Guest Column: On the Overwatch Matchmaker as a Trolley Problem

It's sometime before nine AM. I'm hoping to get one quick game of Overwatch in before starting work. I launch the game, get on Quick Play, and the game has me backfill a team. I choose the best class for the team composition, and the game ends in defeat for my team before character models have finished loading.

It's sometime in the mid-afternoon. I'm due to leave in twenty minutes, but I figure I have time to play some Overwatch before that. I get on a quick play match, pick Lúcio, get on the payload, and stay there. We never stop moving until the match ends a handful of minutes later. I quit Overwatch to do something else while I wait, since I don't have time for a second game.

Games, as design objects, have failure modes. You can sit down to play a game and have a bad experience that was never intended by the designers. Motion sickness in first-person games, unbalanced multiplayer matches, and so on.

Bad matches don't require a matchmaker. I've played enough Team Fortress 2 to encounter numerous matches that were grossly unbalanced. But, of course, there's something particularly infuriating about the matchmaker's role in this. You didn't choose a server with bad team balance; that server was chosen for you. At one point, one can start to feel like what you do in a match is beside the point; the outcome was decided by the algorithm ahead of time.

Blizzard has grappled with matchmaking for a long time now, starting with PvP gameplay in World of Warcraft. If Overwatch sometimes feels a little bit clumsy, with its too-complex level design and its too-ambitious class structure, one is often tempted to write this off as the result of a studio with vast resources and expertise, but who is new to the genre. But one would think that matchmaking systems are something that Blizzard has to have down; something that they certainly have done before.

And yet, Overwatch matchmaking is a mess at times. The truth of it is that matchmaking is a hard problem, even more so in a game like Overwatch. Optimal results are never going to be guaranteed. But Overwatch is also the latest in a long run of incremental improvements to matchmaking.

In the beginning, we had self-selection; players would pick their own opponents. In Chess, the grandfather of competitive gaming, this was mostly a reputational process. In the server lobbies of Starcraft, this was performed through some combination of trash talking ability and pure randomness. In games with permanent dedicated servers, like Team Fortress 2, players would not so much pick opponents as a server to play on, with individual servers eventually finding some equilibrium of regulars that knew one another.

And then, we had Elo. Invented in the mid-20th century as a way of rating Chess players, Elo is a simple statistical model that can be calculated by hand. The key assumption is that if you play against a higher rated player and win, your rating should go up because your “true” rating is probably higher, while the other player’s rating should go down, since they are probably overrated. This sounds sensible, but it’s based on the assumption that the winning player performed at a higher level than the losing player. That’s true of Chess, but doesn’t translate to most digital games that have hidden information, variance, asymmetry, and nonlinear player skill.

Elo as statistically sound enough to tell us that two players with similar ratings should have a doubtful outcome; it told us who the underdog was, and it let us arrange good matches. Elo also had many problems: it was a transparent system that could be gamed by savvy players who turned its assumptions on their head. It encouraged not playing the game to "protect one's rating.” It was widely misapplied to rate players in games that are not at all like Chess; and in those games, it produced results that were often quite bad.

So, developers of multiplayer games made a bit of an evil pact. Elo, flawed though it is, is a product of a time when computer programming involved physically switching cables around. Anyone with a calculator and some know-how can calculate Elo ratings. But its inadequacy has driven video game developers towards using increasingly opaque methods that are usually proprietary, complex, and subject to constant adjustment. Where before we had a simple equation, we now have algorithms.

Here's the thing, though: Matchmaking systems, strangely enough, are an ethical issue. Because when they fail, they create a bad experience for someone, sure. But more than that, matchmaking systems inevitably end up prioritizing one player's experience over another. Someone is always going to be on the bottom end of the skill gradient of a match.

At times, it's hard not to feel like the matchmaker is press-ganging you into being a jobber for someone else's fantasy of being a hero. And how do you, as a game designer, manage this? You can wash your hands of the issue and live with the fact that some players will experience frustrating runs of bad matches. Or you can try to compensate somehow and make sure players are not being placed as the underdog in matches too often; but when does that cross over into trying to decide matches yourself or micromanage player's win rates?

