Sombra Leaves Damage, Roadhog Loses the One-Shot Combo: Overwatch 2 Season 5 Is Rewriting the Rules

In an old VOD I still keep on a hard drive, Roadhog lands a hook on an Ana hiding behind a corner, yanks her in, and the fight ends before the caster can finish saying its name. No ultimate. No elaborate setup. One hook and one shot. That is why Roadhog exists in the eyes of a certain kind of player: clean, brutal, and rewatchable.

Season 5 of Overwatch 2 has just taken that moment away. The Scrap Gun now fires in two bursts. The one-shot combo is gone from the game. On the other side of the patch, Sombra — a hero who was crossed off in virtually every ban draft as a damage dealer — has been moved down to the support role, carrying a kit that keeps its old shape while changing its effects entirely.

Two heroes. Two overhauls. The same moment. Across years of watching Overwatch competitions, I have developed a habit: wait three weeks before saying anything about a major patch. This time I could not wait, because what changed is not the power level of two heroes. It is who this game allows to occupy which role.

To understand why these two changes weigh more than they appear to, you have to remember how Overwatch behaves at the competitive level. There, a hero's strength is measured by two dry numbers: pick rate and ban rate. A hero banned in nearly every match is not necessarily the strongest hero. It is the hero both teams believe they cannot afford to let onto the map.

Sombra lived in that state as a damage hero. She did not dominate through raw damage. She dominated by switching off other people's abilities. Hack locks abilities. Translocator gives her an escape that is nearly impossible to punish. In a game where most value comes from pressing the right button at the right second, a hero who can erase the opponent's buttons is a structural problem, not a tuning problem.

The ban phase became the release valve. When both teams cross off the same hero, the league keeps running, viewers keep watching, but inside the room everyone knows a hole is being covered by procedure.

Some matches do not happen on the pitch. They happen deep inside people. For Overwatch, that match happens in the ban room: a small, silent room where two teams agree that one hero will not be permitted to exist in the upcoming game. Nobody records that moment on the scoreboard. It decides the match more than any play.

Season 5 walks straight into that hole and does two things at once. With Sombra, it does not tune the hack mechanic — it moves her into a different role, where that mechanic means something else. With Roadhog, it tears out the kill toolkit and installs a teamwork toolkit in its place.

In the taxonomy of esports updates, this belongs to the rework class: changes designed to break structure rather than rebalance it. Two reworks arriving together force the community to learn two new variables at the same time.

One technical point needs stating up front: the patch notes ship with no win-rate, pick-rate or ban-rate data whatsoever. Every conclusion below, mine included, is inference drawn from ability descriptions. I say this first so that the rest of the piece is read at its proper weight.

Sombra becomes the third character in the game's history to change competitive role — a milestone the developer notes themselves confirm. Her path runs from lone assassin, a hero banned almost permanently, down to support. That phrasing sounds like a demotion, and in theory it is one: support is the least damaging role, the least capable of creating sudden swings on its own, and the most dependent on teammates.

But the new kit tells a different story. Sombra's abilities keep their familiar shape — the same hack mechanic, the same Translocator — while the effects have been rewritten to serve a new role. The most notable detail: EMP is untouched.

EMP is one of the heaviest ultimates in the game. It does not need damage to change a fight — it strips shields, locks abilities, and cracks an enemy formation open for a few seconds. When Sombra was a damage hero, EMP came packaged with finishing power, making her a double threat: she broke the defence and then killed through it. In the support role, EMP loses the finishing power but keeps the defence-breaking.

Here is the paradox: pushed down into a role that is weaker in theory, Sombra gains something close to exclusivity within it — a support carrying the strongest fight-initiating ultimate in the class.

Four perks are recorded for her — CTRL ALT ESC, Life Hack, Data Packet, Cybersecurity — adding another layer of adjustment. That is the part I want to return to, because it carries longer-term consequences than the role change itself.

The immediate consequence is clear. An entire class of compositions disappears from the strategic space. Teams that spent months building plans around Sombra as a damage dealer must start again from a blank page. Teams that used a ban slot to neutralise her must now recalculate where the threat sits: it is no longer where they are used to guarding.

Roadhog moves in the opposite visual direction but follows the same logic. His old kit was built around a single moment: land the hook, pull the target in, finish the kill. It was a solitary playstyle, selfish in the tactical sense, and deeply unpleasant to play against.

The rework pulls that moment apart. The Scrap Gun now fires in two bursts, making burst damage harder to stack. The compensation arrives through an entirely new ability: Trash Compactor, an alternate fire that absorbs enemy projectiles and fires them back as an explosive ball — a function compared directly to D.Va's Defense Matrix.

Anyone who has watched D.Va understands the meaning immediately. Defense Matrix is a collective defensive tool: it kills nobody, it destroys other people's plans. Trash Compactor turns Roadhog from hunter into cleaner, from lone gun into a link that protects a formation.

