The Overwatch Ranked Tier List shows the most popular heroes during April, Season 15. Find out the most successful support, tank, and DPS heroes in this meta.
Based on popularity. The tier list is based exclusively on the popularity of a hero in ranked games. The win rate is considered only to determine the most outstanding heroes in a particular tier.
No pro data. The performance of a hero when used by pro players and teams doesn’t influence the tiers.
General and rank-based meta. The best picks at GrandMaster can differ a lot from the ones at Gold, so I produced both a general ranked tier list and many rank-based ones to point out the meta differences in each rank.
Only PC. This list includes no data from console.
Blizzard has not released an Overwatch API to access the data of all players in the game, so this article displays no official info. There are several websites that gather data from their users, and I decided to utilize the statistics provided by Overbuff as it is the most popular one.
Jeff Kaplan shared the official data on season 9, and the figures were mostly similar to the ones available on Overbuff. We can trust the data even it is isn’t completely accurate.
I have published updated data (based only on PC players) during each season. All the previous meta reports are listed below:
Past Meta Reports:
Most popular heroes by rank
The table is responsive. If you are using a phone, turn it sideways for a full view. Otherwise, zoom in or click the grey area to swipe the table.
Ranked hero tier list - April, season 15
Tier 1 (>10%): Reinhardt 10.3%.
Tier 2 (>7%): Ana 9.4%.
Tier 3 (>4%): Moira 6.2% - Mercy 5.3% - Zarya 5.3% - D.Va 4.7% - Lucio 4.7% - Zenyatta 4.4% - Genji 4.1%.
Tier 4 (>2%): McCree 3.5% - Roadhog 3.3% - Hanzo 3.3% - Winston 3.0% - Wrecking Ball 2.8% - Widowmaker 2.6% - Ashe 2.6% - Orisa 2.5% - Baptiste 2.5% - Soldier: 76 2.3% - Reaper 2.3% - Junkrat 2.0%.
Tier 5 (<2%): Brigitte 1.7% - Doomfist 1.6% - Tracer 1.6% - Pharah 1.5% - Torbjorn 1.3% - Mei 1.3% - Sombra 1.3% - Symmetra 0.9% - Bastion 0.5%.
Most successful heroes (positive win rate, and pick rate above 4%):
Symmetra is still included among the supports because of the statistical tool used. Removing her would have caused problems with the data.
The graph shows data on the top 8 most picked DPS heroes at all ranks.
Ranked hero tier lists by rank - April, season 15
GrandMaster hero meta
Tier 1 (>10%): Ana 10.0%.
Tier 2 (>7%): Lucio 7.6%.
Tier 3 (>4%): Zenyatta 5.6% - D.Va 5.5% - Reinhardt 5.2% - Widowmaker 5.2% - Winston 5.2% - Mercy 5.0% - Wrecking Ball 5.0% - Zarya 4.5% - Hanzo 4.4%.
Tier 4 (>2%): Orisa 3.8% - Roadhog 3.6% - Tracer 3.6% - Genji - 3.4% - Sombra 2.6% - McCree 2.5% - Soldier: 76 2.3% - Doomfist 2.3% - Baptiste 2.0% - Torbjorn 2.0%.
Tier 5 (<2%): Brigitte 1.9% - Ashe 1.9% - Mei 1.2% - Moira 1.0% - Symmetra 0.8% - Pharah 0.6% - Junkrat 0.5% - Bastion 0.4% - Reaper 0.4%.
Most successful heroes:
Wrecking Ball: 52.7%
In the past meta reports, I have explained how the Overwatch developers are gradually weakening GOATS patch after patch. Despite their efforts, it is still the most dominant meta in the game, particularly at very high ranks and in the OWL.
I have also pointed out the rise of Wrecking Ball and the impact of the changes brought by the patch released on March 19th. Indeed, his popularity increased, and now we can find him in the top 10 both at Master and GrandMaster.
This month we will review a hot topic: role queue.
Here is a summary of the sources and arguments showed in the article:
MonteCristo analyzed how the role lock improved League of Legends, and the potential impact it might have on Overwatch.
Jeff Kaplan clarified the queue time necessary for a DPS player, the pros and cons of the system, and affirmed they are working on several changes necessary prior its release.
I will provide detailed info on the role queue in Dota 2, its issues, and how the same problems could take place in Overwatch if not addressed preventively.
Finally, another argument strictly connected with the meta: hero bans.