Ultimately, the Overwatch matchmaker is tasked with deciding whose fun is more important, and which standards of fun should apply. Back when the "avoid this player" feature was in place, Blizzard found that one of the world's top Widowmaker players was finding it hard to get a match; people were misusing this feature to steer clear of people who were too close to the skill ceiling.

But who is to say that someone who is so good at the game that they frustrate the players around them is entitled to getting a match in a timely fashion? I'm not saying that isn't the case, but the counterfactual idea that players have an expectation that they won't encounter pro-level quality play in pub games doesn't seem totally unreasonable. Those two preferences can't be satisfied simultaneously, so someone at Blizzard has to make a decision about which preference is more important. Do we care that the high-skill players can find matches in quick play more than we care about isolating players from frustrating experiences on the other side of that rifle?

We are not used to thinking of game design as zero-sum. We generally assume that we can only make the game better, we can only add enjoyment to it. But when it comes to multiplayer games, those situations arise where one player's enjoyment is another player's frustration. We don't have an ethical calculus that tells us that making one player wait for one extra minute is worse than making six players experience a game skewed by a single player operating near the skill ceiling of their class. And yet, an answer has to be provided one way or another.

Those are complex questions without direct answers. They're also not terribly important. Ultimately, the consequence of failure when designing a matchmaking system is that someone has a bad, frustrating time with your game. This is pretty good compared to the consequences of failure when designing, say, a car or a skyscraper. Nobody is going to die because Overwatch's matchmaker is sometimes stupid. It's a toy problem, literally, one where the stakes are defined in notional units of fun. But the wonderful thing about toy problems is that they're clarifying; that it's easier to talk them through when there's not the pressure of life or death hanging over the conversation.

Important or not, someone at Blizzard has to answer those questions. In the case of "avoid this player", they chose that individual high-skill players deserve to get a match quickly even if it's not ideal for the enjoyment of lower-rated players. But the important thing here is that decisions like this then get encoded into an algorithm. They become a policy that is implemented as software.

And this is why this is a conversation worth having: Because we live in a world where, increasingly, policy is implemented in the form of software. Someone at Uber makes decisions about how their drivers get compensated and how their users are charged, but that decision exists in the form of an algorithm. There is an entire strain of Silicon Valley salesmanship built around using algorithms to replace human decision-making.

Except at no point are human decision-makers ever replaced. Instead, they're put behind a curtain made out of software. Where before you had bosses cutting pay and jacking up prices, now you have an algorithm telling you the value of something. Algorithms are useful in many ways, but Silicon Valley has made an artform out of using them to sublimate responsibility. When an algorithm targets you with a sales pitch in a moment of vulnerability that it detected by trawling your data; when an algorithm slashes your compensation; when an algorithm decides that you are not worth prioritizing relative to some other user, to some other actor in the system, or simply relative to sheer profit: Who do you blame? Who do you get angry at? Where do you picket, how do you strike? The buck must eventually stop at a human; algorithms are not laws of nature, they are the product of programmers working under the direction of management. But Silicon Valley PR has given them an aura of infallibility, fairness, and impersonality. As though being hurt by a mathematical abstraction is supposed to hurt less.

And yet. And yet, when the Overwatch matchmaker wastes my time or frustrates me, somehow I am more, not less, angry than I would be if I had just joined TF2 server with bad team balance. Sometimes, it takes something that doesn’t really matter to crystallize the import of something that actually does. There’s an immediacy to it that doesn’t come across most of the time; the software systems that increasingly bind and direct our lives are soft, slow, invisible. They’re doing things very incrementally. The effect of all the advertising that gets served to our eyeballs, all the ways we are deprioritized, all the surge pricing, is aggregated over a matter of years. Often, we don’t see it at all; if a company uses algorithms to filter job candidates, you don’t really know that, and you don’t really know an algorithm sent your CV to the bit bucket.

But the matchmaker? The matchmaker expresses its preferences in a timescale of minutes. Sometimes seconds. Sometimes you know the algorithm fucked you before the character models load.

