All Minor and Major Perks in Overwatch 2 — the Competitive Variable the Overwatch League Cannot Yet Measure

2:47 in the morning at an analysis room in Gangnam. The third monitor freezes on frame 214 of a closed scrim. A hero has just hit in-match level 2, and its Major Perk unlocks at the exact moment two teams are locked against each other on the objective. Seven seconds later, the defending formation collapses. There is no individual play worth clipping. No ultimate is fired. Only a status bar changes colour in the lower corner of the screen.

When the numbers do not lie, my heart only then begins to listen.

I grew up with football, where every match starts with two lineups that are identical in terms of tools and ends with numbers that can be verified. Eleven a side, same rules, same clock, and every difference lives in how they use identical cards. Overwatch 2 has just torn that assumption apart. The Perks system inserts a layer of evolution inside the match itself: your hero grows stronger while the match is running, and grows stronger in different ways depending on the choices of each player, each team, each moment.

That is why I stayed up until almost three in the morning. A system that generates power while a match is running will collapse nearly the entire analytical framework I have used for twelve years.

Context: How Perks Actually Work

Blizzard announced the Perks system during the Overwatch 2 Spotlight on February 12, 2026, and put it live with Season 15 on February 18, 2026. Formally, it is an unlockable upgrade layer available mid-match, running in parallel with each hero's base kit.

The core mechanic fits into a few lines. Every hero has a separate in-match level bar, and that bar levels up twice in a standard match. Level 1 unlocks a Minor Perk — a light upgrade that usually tunes a stat or adds a small effect to an existing ability. Level 2 unlocks a Major Perk — a heavy upgrade capable of changing how a hero is played, rather than stopping at added damage or a shorter cooldown.

Experience points come from three sources: playing the objective, kills or assists, and healing or saving teammates.

Each hero has its own perk set, designed according to a philosophy Blizzard calls reinforcing hero uniqueness. Upgrades must carry that hero's signature, not use generic stat packages applied across the whole roster.

This is where I want to pause, because it determines everything that follows. Blizzard chose to expand depth along a vertical axis rather than a horizontal one. They did not give every hero five percent more movement speed or two seconds off a cooldown. They built a progression layer inside the match, where each hero has its own path, and that path only opens once the match has already run a quarter of its length.

I have counted every empty space on the pitch when the crowd disappeared.

Where Minor and Major Diverge

Most community discussion stops at sorting which perk is strong and which is weak. The two-tier structure matters more than that.

Minor Perks are tuning tools. They address small weaknesses in a kit, or accentuate a behaviour the hero already leans toward. They arrive early, meaning they shape how a hero is played across the first two-thirds of a match.

Major Perks are reshaping tools. They change how a hero interacts with space and with opponents. This group arrives later, meaning they shape the decisive fights, when the final objective is being contested.

The power gap between Minor and Major is an unquantified variable, and it determines the entire tempo of a match. If Major Perks are overwhelmingly stronger, the level-2 breakpoint becomes the true win condition, and level 1 becomes only a psychological stepping stone.

Blizzard has published no quantified table for that gap. Based on their design history, the leaning is toward Major Perks carrying influence on the order of a strong secondary ability, meaning large enough to swing a fight.

Three XP Sources and the Trap of Passive Play

The experience allocation table sounds harmless, but it is the part I care about most in the whole system.

The three sources are playing the objective, kills or assists, and healing or saving teammates. All three revolve around a single behaviour: being present in the team fight.

Compare that with how a solo operative generates value. A flanker who cuts angles, a sniper who applies pressure from range, a Tracer who harasses the back line — they create value by selecting small fights, not by attending every five-on-five collision. The new experience system does not punish them, but it does not reward them at the same rate.

The consequence is a structural subsidy for collective play, and a hidden tax on single-lane play.

I have seen something similar in football, when a league adjusts substitution rules. Teams that play in structural blocks, holding spacing and rotating the ball steadily, benefit immediately. Teams that live on individual moments need a few rounds to adapt, and a few never adapt at all.

Here the dynamic is stronger. Experience does not flow into players simply for joining a fight. It flows in when they join in the right place. Standing on the objective, holding the objective, healing the person holding the objective, supporting whoever is fighting — all of it is behaviour rewarded directly with a resource that exists only inside this match and vanishes when it ends.

