How Competitive Skill Rating Works (Season 12)

The rating and matchmaking system is confusing, and a good overview does not appear to be available. The official overview (1, 48) is incomplete and does not answer a number of common player questions and concerns. This information below is gathered from sporadic blue posts and developer update videos, and salted with my own experience and experiments, various forum threads, and watching streams. Note that since Blizzard does not give exact algorithms, I do have to fill in some gaps, or leave some items unknown. I will edit this post if information becomes available before the end of this season. Next season I will make a new post and include changes.

Whenever you see a number in parenthesis, look to the references section and follow the link to see the support for the statement in question.

Changes in Season 12

I’ve removed the section on Skill Rating -> High win percentage debuff / Low win percentage buff. It was always a bit speculative, the data it is based on is rather old, and more recent data ( -> Dmum 11) does not show the effect.

Similarly, I am no longer confident that typical MMR movement is 19.0, so I am removing that statement. At least for diamond+, for non-decayed accounts, MMR and SR seem to be moving in sync, but my data is not particularly solid. I now consider MMR movement to be unknown.


See the later sections for a more in-depth exposition, as well as references to supporting material.

How do I maximize my Skill Rating?

In order:
1) Win games
2) If diamond or above, play a minimum of 4.67 games per week (starting one week after placements).
3) If platinum or below, perform well, statistics-wise, with the chosen hero.

In my last game, my Skill Rating went up/down by X. Why did this happen?

The details of Skill Rating (SR) movement in a match depends on whether a player wins or loses, the Match-making-rating (MMR) mismatch between the two teams, the mismatch between the player’s SR and MMR, whether a player is new, whether a player is active, the player’s statistical performance, the player’s position on the ladder, and possible bugs. Add it all together and it is nearly impossible to determine why a player’s SR has a specific movement in a particular match. Sometimes over the course of twenty to one hundred matches it can be seen that one particular cause is dominant, but that often depends on careful data collection and some guesswork.

Summarize matchmaking, rating, and progression for me

MMR is a (hidden) number that goes up when you win, and down when you lose. How much it goes up and down is a complicated formula, based on many factors, but the essential truth of wins are good and losses are bad is certain (4).

That MMR is then used to match people with and against people of similar MMR, in an attempt to create a 50% match (32). Here is the key part:

If a player’s MMR is wrong and too low, then the odds to win will be greater than 50%, and the player will win more games than he loses, which will cause his MMR to rise over many games played. He will then be placed with stronger and stronger opponents (and stronger and stronger allies) until his MMR is correct, and his win percentage approaches 50% (with some random oscillation around 50%) (7).

Once a player finds himself trapped in a rating range, the only way to break out is to improve as a player and play enough games to overcome any random factors (47, 48 at the end).

What is SR then? SR is a visible and friendlier approximation of MMR (23). It has no more meaning with respect to matchmaking than tier icons. However, except for top players who have decayed, MMR and SR are closely linked, so inspecting SR typically gives a reasonable estimate of a player’s MMR (23).

< 500 Bronze, SR not listed
500-1499 Bronze
1500-1999 Silver
2000-2499 Gold
2500-2999 Platinum
3000-3499 Diamond
3500-3999 Master
4000-5000 Grandmaster
Top 500 players in region.

For season 3, the rank distribution was Bronze: 6%, Silver: 22%, Gold: 34%, Platinum: 23%, Diamond: 10%, Master: 3%, Grandmaster: <1% (3). For season 8, the rank distribution was Bronze: 8%, Silver: 21%, Gold: 32%, Platinum: 25%, Diamond: 10%, Master: 3%, Grandmaster: 1% (6). Third-party sites such as masteroverwatch and overbuff do not give reliable distributions because players have to actually log in to those sites for it to count the data. This skews those sites heavily toward higher ranked players.

As a player gains SR, he will be promoted to the next tier when appropriate. He will drop out of each tier as he loses rank. However, there is some loss protection for Diamond and below. After a loss (but not a win), the game will check to see what a player’s current tier and skill rating are. If his skill rating has not exceeded the minimum for his current tier for the past five games, he will be demoted (16). For Master and Grandmaster, a player will be demoted immediately if his skill rating is not high enough. Competitive point rewards are based on the highest rank achieved during the season (16). Ranks and SR are wiped each season (but not MMR, see below).

The top 500 leaderboard becomes available two weeks after season start. To be eligible, a player must have played at least 50 games that season (45), all from one region. The player must have had SMS protect enabled for all 50 of these games (50, 51, 53). The top 500 spray and icon are not awarded unless a player is in top 500 at the end of the season (18).


Matchmaking is based on hidden match-making-rating (MMR), not skill-rating (SR) (3, 21, 25, 40). Competitive MMR is separate from the MMR of other modes (37).

The game uses MMR to determine whether a match is fair, and attempts to match teams such that the game is fair (each team has a 50% chance of winning) (32). Upon victory, the winners receive a bump up in SR and MMR, and the losers get bumped down. At times (off peak-hours or very high/low rated players), finding a fair match will be impossible in a reasonable amount of time, and the match maker will match teams based on a best fit. However, note that if no matches can be made with an expected win percentage between 40% and 60% (the infamous Brazilian six stack playing at 3 AM), then no match will be made and queue times will become extreme to infinite (48). In the middle ground, top (or bottom) players will have a longer, but not infinite queue, to attempt to find a better match (15).