And so: If we can resent Blizzard for how they tune their matchmaker, then maybe we can resent Uber for how they pay their not-quite-employees. Maybe we can resent Facebook for their abuse of our personal lives to serve us ads. Whenever games act as a microcosm of broader changes in society, they act as an immediate barometer for our own feelings about those changes; and so, as online multiplayer moves more towards a matchmaker-based approach, examining what that means ethically becomes very important. Not so much because of how it affects your skill rating, or your play experience, or your feelings towards the game itself, but because of what it reveals about how we relate to algorithms and the control they exert.

Article headline is click-bait for Ethics nerds who like games.... rare! :D

Read this article twice and still don't understand the leap from the bad matchmaking to Uber fucking over its employees.

Wonderful article! It's really fascinating how this process works for each individual game. I'm sure we've all had some anecdotal evidence of this.

I personally just got done with season 2 qualifying matched and all 10 were clean sweeps where I went 5-5.

It's...It's...(long sigh) It's LITERALLY about ethics in Video Games.

I loved this article, thanks a lot for writing it.

It's a shame to hear about how Overwatch's matchmaking prioritised higher-skilled players though, considering they make up a far smaller proportion of the player-base and would probably train against each other anyway :/

@walreese55 said:

I feel like you may have skimmed over the last paragraph.

It was about how algorithms affect us in some way in our day to day lives and its apparent in applications outside of games and used Uber as an example.

Really interesting topic about how algorithms have the illusion of being unbiased.

It's the inherent randomness of matchmaking that's pushed me away from multiplayer games when I'm limited on time. Playing Overwatch over an hour or two and the good and bad matches even out. But for 20 minutes? I know that I'm risking a play session that leaves a bad taste in my mouth. Better to stick with a single player round of Magic Duels or get through a puzzle in Deus Ex Go for my sanity's sake.

Really interesting, thanks for writing...

Article seems to imply Warcraft III did not have matchmaking...

Outside of chess, tennis also has a rating system. In tennis, however, it's based upon a self-rating & match results. I wonder if a system like that (a combination of self-assessment and actual results) would ever be viable in video-gaming.

This article allowed me think about the match-making process in a way I wouldn't have otherwise.

Thats uhh thats quite the jump there at the end of the article

Remember the times where we didn't need matchmakers, only a server browser?

@foggen: And don't forget about Hearthstone! That came long before Overwatch, and is based entirely around match-made games.

I think the interesting thing about Overwatch is that even while acknowledging matchmaking is never going to be perfect, a lot of its systems are designed to make losing remain fun (or at least as fun as it can be to see DEFEAT appear on your screen). Characters have achievable ultimate moves so that most of the time you'll have at least some period of time where you feel powerful or do something awesome. Maps are balanced such that you'll have push and pull swings in matches where different teams have advantages - spawn advantage switches over the course of the match, defender advantage lasts until one team breaks through, capturing a control point is typically easier than defending it. Overtime allows for crazy comebacks and desperate swing situations. And matches have enough length to let those swings happen, while not being long enough to force you through a bad situation for an interminable length of time.

And for me at least its generally worked fairly well. Most matches of Overwatch I come out of wanting another match, even if I want to dive out and find another pub group, and I'm almost never discouraged to the point of wanting to quit or feeling unhappy with the game as a whole. Which is rarity in multiplayer games, at least for me, as I'd commonly bomb out other matchmaking-based games after having absolutely no fun for a few matches in a row. The systems can only go so far and a complete matchmaking failure where you get stomped and you and your team do nothing right is still un-fun to play, but at least for me that's not a common occurrence.

@freddiefiasco: I was specifically talking about where he said Blizzard started grappling with matchmaking in WoW PvP, which predates Hearthstone but not WCIII.

I don't know.... during season 1 of overwatch after playing a fair share of competative mode to get a solid rating, every game for me was a close battle. So I can't really relate to this article..... Maybe try competative mode if you haven't already?

Not everything about match making in OW is perfect, but so much of the actual game design and feedback loop has been designed around making losing less of a impact on the players ego that I feel being mismatched in skill is somewhat mitigated on the casual level. Highlights, cards and medals can make the sting of losing less.