A player standing at the edge of a fight looking for an angle receives nothing from that stream. A player standing in the middle of the objective and absorbing damage receives it. This is a very clear design statement, delivered through an allocation table rather than through a patch note.

Level 2 Is the Real Win Condition, Not the Perk

This is the part I consider the most under-rated in every discussion of the system.

Explainers usually stop at listing which perk belongs to which hero, which is strong, which is weak. But the more important operational question sits elsewhere: when does one team reach level 2 on its core heroes, and how long does the window last in which the other team is still at level 1?

In a match where both teams play correct structure, that gap may be only a few dozen seconds. In a match where one team is pushed off the objective across three consecutive phases, that gap can stretch into two minutes. And across those two minutes, the team that reached level 2 first is playing with a qualitatively different toolkit.

The win condition lies in securing the level-2 breakpoint across multiple heroes at once, then using that power window to take the objective and accumulate more experience.

This is a self-reinforcing loop in the literal sense. The team that hits the breakpoint first is more likely to win fights, which makes them more likely to hold the objective, which gives them more experience, which lets them reach level 2 on their remaining heroes sooner. In football xG analysis, I call loops like this a skewed equation. There is no surprise here, only a skewed equation accelerating.

What is worth noting is that this loop does not require an outstanding play to start. It only requires one team to hold the objective roughly fifteen seconds longer than its opponent across two consecutive phases. Fifteen seconds is a small enough window to be ignored in a high-tempo match, and large enough to create a level gap that individual skill cannot close.

Which Heroes Gain and Which Heroes Pay

From the structure above, several groups emerge clearly.

The first group is tanks with strong baseline kits and good objective-holding ability. DVa and Winston are the two most frequently cited examples in early discussion. Both already have base kits strong enough to apply pressure, meaning perks act as force multipliers on an already-solid foundation. When you add to something already strong, the resulting gap is not addition, it is multiplication.

The second group is supports with strong utility kits. This is a direct consequence of the experience table: healing and saving teammates are level-up sources, meaning a support playing the correct role levels steadily regardless of how the match unfolds. A sniper, by contrast, earns experience only on successful kills — a high-variance income stream dependent on whether opponents expose angles.

The third group, and the most concerning, is heroes with weak early games. A match runs for a stretch before the first perk appears, and during that stretch, any hero without the tools to affect the objective falls behind both tactically and in experience terms.

Put another way, the system creates a longer opening phase for the very heroes that were already weak in the opening phase.

Callback Perks and the Ghost Power Problem

There is one technical detail I consider the most important in the entire perk list: some perks restore abilities that were removed during the transition from Overwatch to Overwatch 2.

The clearest example is Bastion's Self-Repair. It was once a signature ability in the original Overwatch, removed during the rebuild for the sequel. It now returns as an in-match perk. That means Bastion can re-establish the sustain model that once made it considered overpowered, except the model now appears later in a match rather than being available from the start.

The second example is Orisa's Barrier. Orisa lost her shield entering the sequel and was redefined as a brawling tank. Restoring the barrier through a perk does not stop at adding an ability — it changes the hero's role identity. An Orisa with a barrier is an Orisa who can play the front line in a completely different way, with an entirely different tempo of corner pressure and space denial.

This is ghost power: capability from earlier versions of the game returning through a side door, carrying an entire body of old balance data that nobody still controls.

In football transfer analysis, I used this term for players priced on form from two seasons ago. Here the problem is more systemic: some heroes get back exactly the tools that were deleted for balance reasons, while other heroes get only minor tuning. The scale is asymmetrical at the design stage, and that asymmetry only becomes visible once the system runs long enough to generate data.

Hero Switching and the Perk Banking Strategy

A secondary mechanic, rarely discussed but with the potential to create a new strategic layer, is how the system handles hero switching.

When you switch heroes mid-match, the new hero's level bar starts from zero. But if you return to the previous hero, they retain the perk level already unlocked, and the next unlock happens faster.