If the match making algorithm determined before the game that it was not fair, the higher ranked team will receive less SR/MMR on a win, and lose more SR/MMR on a loss. The lower ranked team will receive more SR/MMR on a win, and lose less SR/MMR on a loss (2). This is standard for Elo (rating system from chess) type systems. The SR/MMR penalty/bonus is designed to be fair, such that if you win 40% of your games with a predicted 40% expected win probability, your SR/MMR does not change (48). Draws cause no change in SR and have an unknown effect on MMR.

Matchmaking and groups

In addition to MMR, the match maker attempts to match on ping and group size (32, 48) and to make matches reasonably quickly. (Jeff also mentions quarantining new players from the general population, but this effect does not apply to competitive).

Predicted win probability (which is used to calculate fair matches and MMR/SR rewards) does not depend on grouping (49). This leads to a small handicap for solo players vs groups, and smaller groups vs larger groups, for when the matchmaker cannot find groups of equal sizes to match against each other. The actual (as opposed to predicted) average win rate for solo players is 49.94% and for six stacks is 52.88% (48, see this reference for more numbers). Smaller groups assumedly have win rates greater than 49.94% and less than 52.88%.

However, if a player groups with someone with whom they have poor synergy, it acts as a penalty on their practical win likelihood, because they are likely to be matched against groups of equal size, with more synergy and experience. Finally, large groups tend to have longer queue times while the matchmaker tries to find other large groups. With the LFG tool now being live for competitive, this may be less of a problem as there are more groups in the system.

If the match-maker says most games are fair, then why are there so many stomps?

There are many reasons:

  1. Overwatch, as a game, has a tendency to snowball. The winner of the first fight has an ultimate advantage that has a tendency to last the round. Losing teams tend to tilt and start playing poorly, which can continue in the same round and carry into the next round.

  2. Random variables are random. Maybe a cat walked on a keyboard or someone fat fingered an ult (7). Maybe someone who has 99% up-time for their internet had their 1% failure that day. Maybe one team has all dps mains, and the other team is well balanced. MMR and predicted win percentage only has validity over many matches, not each individual match.

  3. Not every player tries hard every match. Sometimes this is subtle, like playing with a beer or two too many. Sometimes it is less subtle, like practicing a hero a player is not good at, regardless of team comp or map.

  4. Some players actively break the system, by either hard throwing, that is intentionally losing (as opposed to soft throwing as in item 3), playing on someone else’s account to boost it, or other similar actions. These actions are bannable and should be reported when seen. Blizzard has promised to take stronger actions against this sort of thing (17). However, soft throwers and hard throwers can be difficult to distinguish, and Blizzard has to error on the side of caution to avoid banning the wrong people, so it will always be a problem. The in-game report UI, as well as a key blue post (36), give guidelines on what behavior is or is not bannable.

  5. New accounts in general have less data and will have a less accurate MMR. There is no required minimum number of games to play, so new accounts can stay “new” for a very long time.

  6. MMR itself is only approximate. See the section on “How Accurate is SR”, below and realize that MMR generally will have the same issues as SR, with respect to accuracy, except it is harder to measure what is going on with MMR.

Many of these effects are removed or mitigated by the Looking For Group feature (or building your own team by some other means), so I recommend using Looking For Group if you are in a rank/region/platform/time/day with enough players for it to be viable.

Okay, stomps are going to happen, but why are there so many win/loss streaks and large SR/MMR movements? Shouldn’t a player’s SR/MMR be fairly stable once he gets to the proper rank?

Win probability changes slowly with rank because there are so many random factors in each individual match. Unfortunately, it follows from this that frequent and long streaks will occur, and a player’s rank will oscillate widely. Essentially, a player will tend to bounce between the range of where he is nearly guaranteed to win and where he is nearly guaranteed to lose. The range varies from player to player, but +/- 250 SR/MMR is common and +/- 500 is possible. This problem can be analyzed in depth, mathematically (34).

Skill Rating

Skill Rating (SR) should be thought of as a visible and friendlier approximation of MMR, similar to tier icons. With the exception of diamond+ players who have decayed (see below), MMR and SR are closely linked (22). The SR/MMR convergence is implemented by having each match pull SR in the direction of MMR. That is, if SR is lower than MMR, a player will win more SR than MMR on a win, and lose less SR than MMR on a loss. When I refer to an SR buff, this is how the buff is implemented (14).

Typical SR movement is 20-30. Typical MMR movement is difficult to quantify as it is invisible, but is likely similar. If an account is inactive for a while, after the player returns SR/MMR gains will be exaggerated again for a time (48). For this reason, players that only do placements each season will tend to see high volatility in their rating each season.

In addition to Elo type effects for high rated players (in which they gain less SR/MMR because there are no fair matches available), there is an extra SR debuff for high ranked players to prevent them from getting to the SR limit of 5000. The concern here is that players would start to pile up at the limit of 5000 and break the leaderboard (35, 48).

New accounts

If a player has never played competitive, his initial competitive SR is around 2350 (38). For new players, MMR and SR are more volatile (35) and SR has been measured (for average players) to approximately follow the formula: SR change = -3.85 * game + 99.34 (38) until about game 18. Top players gain more than this, and have increased volatility for 25 games or so (38). Bottom players assumably gain less, but no one has tested and shared the results. The maximum possible range of placements is roughly high bronze to low master, but most accounts tend to be artificially close to high gold. In addition, because initial placements depend on only 10-25 games, initial placements have substantial random causes and are frequently inaccurate.