When I reflect on my personal experience with OW the bad matches I have had are not something a better algorithm could solve. The human elements not logged by match making is what causes all my bad experiences. Toxic team member, annoying chatters, gimmick parties and team trolling behavior are all incalculable. They would all be mitigated by the "Avoid this player" function that was taken out in the interest of fairness to high skill players. Even then, I imagine players that exhibit this bad behavior do not do so for the majority of their play time.

The frustrating part is seeing how the game design has fed into a lot of this bad behavior. The feedback of medals ranks players during the match against their teammates based on damage/elims/objective/healing. These shift the discussion from "The other team was better" "we made mistakes" and "match making was bad" to "We have weak links, but its not me, you guys suck, kill yourself.".

Some people think the two distinct hoppers of Ranked and Quick Play has encouraged healthier behavior. They are wrong. Quick Play has been interpreted as two different things, and when players with opposing views are on the same team it is ugly. Some think QP is low stakes, low stress way to play OW and get more experience on characters you are not comfortable staking your ranking on yet. Some think QP is a mode where no one is interested in playing OW, being on an effective team, or trying even a little to actually win a match. You'll run into Mercys who only use their pistol, people who just walk around emoting instead of fighting, people who leave at the perfect time to swing the whole match, team killing symettras. All these people will claim "Dude, its quick play, settle down." after wasting around 10 minutes of 11 peoples time. Better matchmaking isn't going to do anything to protect players from matches ruined by this behavior, the stuff that actually upsets people.

Blizzard word filtering GG EZ has shown they actually acknowledge that shitty behaviour is effecting widespread player enjoyment, but it appears they are at a loss of what to do but poke fun at it. It's not an easy problem to solve.

It's is an interesting viewpoint to technology and ethics in general. Is technology used to obfuscate the responsibility of actions and decision by creating algorithms so people don't have as direct and clear view of the actual decision making.

For example, would people be so angry about Daraprim prices if they had said that they created a complicated (and black box) algorithm to analyze market prices and set new price according to the market instead of having Shkreli standing there and saying "this is what we think it's worth."? I'd guess that at least some would have accepted it more easily, because "hey, it's what market says". Of course this is an extreme example but I think the idea is worth thinking about.

Destiny has the same kind of dilemma going on. Being a P2P game, having parameters prioritizing skill level over ping meant better players were always playing the hardest laggy matches for a while there. It seems like the balance is back now but skill based matchmaking still generates some goofy consequences.

Consequences like trials (the PvP competitive mode) being the most casual, laid back mode for good players because its matchmaking only looks for good connections.

"If Overwatch sometimes feels a little bit clumsy, with its too-complex level design and its too-ambitious class structure."

Did people just skip this part? I can kind of see the class problem but "too-complex level design" seems incredibly dense. If anything, the current maps are far too simple and predictable to be interesting for very long and thus the classes HAVE to be overly ambitious to make up for it.

If OW didn't have extremely high class diversity, the simple maps would become boring and the game itself would be uninteresting within hours.

Definitely an interesting problem with few easy solutions, one I think everyone is aware of on some level but rarely gets articulated.

Blizzard found that one of the world's top Widowmaker players was finding it hard to get a match; people were misusing this feature to steer clear of people who were too close to the skill ceiling.

The first half of the sentence is a fact, the second half is an assumption predicated upon that fact. It's the same assumption Blizzard made, but the truth is that if you're being matched with that guy you're probably being matched with Seagull and Moonmoon and many others.

What's sad and hilarious is that it was never considered that the guy was being avoided not because of his skill, but because chooses to apply it with the game's sniper character. Everything about Overwatch is based around medium-range combat, with a few characters giving you bonuses if you can bridge that gap and get in close. Widowmaker is the only character that is effective at very long range, and her existence means the game's maps can't have large unobstructed areas or else she'll become too powerful taking everyone out at a distance from which can't be countered.

