overwatch-obsessed brains. Mine included! So if you know anyone who works in the scene, are fans of any tournament support staff, or are even vaguely aware of the grassroots side of Overwatch, be sure to give that person a bit of thanks – they’ll greatly appreciate it!

Also, as a companion to this week’s Meta Report, I’ll be having Selfless’ shotcaller, dhaK, on the Around the Watch Podcast which should go live Thursday evening, time permitting. If you have any questions you’d like me to ask dhaK, be sure to hit me up on Twitter, or in the comments of the reddit threads that accompany the report.


Until next time,