If MMR are SR are so similar, why have both? Isn’t that unnecessarily confusing? Can’t we handle the problem of rank camping (see below) some other way?

In a word, yes. Here I’m going to engage in editorializing and reading between the lines of the blue posts. See this as my informed, but unverified, opinion. The real purpose of MMR is to provide Blizzard’s absolute best estimate of each player’s ability. However, since it is invisible to the players, Blizzard can change how it is calculated at any time and can have movements or changes that feel bad (23). SR, on the other hand, is player facing, and subject to player psychology. Aside from players’ tendency to over-react to every little thing, blizzard uses SR to influence player behavior (SR buff at the beginning of the season (now removed), SR penalty for leaving, SR penalty for being inactive) and to coddle players’ tender egos (23).

The good news is that as Blizzard is getting more confidence in the system, goofiness like this is slowly getting removed. Streak bonuses were removed. The knock down / build up in SR at the beginning of the season was removed. Performance modifiers are being dialed down. There may be some day in the future when Blizzard has enough confidence in MMR to make it public, and to abolish SR. I propose that we deal with the problem of rank camping by having tier decay (top 500 -> grandmaster -> master -> diamond) but not SR/MMR decay.

But I just read this post from a developer, and it said matchmaking is based on SR, not MMR

I’ve already referenced all the posts indicating that indicate that matchmaking is done by MMR alone (3, 21, 25, 40). However, there are also four posts that indicate that matchmaking is either based on SR or matchmaking is limited to a given SR range (5, 7, 35, 48). Scott Mercer is the author of two of the yes SR posts, as well as two of the no SR posts, so he contradicts himself. The contradiction must be resolved somehow, and I decide in favor of SR being used in no way during matchmaking for the following reasons:

  1. Whenever MMR and SR are discussed in the same post (including by Scott), it is to say that SR is not used in matchmaking.
  2. Master+ players regularly see decayed “diamond” players in their matches. And when someone like Seagull decays down to diamond on stream, he is still placed in grandmaster / top 500 matches. If SR was either used for matching, or the limits on matching were restricted by SR, this should not happen.
  3. SR and MMR are closely linked (22), except for decayed players (who only exist in diamond+) so saying matching is done on one or the other is a distinction without much of a difference. This makes Scott being sloppy with the terms not as surprising and makes the answer to this question less important.
  4. There hasn’t been any clear revoking of the old system on this point so the belief that we should only keep the newest non-contradictory statements (35, 48) is questionable.
  5. Generally the people who are insistent that SR and MMR are both used in matchmaking are using it to infer some broken or rigged system with horrible and implausible outcomes that are contradicted by careful analysis, common sense, and unrelated blue posts.

An alternate (but acceptable) hypothesis, in which I completely believe new posts (and declare obsolete inconvenient old posts) would be the following: MMR does not exist anymore. Posts describing MMR in competitive are obsolete. Match making is based on SR alone (48), except in the case of decayed players, where there is a specific hack in the code to use their “internal skill rating” instead of their “displayed skill rating” in match making. I’m not ready to change my whole post to support this hypothesis, as it hasn’t been explicitly endorsed by Blizzard, though it would be the correct conclusion if you ignore posts from August 2017 or earlier.

What would not be a reasonable conclusion is that matchmaking is based on some amalgamation of SR and MMR that ends up rigging matches or breaking the system. Regardless of what we call the number used to make matches, it goes up when you win, down when you lose, and is a single number.

I have tried to ask Scott for clarity on this issue, but have not gotten through the noise.

Rank Decay

To prevent rank camping, for players ranked 3000 SR or higher, their rank will decay 25 SR per day if they do not play. Each game a player plays increases his buffer by 36 hours, to a maximum of one week (20). Each day a player’s decay buffer decreases by 24 hours. If it hits zero, decay starts. To determine how many games must be played per week to avoid decay, we can calculate (1 game / 36 hours) * (168 hours / week) = 4.67 games / week. This is slightly lower than the 5 games per week that was originally reported (15). To see if decay is imminent look on the right hand side of the information screen of the competitive play card. After returning from decay, the player will have a substantial SR buff (gain more SR from wins than he loses from loses) until he is back where he was (3, 48). While decayed, a player’s MMR (and hence matchmaking) does not change (3). The decay clock doesn’t start until the player does his placement matches.

Performance Modifier

In platinum and below, SR/MMR gains are adjusted up or down based on the performance of the player. This is a minor factor (2, 48). This is done based on a numerical comparison of measureable quantities such as elims, deaths, assists, damage blocks, ults cast, etc. between a given player and other players of that hero at that MMR (8). Generally it is assumed that the measured stats are those visible to the player, but that has never been confirmed by Blizzard. Most of the details of this implementation are fuzzy and not published (probably to reduce exploits). This performance measure is correlated but not identical to “on fire” calculations. “On fire” compares a player to his teammates, while SR/MMR bonuses compare a player to other players that are not in the current match, but in a similar skill bracket, and playing the same hero (8). The effect of the performance modifier is generally small, but there are plausible reports of it causing people to have to maintain a 55%+ win rate to maintain their SR (11).