Widowmaker can't be anything better than a periodical semi-serious novelty pick or she turns certain players into demigods, with lifetime Counter-Strike players probably having the biggest edge over the crowd migrating from TF2, Battlefield, Quake Live, etc.

@r3dt1d3:In quick play, I often find that the maps seem to have side passages that have poor wayfinding, causing them to become underused. There's a decent number of weirdly large alcoves and cul-de-sacs in some maps, too, which seem to serve little purpose. I'm not enough of an expert on level design to comment on how it should be different, but it's clear to me that map complexity doesn't equate to gameplay longevity. People played de_dust without getting sick of it for a decade, and that map is so simple I can hold it in my memory and walk through it in my head even now.

Related to that, but more in the realm of environmental design rather than strictly level design, is stuff like badly placed props with collision boxes in this game. There's a camera mounted on a tripod in Hollywood that looks to all the world as though it's a physics prop, but it's actually map geometry and it creates a blind corner you can get stuck on it while moving close to the wall. There's stuff like the banners on the opening approach in Numbani which I cannot, still, tell whether they are a mistake or exist specifically because Pharah players had too easy a time overlooking that approach, but they seem like a clumsy solution to that problem if it's the latter. And there's stuff like the fact that the only way to know whether you can wallgrind a wall as Lúcio is to know it from having tried it; there's no consistent visual indication. I don't even disagree that the level design feels a little simple and repetitive in at times, but to me it also often feels clunky in a way that calls for streamlining.

Part of the repetition problem is that that the maps are very small and tend to funnel players (particularly uncoordinated QP players) into the same combat zones over and over again.

@psyael: This is an inherent problem with class-based shooters; some characters will invariably work better or worse across skill levels, and thus be hard to balance. Widowmaker's problem isn't even the long range, but rather the fact that she had the ability to get one-hit kills in a game where players get a very strong expectation that this isn't supposed to happen. Popping your head out of cover and immediately getting clipped by some guy across the entire map is the core experience of Counter-Strike, but it feels very out of place and seemingly "unfair" in Overwatch. Snipers also have the problem that you sometimes only encounter them when they're killing you, creating a perceptual bias towards thinking they're too powerful.

The part about the skilled Widowmaker player being cock-blocked by the wider community due to his / her skill (which is a bit of an assumption, but it's not unreasonable, so we'll go with it) is interesting, but on a general level, I think that speaks to a lot of players' distaste for being sniped. Even if they respect the skills on display, or enjoy partaking in sniping themselves from time to time, it can still be a rather bitter pill to swallow when you're repeatedly on the receiving end. Maybe I'm projecting slightly there, but when I see Widowmaker or Hanzo in games, there's usually a lot of salt being thrown around. While I don't partake in that, I do at least understand why it's so common.

For me at least, it simply boils right down to the fact that there is nothing fun about being killed by something you had no way of preventing. There are of course situations to circumvent this, depending on team set-up etc, but I'll ignore that for the purpose of argument. My general point is this: I'd happily be smoked over and over again by a skilled Winston, say, as opposed to a skilled Widowmaker. My patience for long distance insta-kills is exactly zero.

Yes, I'm (hypothetically) being very selfish by denying players their fun (or at least I would be if I avoided players, or heck, even still played the game!), but quite honestly, I couldn't give two shits. I'm playing a game to have fun, and being 360 no-scoped repeatedly is the absolute antithesis of that. If there's a way to cut that out of my experience, so I can have my own subjective idea of 'fun,' I'll happily do that and in such situations, I'd be pleased that algorithms can make that possible for me.

At the same time, I freely admit that I'm a filthy casual, and Overwatch (along with all competitive FPS) is not my kind of game, so think of the above what you will. But for my money, the game would be much better overall if sniper characters didn't exist.

Either way, really interesting piece. I'm not sure why the overall point towards the end is being lost on some.

On one hand, I miss dedicated servers as well.

On the other hand, even as the type of person who DOES take little examples and extrapolate them over larger societal issues via my own heavily flawed sense of pattern recognition, this seems like a stretch.

We live in a dystopian cyberpunk future where all our personal data and many of the details of our lives are designated in an automated fashion by computers owned by giant corporations. This seems like the smallest example.