Read those two lines closely and a clear tactical opportunity appears. A team can deliberately rotate to generate a perk advantage, or to break an opponent's accumulation rhythm. A concrete example: if you force a key enemy hero to change role by pressuring their composition structure, that hero resets to the starting line, while their replacement needs time to climb to level 2.

I call this perk banking, and it has the potential to become a pre-match preparation layer comparable to ban-pick in MOBA titles.

I do not believe in inspiration — I believe in standard error.

At this point, no professional team has published data on hero-swap strategies for perk advantage. But every system that has ever created an in-match progression layer has produced this kind of tactic within its first few weeks, and I have yet to see an exception.

Environmental Variable: The Patch Lock Problem at LAN Events

This is the risk I consider most serious, and it sits outside the control of any team.

Blizzard has stated that new heroes will receive perks, and that the system will be updated every other season. That means the competitive server and the live server can diverge in version at exactly the moment an international event takes place.

In esports history, this type of controversy has repeated many times. A major event is forced to choose between two options: lock the old version to guarantee fairness, or use the new version to reflect the game's actual current state. The first option makes the event look outdated compared with what viewers are playing at home. The second gives teams less preparation time for a new system.

With Perks, the severity is considerably higher than with an ordinary stat patch. A stat patch is published publicly, applied uniformly to everyone, and can be simulated in practice. Perks create power divergence that depends on individual player choices and individual match tempo. There is no way to fully simulate a variable like that before stepping onto a competitive server.

Tournament formats are also affected differently. In single-elimination, early perk choices can decide the shape of the very first contest, turning the match into a gamble over which team understands the system better. In a losers bracket, the team that drops down gains extra data and learning time, creating an adaptive advantage that older formats did not have.

Correlation Is Not Causation: The Contrarian Angle

Now to the part where I argue against myself.

One reading of the Perks system is spreading through the community: the team that wins early fights wins the match, because perks create a snowball effect. That reading sounds reasonable, and it has numbers behind it if you look at the correlation between level-2 timing and win rate.

Correlation is not causation.

The team that reaches level 2 first is usually the team that was already playing better in the opening phase. They control the objective better, they win more fights, and therefore they level faster. In most cases, match control itself is the cause, and the level-2 breakpoint is only an accompanying variable.

To separate the two, you need a specific dataset: matches where the team that reached level 2 first then lost the immediately following fight. If that rate is low, we have evidence that the perk itself creates an advantage. If that rate is high, we are looking at an inflated effect.

My hypothesis, at the current level of data, is that perks amplify gaps more than they create them. The snowball effect exists, but it operates on a trajectory established earlier, rather than initiating that trajectory.

In my world, luck is only the residual I have not yet explained.

The Value of Flex Players and the Cost of One-Hero Specialists

There is a personnel consequence I believe will reshape how organisations build rosters.

The hero-switching mechanic creates an advantage for players who can competently operate multiple heroes within the same role. A support who plays three different heroes has more tactical options than a specialist who plays one. In a system where swapping heroes can be a strategic move, the ability to swap becomes a marketable skill.

Conversely, one-hero specialists face a double risk. First, if their signature hero's perks are weaker than the opponent's hero's perks, they cannot compensate with individual skill as before. Second, if the meta forces them to switch, they fall back to the starting line while the opponent already has perks banked.

I expect the market value of flex players to rise across the next two to three transfer windows, before the market corrects itself.

This is what I saw in football when the full-back role changed: the price of a player who could run both flanks spiked for two seasons, then re-balanced once supply caught up. Esports markets tend to have a shorter lag, but the mechanism is the same.

This also creates a new coaching role. Teams will need someone dedicated to analysing perk timing, selection priority, and counter-perk strategy. That is a job title that does not exist in most organisations' current staffing structures.

The Spectator Problem: One More Layer of Information

I have to be blunt about this risk, because technical analyses rarely mention it.

Overwatch is already a hard sport to follow for a general audience. Fight tempo is high, visual effects are dense, and most important information lives inside an individual player's perspective rather than in the wide view.

Adding an in-match progression layer substantially increases the volume of information casters must communicate and viewers must track. A fight one side wins because a perk just unlocked looks identical to a fight one side wins because of better mechanics, unless the system displays it clearly.