This is the most controversial (among many controversies) part of the SR system. There are two camps, those who don’t want their SR/MMR to be so heavily influenced by those in their group, and those who worry that having a performance based system will cause people to not play the objectives / win conditions and instead go stat hunting. In addition to people who go stat hunting, people can unintentionally be at the wrong rank because their good / bad play is not reflected in their stats. In the early days, these groups were roughly equal, but lately those advocating for no performance modifier are more numerous.

Starting in season 8 (January 2018), the performance modifier was removed for diamond and above (35). Blizzard has started to realize that having a motivation other than winning causes all sorts of non-ideal behavior and effects. The immediate forum response has been to request this change for lower tiers as well (at least whatever tier the poster happens to be in). However performance modifiers do help move new players where they belong much faster than an Elo type system, so I propose that the performance modifier be removed for Bronze to Platinum players after a sufficient number of competitive games has been played. Blizzard can use their data to determine how long it takes for people for their rank to stabilize, which would be a good place for a cutoff.

Season Transitions

Players’ 10 placement matches will start with Blizzard’s best estimate of each player’s MMR for matchmaking purposes (15). A player’s initial SR will generally be close to his previous season ending SR, with a relatively ordinary adjustment based on 10 placement matches.

However, occasionally a player will have a large movement (24) in either direction. This happens on relatively inactive accounts (39), and can be explained by Scott’s statement: “Play a lot of games, it (MMR) gets more certain. Don’t play Overwatch for a while, it gets less certain. The more certain the matchmaker is about your MMR, the less your MMR will change in either direction based on a win or loss” (2, 48). Essentially, if a player only does his placements, or slightly more than that, in a season, his MMR/SR is prone to much wider swings when he returns.

Because of reasonably high quality data for seasons 6, 7, 8, and 9 (24), it is possible to derive a formula: (New Season Starting SR) = (-173.7 +/- 20.8) + (1.0015 +/- .0059) * (Previous season Ending SR) + (37.94 +/- 2.23) * (Placement Wins). New Season Starting SR has a standard error of 181 SR (24), which implies a possible change in rating during placements of +/- 500 SR for inactive accounts. For active accounts, the max range is more like +/- 200 (from winning or losing all matches).

Decay does not persist through season transitions. That is, if a player decays from 4000 to 3000 before the season break, he will typically place around 4000 after the season break. Of course, if he is an inactive player (which decayed players often are), he may see an unexpectedly large swing in either direction.

Prior to season 6 (September 2017), players’ SR would be bumped down below their MMR at the beginning of the season, and they would earn it back with an SR buff over the course of 50 matches. This felt bad and was removed (15).

How Accurate is SR?

On the forums there are generally two factions with respect to SR, those who think that SR is essentially luck and a meaningless value, and those who believe that SR is essentially correct. Neither of these factions have the correct approach. The correct question is to ask how accurate is SR? Clearly top 500 players are better than bottom 500 (see any bronze to GM series or watch low bronze play vs top play), so SR is not completely random, but how accurate is it really? For this discussion, I am assuming that we are talking about a player that plays to win every game, doesn’t share his account, and has played at least 100 competitive games, and continues to be an active player.

There are number of ways to approach this question. One is to start a completely new account, and then play 100 games on the new account and see how it performs compared to the old one. This shows that SR can vary by 1000 SR in extreme cases, and 500 in normal cases (27). There is some evidence that reroll experiments show less variance at higher ranks (28) likely because there are less random variables, such as smurfs, throwers, and inconsistent play.

Next, any player can see how his SR changes during a season. A range of 500 is completely normal here.

In addition, if an account needs a 55% win rate to maintain SR (11), and if win % changes slowly with rank, then it is expected that this will be an additional source of significant error in SR.

If a player gets noticeably better, it can take a long time to get to the correct rank. If a player maintains a 55% win rate, he will only go up approximately 220 SR/MMR every 100 games. Because there are twelve players in a match, one player’s contribution (and ability to carry the match) is limited unless he is playing at a vastly different skill level, so a 55% win rate for a player that is moderately under placed is to be expected.

Put this all together, and we can state that an active and motivated player’s SR is only accurate to +/- 250 SR in normal circumstances and +/- 500 SR in extreme cases. Of course, new players, barely active players, or players that actively break the system can be off by much more.

MMR/SR Reset

Periodically, there are calls on the forums for all MMR/SR information to be wiped and to start over. The justification here is that because the system has had many flaws, and that MMR/SR takes a very large number of games to move, and that people are misplaced by 250 to 500 SR (see previous section), that it would be good to wipe everything and start over. The chief problem with this is that before the wipe, the typical error is +/- 250 SR, with +/- 500 SR in extreme cases. Immediately after the wipe, the typical error would be +/- 2500, as everyone would have the same rating. Top 500 would be playing against bottom 500. This would cause matchmaking and rating quality to fall through the floor until a sufficient number of games are played. And because many people do not play a huge number of games, or play inconsistently, this inaccuracy would persist for months. And at the end of that, the accuracy would be similar to what it was before the reset, because the underlying system has not changed. My reasoning against a reset is similar to that given by Scott (12).

Popular Myths

There is a lot of repetition in these subsections. My apologies, but the various myths tend to have a lot of overlap. They also have incompatibilities, so it is difficult to address them except individually, with repetition.