The accompanying perks reinforce that direction. Scrap Hook Minor is buffed to reload four ammo instead of two after each successful hook — a small technical change that says a great deal about design intent: the developers want Roadhog to stay in the fight longer, to apply sustained pressure instead of waiting for one perfect moment. Piggy Piggy and Toxic Exhaust complete the picture.

In football language, this is the story of a number nine who only knows how to finish being converted into a target man. For a team whose striker is the last link in the chain, losing the finishing touch removes the whole system. For a team that needs someone to hold the ball, drag defenders and open space, he becomes considerably more valuable.

Roadhog loses individual power in exchange for team value. It is a design gamble, and it only pays out if teams agree to play the new way.

For anyone who has tracked enough patches, the pattern here is familiar: when a hero is stripped of the tool that creates sudden swings, teams do not immediately mine the compensation. They drop him. It takes weeks, sometimes months, before somebody notices the compensation is stronger than what was removed.

This is the most important part of the patch, and the most underrated. The perk system is a modular customisation layer that lets players steer a hero's abilities in specific directions.

The real value of this system is not how strong it is, but that it adds a preparation layer to the game — and every new preparation layer is an opportunity for the team with the better analytical infrastructure.

Previously, a team's preparation revolved around picking heroes and building a composition. Now the same hero can be played in several different ways across different games, depending on the opponent. In best-of-three or best-of-five formats, this opens the possibility of adjusting between games — something that barely existed at the individual level before.

Picture a football team able to change formation between the first and second half, not by substituting players, but by changing how those same people move. That is the degree of flexibility the perk system introduces, and the degree of complexity it imposes on coaching staffs.

The cost of that flexibility is not cheap. Every perk choice has to be verified in scrims, logged, and cross-referenced against specific opponents. There is no public win-rate data for each combination. No leaderboard tells a team which build beats which in a given situation. That work belongs to people, in a room, after practice.

Two reworks landing together create what analysts usually call a meta chaos window. In sport, when the rules change at several points at once, the gap between strong and weak teams tends to narrow in the short term — not because weak teams improve, but because strong teams lose accumulated advantage.

The advantage of a strong esports team comes from thousands of scrim hours inside a stable environment. When that environment is shaken, the advantage evaporates first and the base skill remains. For a short stretch, two teams at different levels can look nearly equal.

This is why I always watch the second and third weeks of a new season. That is the window in which results reflect true strength least accurately — and the window in which people betting on emotion lose the most money.

One thing the patch notes do not say needs saying plainly: there is no win-rate, pick-rate or ban-rate data yet. All we have are ability descriptions and a belief that the developers know what they are doing.

That belief is not always repaid. The history of esports reworks is full of cases where a hero was pushed into a new role, became overpowered for two weeks, was hotfixed in a hurry, and then became useless for the following six months. That spiral is not a sign of incompetence. It is a sign of balancing a game with dozens of interacting variables, where changing one point always shakes another.

Here I want to step away from the technical description and say what I actually think.

The community is arguing about whether Sombra is strong or weak in her new role. That framing is wrong. The real question sits elsewhere: why did a hero have to be banned in nearly every match for months on end?

The ban draft cannot fix a design flaw. It only hides that flaw from the audience. When two teams agree to erase a hero from a match, they are telling each other that this game, in its current state, cannot handle that hero on its own. Moving Sombra to support keeps her inside the game and hands her a new reason to exist. It does not solve the underlying problem — the hack mechanic is still there, wearing different clothes.

Put another way, the developers have legitimised the existence of a hero who used to be permanently banned. That is a reasonable commercial solution. It is not yet a design solution.

The second point concerns perks, and it is the part of this piece I believe most firmly. The community is praising the new strategic depth. Depth only has value if somebody mines it. Most esports organisations spend their money on star players — a version of the youth-academy model I have seen everywhere in football: glitter at the top, emptiness underneath.

What this patch rewards is not mechanical talent. What it rewards is analytical infrastructure: the editor watching VODs at two in the morning, the data assistant, the person logging every perk build against every opponent. That infrastructure cannot be bought in a single transfer window.

Chicago Fire taught me that football always knows how to trample the script. I started the blog Hiep Ba in 2026 to write about the same things in football, and I learned that the script gets trampled in the same way in every sport with enough data for people to believe they have figured it out.

Where could I be wrong? If Blizzard ships a hotfix within three weeks, if Sombra support gets buffed because she is too weak, if Trash Compactor turns out to be too strong inside a specific combination, everything I have written above will need rewriting. That is the risk of analysing a patch in its first week. I accept that risk rather than waiting, because the value of speaking early lies precisely in the period when you are not certain.

What I will be watching over the next six weeks, and what you can verify alongside me: upset rates rise in the opening stretch of the season; Sombra support appears mainly in fast compositions where EMP is used to open fights; Roadhog disappears from burst lineups and returns later, in a different role.

I write Hiep Ba to talk about football, and it turns out I have been talking about myself — and now, about the people playing a game I only get to watch from the outside. Esports keeps lying to every chart we have. The job of the writer is to record that lie early enough for it to become a verifiable fact, rather than a late excuse.