2/2/2 role lock system
In the video below, Christopher “MonteCristo” Mykles (analyst for the Overwatch League) talks about pros and cons of a role lock system in Overwatch.
Role queue in League of Legends. After its release, it took a few years for the meta to develop on its own, but then Riot solidified it via items, runes, masteries, and new champions. From that time onwards, both casual and pro players had "an idea” on how to play the game:
As a casual gamer, you can queue for the role you like best. There aren’t anymore fights within the team to decide who is going to play a certain role.
Pro players can focus on a specific skill set and refine it over the years. Their careers are now longer and safer because the meta doesn’t shift wildly after each patch.
MonteCristo believes that 2/2/2 is the funnier meta to play and watch for the majority of the players.
Role lock prevents some overpowered compositions to be abused - GOATS would have never existed - and it also makes game balancing a lot easier for the developers because they can somehow predict how the meta will evolve.
A role lock system would also require massive balance changes because you couldn’t pick anymore a third healer to make up for the healing weakness of a support such as Brigitte. The developers will have to both balance the heroes and redistribute them among the tank, support, and DPS categories, which is particularly difficult for hybrid heroes.
Jeff Kaplan on the role queue
In the following interview by Unit Lost, Jeff Kaplan talked about the possible role queue system in Overwatch.
Jeff is a huge fan of 2/2/2 compositions, and he believes that it is the way the game should be played for most of the player base.
The Overwatch team loves the idea, but the players must understand that there are several downsides and costs necessary to enable role queue in the game:
It requires months and months of work to rewrite the matchmaking system.
Right now, the average queue times is one to two minutes, but in a role queue system the average queue times as a DPS player would go up 13 to 26 minutes!
They might need to assign each player a different rank depending on the role. If a main tank player suddenly decides to play as a support, the matchmaking can’t use the same rank or he will be matched with and against players that are stronger than him on that particular role.
The issues of the Dota 2 role queue
I was a great fan of the role queue system when it was initially introduced in Dota 2. If you don’t know it, basically you can select a role based on your position on the map, so each game will have one off laner, one mid laner, one safe laner, and two supports.
As an off laner, I thought the system was amazing because my queue time was not a lot longer, and I could finally enjoy games without several DPS heroes.
Now, a year later, the role queue is almost dead. There are certainly still players utilizing it, but if you search on the Dota 2 subreddit, you will find a massive amount of people complaining about it.
A short summary of its decline
Initially, more players used it, so the queue times were mostly acceptable, depending on your preference.
Safe and mid laners had longer queue times because the great majority of the players wants to play carries and DPS heroes.
A portion of these players was unsatisfied by the waiting time, so they started using again the standard matchmaking. Others decided to queue as the off laner, which in turn increased the queue times also for this position.
More and more people abandoned the role queue over time. For example, none of my friends use the role system in Dota 2 anymore.
DPS players playing as supports. Some heroes can be played in various ways, so many DPS players frustrated by the long queue times started to select the support role, which they played as semi-carries.
Of course, you can report the players who do not strictly respect the role lock, but it is easy to bypass the punishment. Don’t forget that all these report systems are automated - there will never be people watching entirely all the games reported daily - so if you do certain actions and/or buy supportive items during the game, the punishment system won’t detect you as a griefer, but just as a bad or inexperienced player.
This factor also contributed to the decline of the role queue because the other members of the team waited the long queue time just to find a situation similar to the standard matchmaking.
The same issue could happen easily in Overwatch as there are many support heroes who can deal a decent amount of DPS, such as Moira or a sniper Ana. You could wait 20+ minutes for a game and find Ana acting as Widowmaker for the entire match.
Actual queue times. The players at most ranks require 2-5 minutes to find a game via the standard matchmaking, but the same players need even 40-60 minutes if they try queuing as a carry or a mid laner in the current role queue. The situation is even more dramatic at very high ranks where you might have to wait over an hour.
I still hope that the role queue in Dota 2 will be improved someday, but its relative failure demonstrates that Jess Kaplan is telling us the truth: you can’t just put in place a role queue system without a massive redesign of the matchmaking and of several aspects of the game.
Note: the role queue was not introduced in Dota 2 because the pro or pub meta were stale, but just in an attempt to help pub players finding games where they could select the role they wished to play. Valve never made any direct change to improve the experience in role queue.