The viewing experience depends on the quality of the interface, not the quality of the system design. A perfectly balanced system with weak display produces a feeling of chaos. A temporarily imbalanced system with clear display produces a feeling of strategy.

The empty-stadium season was the largest laboratory I ever walked into, and the lesson I took from it still holds: viewers do not need to know every detail, but they need to know who is stronger right now and why.

Comparison with Similar Systems Across the Industry

Perks did not appear out of nowhere.

Apex Legends uses Evo Shield — a system where a player's armour strengthens based on in-match performance. Fortnite uses weapon rarity upgrades. Paladins uses a loadout system that lets players customise abilities per match. Paragon once tried a card system before the game was shut down.

The common thread among successful systems is that they reward behaviour the game wants to encourage, and they display that reward clearly enough for players to understand where they sit on the power curve.

The common thread among failed systems is that they create too large a power gap between the leader and the chaser, deciding the match before it ends.

Perks sit between those two zones, and its exact position will be defined by balance data over the coming months. What makes Overwatch different is that this system is being applied to a game with twelve years of balance history and a community intimately familiar with every frame of the original kits. The expectation bar here is far higher than for a new title introducing a similar system on day one.

Competitive Rules and Integrity

At this moment, no rules framework has been published for how Perks will apply in professional play.

Three questions need answers before any tournament enables the system. Are perk selections visible to both teams? If yes, coaches can react in real time. If not, the opposing team has no way to counter-play, and the strategic value drops sharply.

If a perk is discovered to be too strong mid-season, who is accountable for the fallout? That is a genuine governance question, not a technical one.

And finally, does the system run fully server-side or does part of it run client-side? Any new system opens a new exploitation surface, and esports history shows exploit bugs typically appear within the first two months after a new mechanic launches.

A worst-case scenario worth imagining: an unintended perk combination is discovered during an official match, that match result is suspended, and the tournament must handle a validity controversy with no precedent to reference.

Impact on the Content Ecosystem

There is one positive side I want to acknowledge.

Perks create an entirely new stream of educational content. Tier lists by perk, situational perk guides, counter-perk strategy, breakpoint timing analysis — all of it is material for long and short content. During the period when the community has not agreed on which perk is strongest, every analysis has value and every position is arguable.

This is something organisations with strong content arms can exploit immediately. The learning phase lasts roughly three months, and during those three months, viewership belongs to whoever analyses fastest and most accurately.

Four Signals to Track in the Next Cycle

If you are tracking this system the way I am, these are the four signals to put on your list.

The first is Overwatch League and international event policy. If Perks are disabled in professional play, everything above becomes irrelevant to the competitive scene, and the system remains only a ranked-mode feature. If Perks are enabled, this is the biggest change to the discipline since the shift to five-player lineups.

The second is the list of perks the community flags as broken. That is an early indicator of balance problems. When a community converges on a specific perk within days, it is usually a sign of a design flaw rather than a tactical choice.

The third is the emergence of deliberate hero-swap tactics in scrims. If teams begin swapping heroes to break an opponent's perk accumulation, we have a genuinely new strategic layer, and every old analytical framework needs rewriting.

The fourth is active player numbers. If the system keeps players in the game longer, Blizzard will keep expanding it. If it drives players away, it will be narrowed or removed within a year.

Forward-Looking Thought

The Perks system introduces a variable into Overwatch 2 that esports analytics has no tool to measure. In football, I have xG, I have PPDA, I have minutes played and touches on the ball. All of them are stable across a match and comparable across matches.

Here, a hero's power depends on the moment within the match, on the player's choice, and on the fight tempo both teams create together. That is a function with more variables, and I need more data before I claim anything about it.

What I know for certain right now is that professional teams are walking into unmapped territory, and the teams that build the fastest learning systems will hold the advantage for the first six months. After that, the rest of the league will close the gap, and the edge shifts to whoever can run the most stable structure under a new system.

If you want to test my hypothesis, here is how. Take one week of matches, record the moment each hero reaches level 2, record the outcome of the fight immediately following that breakpoint, then split two groups — teams that won and teams that lost the fight before the breakpoint. Once you have that data, you will know whether perks create gaps or merely amplify the ones already there.