Forced loss streaks

There is an old and persistent conspiracy theory that Blizzard’s algorithms force players to have a 50% win rate by nefarious means (if a player is doing well, very poor players are put on his team to make him lose, often for many games in a row). This has been contradicted by Blizzard (7, 46), is contradicted by people’s ability to climb (13), and would be a horrible and difficult-to-implement design. The truth is much simpler. If a player wins more than he loses, his SR/MMR goes up. As it rises, he is placed against stronger opponents (and with stronger allies), which increases the chance that he will lose (7). Once in equilibrium, the average person he faces (and is allied with) is at his skill level, and the only way to go up is to become a better player. The win/loss patterns and streaks are fully consistent (in the mathematical sense) with the system as Blizzard describes it, and are not consistent with a system in which Blizzard forces wins and losses (34).

Matchmaking takes into account win percentage

It explicitly does not. “At no point in MMR calculations do we look at your win/loss ratio and win/loss ratio is never used to determine who to match you with or against. We are not trying to drive your win/loss percentage toward a certain number” – Jeff Kaplan (32). However, probability of winning depends the difference between MMR and a player’s skill level, and MMR does depend on how much a player has won or lost lately, so in this way there is an indirect dependence of win probability on win/loss history. This leads to (correct) statements that “Whenever I reach, or am near, my career high, I start losing and fall back down”. This corresponds to a player reaching his skill cap. The only way to break out of an MMR/SR range is to improve as a player and play a sufficient number of games.

Matchmaking takes people at a given SR, and then matches high skill players with low skill players to make a balanced match. Sometimes this is called handicapping, similar to amateur bowling or golf. This is done to achieve a 50% win percentage

This is not compatible with Blizzard’s statements: “All the system does when it comes to matching on skill is attempt to match you with people of a similar number” (32), and “We use MMR for matchmaking, not SR” (25) and “SR closely chases your MMR up and down and is a more “digestible” number. With the exception of top players who have decayed, MMR and SR are closely linked” (22). It is also contradicted by people’s ability to climb (13).

MMR is determined by statistical performance

The idea here is that a player’s MMR is really a summary of their statistical performance, and that because SR chases MMR (14), a player’s SR is basically determined by their statistical performance. If this were true, statistical performance would be, by far, the most important contribution to SR. This is contradicted by Blizzard (2) and by most players having a win percentage of about 50% (if statistical performance was weighted very heavily, stat-chasers would have much lower win percentage at a given SR than team-players, for example).

Matchmaking pushes a 50% win percentage using other broken criteria

It varies what this broken criteria is. The theories typically have a few things in common:

  1. Blizzard has never confirmed (and sometimes has denied) them.
  2. They would lead to obviously broken matchmaking.
  3. Their proponents never have solid data backing them up.
  4. They usually are not compatible with Blizzard statements, “All the system does when it comes to matching on skill is attempt to match you with people of a similar number” (32), and “We use MMR for matchmaking, not SR” (25).

The supposed support for these theories are that Blizzard has indicated that they desire and are happy with a 50% win percentage (32). However, there are many ways of accomplishing a 50% win percentage, and much better ways than the theories these forum writers ascribe to. Specifically, a 50% win percentage can be accomplished using a procedure similar to that which Blizzard describes (7): As a player wins matches, he is placed with and against stronger players. As he loses matches, he is placed with and against weaker players. With time, his win rate will converge to 50%, with some random oscillation around 50%. If he gets better (compared to the rest of the community), his win percentage will go up a bit until he finds his new level.

There is an SR penalty when grouping

This has been explicitly contradicted by Scott (48). If you take into account the entire system, there is a small SR advantage when grouping. This is because grouping increases your average win probability slightly as larger groups are unable to find equally sized groups and get matched with smaller groups. The system does not try to correct for this (49).

Blizzard is a company, and companies are inherently deceptive. See cigarette companies’ studies that indicate that cigarettes are good for you. Blue posts are not a reliable indicator of how the system works. My experience is proof that the system is malicious.

Yeah, well, you know, that’s just, like, your opinion, man.

More seriously, though, I take it as given that Blizzard employees don’t lie or intentionally mislead their customers. If you don’t agree, we don’t have much to discuss here. I can prove some things with data, but much of the system will always be opaque if we disregard all the blue posts. Even if Blizzard open sourced their matchmaking and ranking code, people could claim that the available code is not actually implemented in the game.

Win Streak Bonus / Loss Streak Penalty

Prior to season 5 (June 1st, 2017), there was a substantial bonus to SR for winning many games in a row (about 4 or more), scaling all the way up to 150 SR for one win. To keep things balanced, penalties existed if a player lost many games in a row. This was removed because it lead to people who got lucky/unlucky being thrown far from their true rank (9, 10).

Leaver Penalty

Each time a player leaves competitive matches any time before the Victory/Defeat screen, he will receive a 50 SR penalty. It is not known if leaving penalizes players’ MMR. In addition, a leaver will receive increasing automatic bans, with each leave (19):

10 minutes
30 minutes
2 hours
8 hours
24 hours
Season Ban (with a season ban, no rewards are received)

The season ban cannot be reversed. The reason or method of leaving is irrelevant (43). If a player plays many games without leaving, the leaving penalty resets downward. The rule of thumb is that if the bans are getting into 8 or 24 hours, a season ban is imminent. If a player receives three season bans, a permanent ban is likely (30). The season bans do not need to be consecutive to count toward a permanent ban.