It is also important to mention that the Dota 2 role queue is behind a paywall: initially it was available via The International Battle Pass, while now you must be a Dota Plus subscriber to access it ($3.99/month).
It is still too early for hero bans
After reading the explanations above, I wonder how many people are still convinced that the role queue and the role lock are an easy solution to implement. Honestly, I am kinda neutral on it because I don’t know yet what to expect.
Heroes such as Brigitte and Roadhog would require a rework in order to work in such a system. For example, in a game with only two supports, Brigitte is not a viable healer for the team.
Another system that would improve the meta is the addition of hero bans. IMO, they should be added into the game independently on the implementation of the role queue.
In a stream that took place at Blizzard HQ a few months ago, Overwatch developer Geoff Goodman briefly talked about the chance to introduce hero bans in OW.
“As a spectator, you want to see a player using his best heroes, so hero bans can turn into a potential issue if they become player-targeted instead of being just composition or meta-targeted.
Hero bans bring many advantages, and the main argument is that the meta gets mixed up more, but this isn’t necessarily true for Overwatch because right now we have about 30 heroes. If you pick a meta composition and it has just one real counter, banning that specific counter will strengthen the meta even more.”
Goodman concludes telling they had lots of discussions about hero bans and there is definitely the potential to introduce them, but we will have to wait a while more because they need a deeper hero pool first.
We need more heroes
Right now, we lack hero diversity. For example, there are several tanks who can be played in a Dive composition, but there are only two shield-based ones, and a few choices available for other team formations.
I don’t know exactly how many heroes are needed, but probably we require at least 10-12 tanks to not limit too much the gameplay with hero bans. The same reasoning applies to supports.
In this stream from November, Jeff Kaplan confirmed that they know that they need to release more tanks and supports, and they are actually working on the next six heroes already.
Considering the lack of heroes, I think that a hero lock system is the “easier” way to fix the game. Furthermore, the 2/2/2 lock would get rid of one of the biggest issues for pub players: not being able to play what they want.
Some players are against the role lock because it limits the team composition creativity.
I believe it won’t be a problem. Yes, currently we may think that, but note that the game will be rebalanced for a 2/2/2 meta composition, so some heroes you are playing now might be different in the future. Jeff Kaplan explained that it will be easier to balance heroes; certainly the devs will do their best to be sure we will be able to create creative comps also in a 2/2/2 world.
At the end of the day, it is not possible to make everyone happy. When Mercy was a must pick in every OWL game, the community complained wildly asking for a meta change, so she was nerfed heavily several times.
While many players were happy about it because it caused the rise of new supports, the OW devs also had to deal with hundreds of thousands of Mercy mains who probably still haven’t forgiven them today.
Honestly, it was a necessary nerf, and Mercy never turned into a trash pick: she maintained a respectable pick rate at most ranks even in her darker times.
Master hero meta
Tier 1 (>10%): Ana 12.3%.
Tier 2 (>7%): Reinhardt 7.5%.
Tier 3 (>4%): Lucio 6.6% - Zenyatta 5.7% - Zarya 5.7% - Genji 5.1% - Winston 5.0% - Widowmaker 4.4% - D.Va 4.3% - Wrecking Ball 4.0%.
Tier 4 (>2%): Mercy 3.9% - Hanzo 3.6% - McCree 3.5% - Roadhog 3.4% - Tracer 3.1% - Orisa 2.6% - Soldier: 76 2.5% - Baptiste 2.4% - Ashe 2.3% - Doomfist 2.2%.
Tier 5 (<2%): Sombra 1.9% - Moira 1.8% - Torbjorn 1.5% - Brigitte 1.2% - Mei 0.9% - Symmetra 0.7% - Pharah 0.6% - Reaper 0.6% - Junkrat 0.5% - Bastion 0.3%.
Most successful heroes:
Diamond hero meta
Tier 1 (>10%): Ana 11.4%.
Tier 2 (>7%): Reinhardt 9.1%.
Tier 3 (>4%): Zarya 6.2% - Zenyatta 5.5% - Genji 5.3% - Lucio 5.1% - McCree 4.2%.
Tier 4 (>2%): Mercy 3.9% - Winston 3.9% - Hanzo 3.7% - Widowmaker 3.7% - D.Va 3.7% - Roadhog 3.6% - Moira 3.5% - Wrecking Ball 3.5% - Ashe 3.0% - Baptiste 2.6% - Tracer 2.4% - Soldier: 76 2.4% - Orisa 2.2% - Doomfist 2.1%.