All games on all platforms should have matchmaking as well as a customs server browser.

To say nothing of the gripes against Elo and the weird political swerve into some silicon valley App Economy stuff (I don't like Uber either, but man that's a stretch), I think the basic premise - that it isn't fun to be on the losing team in OW/that the winners have fun at the expense of the losers - is flawed.

Maybe I've been lucky, or maybe I just don't take losing as hard as the author seems to, but the majority of losing matches I've played in OW have still been fun, good matches. With the exception of Widowmaker's range and a few hero hard-counters, OW's mechanics are very oriented towards always having potential, being able to do work and impact a match, and generally being able to play effectively with your class even if your team doesn't come out on top in the end.

Granted, you can end up on a team so bad that they get stonewalled into the spawnpoint (depending heavily on the map in addition to team matchmaking balance), and those matches are legitimately not worth playing. But every player has a counter to that - there's no penalty for leaving! You can jump out and join something else on the relatively rare occasion that matchmaking deals you a game so lopsided that you basically can't even play it.

Oh man, your analysis of algorithm-based business decision making is Spot On (and completely terrifying!), but I'm having a hard time seeing how Overwatch's Quick Play Match-Making meaningfully resembles a trolley problem. The purpose of Quick Play isn't to provide close & balanced matches, it's to get people in and out of Overwatch matches as quickly and easily as possible. Whatever hidden Match Making Rating (MMR) they use to organize Quick Play teams plays far faster and looser with their numbers than the game mode that's purposefully built around pairing you with players of your skill level. Even if we assume that we have some sort of moral obligation towards Maximizing Net Fun Units across the Overwatch Playerbase (and ah geez, what a bunch of assumptions that would take), playing the Ranked Mode should offset the "Hero or Jobber" effect you describe in the article above given that a large enough sample size of games played in a carefully managed Ranked Environment should give every individual in the playerbase the entire gamut of Fun. With very few exceptions at the very top and very bottom of the bell-curve, everyone will stomp and everyone will be stomped from time to time.

Interesting article, all the more because this specific issue is something i have talked with a friend about in the pub just yesterday. There is no easy answer, except that its certainly not perfect right now.

Oof, I'm real tired today, so bear with my double-posting. I suppose the "with very few exceptions at the very top and very bottom of the bell-curve, everyone will stomp and everyone will be stomped from time to time" from my last post is where we get into our Trolley Problem, which is classically meant to suss out whether our ethical intuitions are Consequentialist (based on results) or Deontic (based on fundamental principles). Capitalism & tech culture are inevitably going to lean towards the Consequentialist end of the spectrum, given their primary focus on maximizing profits & growth. Creating an algorithm to place every Overwatch player into a nice little skill-based bell curve is pretty innocuous, but I think it CAN helpfully illustrate an issue with tech culture and late capitalism's version of Consequentialism. Overwatch fits every player into a bell-curve, with only maybe 10% of the playerbase consistently suffering from Actually Poor match-made gameplay experiences. One of the many problems with late capitalism is that it (intentionally!) doesn't accurately take into account the set of people involved in their system, because if it were to do so the benefits of their Algorithms wouldn't fall so squarely upon the Tech Bros & Upper-Middle Class at the expense of the far greater number of people in the set of the global working poor. So first, we have to ask if Consequentialism is worth it: is it worth essentially sacrificing 10% of our population so that everyone else can benefit? Then, we have to ask if, since we've essentially already said "Yeah, Sure" to that first question given that the dominant global ideology is currently Capitalism, whether or not the system is actually working in the way we all wanted it to?

In conclusion: don't be afraid to play support if your team needs it and Eat The Rich.

I don't leave a lot of comments. None of them are useful, or insightful, or even well written. This will be no exception. It might be read by no one; it certainly won't be cared about by anyone that does. I feel compelled, however to say, to anyone involved with picking guest columnists: THIS IS THE FUCKING GUY. God dammit this was such a good piece.