If the player leaves before the game is 30 seconds old, the match is cancelled and no one except the leaver is penalized/rewarded. If a player leaves after the game is 30 seconds old, Players on the leaver’s team get no consideration due to the leaver. That is, they (and the other team) gain and lose SR/MMR as normal. This is to prevent situations where it is the team’s interest for one person to leave and save everyone else from losing SR/MMR (and the winners from gaining SR/MMR) (41, 42).

If a player leaves and rejoins the match (after rebooting / internet comes back, etc.), he will often be able to save himself from getting the penalty, but it is not a sure thing. He needs to rejoin quickly, and not disconnect and reconnect several times (43).

Leaving during or after the Victory/Defeat screen is safe, but it does not save time before joining the next match. The matchmaker waits for the previous match to be complete before attempting to place players in a new match (44).

Exiting the game while in a queue or a group is not safe. The game does not immediately remove you from queue and group when you exit, so a queue can pop (and a group leader can queue you up) before the game figures out you are gone and you can be penalized as a leaver. Always leave groups and queues and wait a minute before exiting the game.


Because of the complexity of the system, subtle bugs with skill-rating and match-making can be difficult to spot and reproduce by the players. And many player reports are not particularly trustworthy for various reasons. However, there are some outstanding issues which are well-known and verified.

There are a number of reports of a bug that will cause people to have very small gains / large losses in the match after the match with a leaver (26), assuming the bugged player left after the game said it was safe to do so, but before the end of the match. To avoid this, never leave a match, even after the game says it is safe to do so. There are also a number of people who a have reported small gains or large losses without any leavers in the vicinity, so we do not entirely understand what is going on. Blizzard has said that there is a fix in the works (33) but it has been a year since that post and there are still several complaints about this issue each season.

There is a rare and serious bug in which players can get incorrectly season banned and lose huge and undeserved amounts of SR. This bug has not happened recently. It appears that what happens is a competitive game gets put in a “Waiting for Players” state. And each time the game restarts with new players, everyone loses 50 SR and gets an increasing ban, up to hundreds of SR and a season ban. Blizzard has acknowledged the problem, is working on it, and has promised restoration to affected players (31). However, restoration can be slow and painful. If you ever see “Waiting for Players” during a competitive match, you should exit by any means necessary to stop the bleeding.

Using statistical measures to rate players is particularly prone to subtle bugs (11), which Blizzard and players have been discovering and Blizzard has been (slowly) acknowledging, fixing, and re-implementing since launch.


(2) The statement, “You go on a large win or loss streak, it gets less certain” is no longer valid, as win streak bonuses (and loss streak penalties) have been removed from the game (9, 10). Grouping synergies are also no longer taken into account when calculating win likelihood (49).
(4) “In Overwatch, whether your MMR goes up or down is contingent on winning or losing. But there are a number of factors that determine how much that rating goes up or down.”
(6) Competitive Mode Tier Distribution
(10) Stevo, a twitch streamer and Symmetra main did a bronze to gm series on twitch after these changes went into effect, and there was no detectable SR bonus, even though he won 51 matches in a row. Analysis at
(14) “When you do come back and actively play matches, you’ll also typically gain more SR from a win until your displayed skill rating and internal matchmaking have again reached ‘equilibrium’ “.
(25) Note however, that the second sentence, “Also players’ displayed icon …” is no longer valid. Players’ icons now change as a player loses SR.
(38) Initial Competitive Skill Rating, Decrypted
(39) Season 9 Placements Analysis, High Variance Explained
(44) Leaving game when seeing defeat or victory
(45) The 50 games restriction is currently in place. Decay has changed since this post. See the section on Rank Decay, above.
(46) Forced lose streak
(48) Groups and Matchmaking in Overwatch
(49) Groups and Matchmaking in Overwatch Note that previously when groups were matched against smaller groups or solo players, the smaller groups or solo players would have higher SR to compensate to make up for the synergy bonus (2). This is no longer the case, and it is unclear when the synergy bonus was removed, though Your Overwatch noticed it in a video on May 21st, 2018 (52).
(50) GUIDE: Enable SMS Protect (required for PC Top 500)
(51) This is from Seagull’s interview with Scott Mercer

Last edited on 8/31/2018


As always, thank you for your continuous research and investigation into the mysterious realm of matchmaking ratings.


One common concern with how people view the MMR and SR relationship is what it means to have a “50/50” matchup or 50% win-rate. This is especially concerning because it’s obvious to everyone that you can’t gain or lose SR if you gain the same amount you lose and you lose as much as you win.

In one of the threads I was asked this question and I spent a lot of time explaining it. Kaawumba goes into everything I say here, it’s not “new”, but it’s a more narrative description of what goes on behind the scenes and how it affects the player. Keep in mind it was written before LFG and any reference to LFG is “potential” rather than “actual”.

It's also a long post. Click here to see it.

First, if your win-rate hovers around 50% you shouldn’t expect to change ranks, right? Even if you just put 2 random people together, in any game, if they both won about half the time you would say that they were evenly matched. One is not better than the other. If you are not better than your competition, you should not expect to increase in rank.

That’s not what you’re asking though. What you’re really asking is “If the MM can produce a 50/50 match, then how can it expect anyone to advance even if their skill is higher than their current rating would indicate.”