Tier 5 (<2%): Sombra 1.5% - Brigitte 1.3% - Torbjorn 1.3% - Reaper 1.2% - Mei 1.0% - Pharah 0.9% - Junkrat 0.8% - Symmetra 0.7% - Bastion 0.3%.
Most successful heroes:
Platinum hero meta
Tier 1 (>10%): Reinhardt 10.7% - Ana 10.1%.
Tier 2 (>7%): None.
Tier 3 (>4%): Zarya 6.0% - Moira 5.8% - Zenyatta 4.7% - Genji 4.7% - Mercy 4.5% - Lucio 4.3% - McCree 4.2% - D.Va 4.1%.
Tier 4 (>2%): Roadhog 3.6% - Hanzo 3.5% - Winston 3.1% - Ashe 3.0% - Wrecking Ball 2.9% - Widowmaker 2.8% - Baptiste 2.7% - Soldier: 76 2.3% - Reaper 2.2% - Orisa 2.1%.
Tier 5 (<2%): Doomfist 1.8% - Tracer 1.6% - Brigitte 1.6% - Junkrat 1.5% - Pharah 1.5% - Sombra 1.3% - Torbjorn 1.2% - Mei 1.2% - Symmetra 0.9% - Bastion 0.4%.
Most successful heroes:
Gold hero meta
Tier 1 (>10%): Reinhardt 11.4%.
Tier 2 (>7%): Ana 8.6% - Moira 7.9%.
Tier 3 (>4%): Mercy 6.2% - D.Va 5.2% - Zarya 5.0% - Lucio 4.3%.
Tier 4 (>2%): Zenyatta 3.9% - Genji 3.5% - McCree 3.3% - Roadhog 3.3% - Reaper 3.1% - Hanzo 3.0% - Junkrat 2.7% - Baptiste 2.6% - Orisa 2.6% - Winston 2.6% - Ashe 2.4% - Wrecking Ball 2.4% - Soldier: 76 2.3% - Brigitte 2.0% - Widowmaker 2.0% - Pharah 2.0%.
Tier 5 (<2%): Mei 1.5% - Doomfist 1.4% - Torbjorn 1.3% - Sombra 1.1% - Tracer 1.1% - Symmetra 1.0% - Bastion 0.6%.
Most successful heroes:
Silver hero meta
Tier 1 (>10%): Reinhardt 11.0%.
Tier 2 (>7%): Moira 8.9% - Mercy 7.1% - Ana 7.0%.
Tier 3 (>4%): D.Va 6.2% - Lucio 4.8% - Zarya 4.0%.
Tier 4 (>2%): Junkrat 3.7% - Orisa 3.3% - Reaper 3.3% - Zenyatta 3.3% - Roadhog 3.0% - Genji 2.9% - Hanzo 2.7% - Soldier: 76 2.6% - McCree 2.5% - Winston 2.3% - Baptiste 2.3% - Brigitte 2.2% - Wrecking Ball 2.2% - Pharah 2.1% - Ashe 2.1%.
Tier 5 (<2%): Mei 1.8% - Widowmaker 1.7% - Torbjorn 1.7% - Symmetra 1.3% - Doomfist 1.1% - Sombra 1.0% - Tracer 1.0% - Bastion 0.9%.
Most successful heroes:
Bronze hero meta
Tier 1 (>10%): None.
Tier 2 (>7%): Moira 9.2% - Reinhardt 9.2% - Mercy 8.0%.
Tier 3 (>4%): D.Va 6.9% - Ana 5.2% - Lucio 5.0% - Orisa 4.4% - Junkrat 4.3%.
Tier 4 (>2%): Reaper 3.1% - Zenyatta 3.1% - Soldier: 76 2.8% - Hanzo 2.7% - Roadhog 2.6% - Genji 2.6% - Zarya 2.5% - Brigitte 2.5% - Torbjorn 2.3% - Pharah 2.3% - Wrecking Ball 2.2% - Baptiste 2.2% - Widowmaker 2.1% - McCree 2.0% - Winston 2.0% - Ashe 2.0%.
Tier 5 (<2%): Mei 1.9% - Symmetra 1.9% - Bastion 1.7% - Sombra 1.2% - Doomfist 1.2% - Tracer 1.1%.
Most successful heroes:
Past Meta Reports:
Next Meta Report: Season 16, May 2019