This is actually a really simple problem to solve.

The MM, as good as it is, assumes in ALL instances that it is wrong. Even if you were perfectly placed, if you win you gain SR. Why is that? Why would you gain if you are perfectly placed even 1 SR?

It’s because the win result shows that it is, in fact, wrong.

When the MM makes the match all it has as information is the MMR value. We tend to think that that MMR value is what the MM thinks we belong, but that’s not the right way to look at it.

That MMR value is a guess. Educated guess, but a guess nonetheless.

That guess needs to be validated. How do you validate this guess? You pit players against each other to find an actual result. So the MM makes a guess. Puts you in a match (it really doesn’t matter against who, I’ll get to that) and takes the result to see if the guess is valid…not CORRECT, VALID.

So, if you (or me) were to be put in a game against NYXL that game would indicate that we were not as good as NYXL. If we kept repeating that match we would be even more confident that we were not as good as NYXL. Each and every drumming by NYXL would validate that, yes, we are not as good as NYXL.

Of course, in that example it would be pitting a team where you estimated had a skill of around 1600 with a team where you estimated had a skill around 4500. The 4500 winning against the 1600 does validate your guess, that 4500 is better than 1600, but that’s not really saying much.

What you want to do is pit two players that you think are the same skill against each other to show who is better. That way, you find out where you are wrong and in which direction.

If you “think” two teams have an equal chance of winning but the final result is that Team A is better than Team B, then your next guess will be that Team A will win again. You no longer think that the result is 50%. You think that it’s 60%, or 51%, or 80%, depending on how you set up the system.

If that same set plays again you can gain information, but that’s not how OW is set up.

OW is set up so that when you VALIDATE that result, when you WIN that estimated 50% match, it updates what it thinks of you. Now, it thinks you’re better than 1600. Maybe it now thinks you’re 1650, maybe 1700, but in our system for most people it appears that it guesses next around 1625.

So instead of getting put against a 1600 team, you are now put against a 1625…STILL A 50% CHANCE TO WIN, JUST AGAINST A BETTER OPPONENT THIS TIME. If you lose, you go back to 1600. If you win you go up to 1650. The cycle continues, giving you 50% matches, until you can’t win anymore. Of course, your win-rate will be above 50% while you’re climbing, but if it weren’t why would you expect to climb. Each validation shows more and more that it’s about right.

Now, say you lost a game that you were correctly ranked at. It happens, in fact it would happen about half the time. You lose again, against a (slightly) worse opponent…you lose again, worse opponent. It happens, because now maybe you have a 60% chance to win. Lose, lose, lose, lose, lose…ASSUMING NO BAD ACTORS this is perfectly plausible…down to the point where you basically have 100% chance to win. Like, 1 v 6 for your entire team. IF YOU TRULY BELONG AT THE RANK YOU WERE…your chances of winning will be increased and you will see a higher win rate up to the point that you are correctly ranked and you are at 50% again. THE MM THINKS IT’S MAKING 50% MATCHES, but you know better, these games are easy and these poor souls don’t stand a chance. Remember, that’s fine, the MMR always assumes that it is wrong.

Now, you’re probably thinking 2 things. 1) This is not what I experience. According to this, I should win all the way up to my rank, then WLWLWLWLWL around my true rank. 2) This is not what they are saying, they are saying that they are happy that people have a 50% win percentage, not simply that they put equal skill sets together.

1_ If you truly are only 25 SR below your skill, you won’t win that game 100% of the time, of course. Your EXPECTED result is 50%, but if we simulated that game over and over indefinitely your ACTUAL result would be, say, 51% BECAUSE THE GUESS IS INCORRECT. You can’t win 51% of a game, but you sure can lose 49 out of every 100 games played. The actual order of wins and losses in that scenario won’t necessarily be WLWLWLWLWL, either. You will get some streaks in there because streaks happen, just flip a coin 100 times to prove it to yourself.

2_ Say you wanted to study some natural phenomenon to see how it worked. One way you could test your idea of how it worked is to create a model. We (humans) do this all the time. In fact, any “theory” in the scientific sense is a model. E=mc^2 is a model. The algorithm that determines who will win a game of Overwatch is a model. You can study the phenomenon, think about how it probably works, build a model that expresses what you believe is true.

How would you validate that model? How would you know it’s correct? You put data into the model and see if the model matches the actual result.

So when we initiate nuclear fission, we take the known mass and take the released energy and see if E does in fact equal mass times the speed of light squared.

It’s the same with the OW MM algorithm. They take the players, guess their skill, and need SOME WAY to validate if their guess was correct.

If they guessed two people were the same skill and they tested it to see if their guess was correct what do you think the result would be?

50%. If their model of how good players are were correct, they would expect a 50% win-rate. They find a 50% win rate, indicating that their model is correct.

That’s part of the reason this “forced 50%” stuff is annoying.

They want a 50% win-rate to validate if their system works, not for some nefarious purpose. Testing mathematical models isn’t something the average person does everyday, so I get the confusion, but this is a process of guess, collect data, adjust, guess, collect data, adjust…not “Lhun is getting a bit above his rate, lets put a leaver on his team.”

This is how the MM can be good (overall 50% win-rate) and also allow people to climb (games are individual validations of SR).

If you want to discuss PBSR, that’s fine. I won’t participate because I really have no strong feelings either way on the issue. I can understand both sides, I don’t believe it’s a significant effect, but I’m not very certain of that belief at all.

Yep. The way they handle groups sucks. It’s really an impossible problem to solve. First, the above system works best when everyone is of the exact same skill. That’s not really realistic but as long as it’s close it’s fine. But you get 1000 SR differences in there? No. It’s that scenario where we were playing against the NYXL, you’re just proving what you already know.

Second, there is a bonus (apparently) for grouping to account for the “synergistic” effects of being in a group. Synergistic effects that don’t apply in the same quantities over all groups. Edit: This appeared to be quite clear when I wrote it, but has since clearly been either changed or clarified as never being the case.

In short, they know groups are screwey. It’s one of the hotter topics out of their own mouths. Unless you have a good idea on how to stop groups from screwing with the MM there isn’t much else to be said.

Maybe I don’t understand what you’re saying, because the only people that will have “nobody to beat” would be the top 1…not 1%, not 100…top 1. Even they have to constantly be beating their competitors.

The reason there are so many conspiracy theories about the MMR is because Blizzard refuses to give us concrete answers and refuses to explain the exact mechanics under the guise “it’s to complicated to go into”. I think it’s doing things we wouldn’t like if we knew about it.


If you want more detail try this:

It’s 57 pages of math of MMR type rating systems.

That doesn’t explain how this MMR system works though. I’ve watched videos on the topic and I understand it’s intent is to create matches where there is an equal chance of failure and success, however there also seems to be something in Overwatch’s MMR that causes losing streaks and winning streaks, based on some hidden value. I just know too many people that will climb only to end up where they were at at the beginning of the season. My personal opinion is that after your first placement in competition you shouldn’t have MMR factoring your SR.

I didn’t read it. I’m saying I’m not a complete novice when it comes to MMRs and that Blizzards is a little hanky in respects to how it measures skill levels and SR. And I’m also saying people have so many conspiracy theories about it, because Blizzard refuses to release how it actually works under the guise people will manipulate it, but my question is, why bother, since they already seem to be with so many smurf accounts, or is it just meant to keep low level players from accidentally hitting higher levels, not the other way around?

It’s not as hidden as you’re led to believe by the conspiracy theorists. Smurfs do mess with the system, but that’s not really the system’s fault, is it?

It’s meant to keep people from overranking through manipulation of the system. Smurfs are underranked, different issue.

Though honestly all that anyone needs to know is that you need to win games to move up. It’s a simple message that is unsatisfying because people have unreasonable expectations of the system. Most things people don’t like are trade offs with other things they REALLY don’t like.

Do you not understand why or how streaks occur? Tilting and other forms of lack of self-awareness (fatigue, etc.) can easily drop you to a point where the next day you are very underranked. Unintentionally Smurfing, if you will. There are other ways, lots, actually, but streaks aren’t strange. Frustrating, but normal.

I don’t have any restrictions on what you post in the thread. I’ll take whatever bumps I get. I’d suggest you label it better, as it appears that you are posting a FAQ, but the contents is hidden in the section arrow and it is addressed to me, rather than everybody.

I won’t usually respond to your posts, because I usually agree with them.

However, this is obsolete:

Any discussion of groups in Overwatch should rely on Groups and Matchmaking in Overwatch, explicitly or implicitly.

1 Like

It would be better if Blizzard communicated more clearly and completely about the system. However, all of the conspiracy theories directly contradict public Blizzard statements. See the “popular myths” section in my original post, for references. Therefore, there is no expectation that more and better communication on Blizzard’s part will convince any conspiracy theorists.


Win probability changes slowly with rank because there are so many random factors in each individual match. Unfortunately, it follows from this that frequent and long streaks will occur, and a player’s rank will oscillate widely. Essentially, a player will tend to bounce between the range of where he is nearly guaranteed to win and where he is nearly guaranteed to lose. The range varies from player to player, but +/- 250 SR/MMR is common and +/- 500 is possible. This problem can be analyzed in depth, mathematically (See

This has nothing to do with Overwatch’s MMR. Any Elo/ladder type system will have this sort of behavior. The only thing that varies is how much the rating bounces around 50% for any particular player and game.

1 Like

SR manipulators cannot be dealt with with a ranking system. They have to be dealt with with a suspension and banning system.

You found the punishing normal player system, you gonna be banned soon

To explain how absolutely normal and common this is, I’d like to link to a tweet from Josh Menke, who used to work in matchmaking and ranking for Blizzard, though I don’t think for OW.

It’s a picture of a sock. With very swingy “graphs” on them. (I think it’s just some random sock design.)

He makes a joke about wearing his “skill graphs” on his feet.

The experts, the guy who has a Ph.D. and has designed and worked with many of these systems for Blizzard and other companies, understands quite well that oscillations are inevitable in these systems. So well that he sees them in socks.

It’s normal. Not even normal, inevitable. There’s no known way to make it not look like that.

1 Like

Better answer, Google “This is how sad Overwatch is”

You’re welcome and have fun actually ranking up!

You have a rather tremendous amount of confidence in a rather silly theory.

Actually you are the one the tremendous amount of confidence in a rather silly and wrong theory, Kaawumba!

1 Like

Well you can’t read something and understand it at the same time so I don’t think my opinion can be less silly than your.

Yeah I totally agree with that video.