This article seems to assume that algorithmics hasn't improved since the 1940's, I can safely say that system used to match players in Overwatch is a lot more complicated than Elo. Then the swerve into computerised business practices seems to misunderstand the role of algorithms in that sector too, often times (especially in areas such as high frequency trading) there is no human input, no 'man behind the curtain'.

MMR is definitely a poor system for games and Blizzard just seems to have slapped it on literally everything, it made PvP boring and formulaic in WoW (in contrast to something like DAoC or Warhammer), it makes netdecking (and to a lesser extent aggro/tempo) the only eventuality in Hearthstone (since legend players are the ones people copy), and it produces various other issues in their other games. The problem is if you're any good at video games then having a 50% win rate is not going to be satisfactory for you and for 90-99% of players that's just how it shakes out, even by design; this is probably less of an issue in Overwatch but I'm sure there's some very competent players that continuously get matched up with shitty compositions to the point of agonizing frustration.

This is one of those things that Blizzard fanboys will not let go no matter how fallacious their argument winds up being so definitely an almost impenetrable topic in terms of Blizzard fixing it, much like the legend grind in HS (which doesn't run based off of MMR exclusively but ultimately has a similar effect).

I really don't mind the matchmaking system in Overwatch. I'd say my only gripe about it is that it will match me with people way below my MMR to stop winstreaks. The system definitely tries to force a 50/50 win-lose ratio. I really don't have an opinion about their other Multiplayer games because I haven't been subbed to WoW since BC and I've been in one online starcraft match not including the custom games :D

Gosh, I really enjoyed this article. All of these guest articles have been good or even great, but this one really tickled my fancy.

Bravo.

Glad you made this article. In the past few months, but most recently since the release of Season 2, Blizzard has been getting flak from fans about their matchmaking. Whether it's about the always talked about forced 50% win ratio, the premades matching up with solo queuers, the SR grind that has you lose more point than gain (unless you're a pro) or just the often unfair team match ups. Something is definitely up with Overwatch's matchmaking and Blizzard's idea of adding tiers now is nothing more than a bandage over a festering wound.

@tehpickle: I get you on not liking snipers. I have been feeling that a lot during the Battlefield 1 beta. I want to be out riding horses and manning machine gun nests and planting anti tank mines and doing bayonet charges, but asshole snipers keep one shoting me from rocks halfway across the map. I wish there were options for sniper free servers or just a complete removal of snipers from the game. The same thing applies to people in Battlefield 4 who spend the entire round in jets dropping bombs on people to pad their k/d ratio. I never had a problem with snipers or air vehicles in stuff like Halo where there was a limited supply, clear counters and you got ahold of them primarily through good map control, but the modern trend of letting people spawn in with snipers and spend the entire game camping and getting one shot kills or running for a jet and spending the whole game dropping infinitely respawning bombs on unsuspecting players is incredibly frustrating. Instant death without warning or the ability to counter it is generally a bad game mechanic and I wish developers catered less to people who play games that way.

I do sometimes worry about the sheer extent a lot of companies refer to or even rely on algorithms to make decisions or interact with their clients. Writing a good algorithm is very hard and I've generally not been all that impressed by them in video game matchmaking.

But then there are other reasons why I vastly prefer server-based communities when it comes to team games as well.

The Overwatch algorithm is not some beautifully abstract system for pairing players together but a designed system with the biases of its creators driving what sort of outputs it strives for. That sounds sort of banal, but it gets lost a lot in discussions about algorithms and automation.

I wouldn't even necessarily make the Uber comparison before I'd make the comparison to the questionable uses of algorithms being deployed to evaluate if prisoners are fit for parole. Because the algorithms are fed on data sets and designed by people they still are at risk of producing biased results.

Basically: we need to stop acting as if, by transforming some decision making process into a data driven algorithm, we're somehow intrinsically making a better more "truthful" system

So just to get it out of the way first, I also don't understand the jump from matchmaking to all this Silicon Valley, Uber/Facebook political nonsense (which isn't to say I disagree with with your hinted-at view points, just that I don't think they were necessary in fleshing out the OP).

Matchmaking in Overwatch (especially the ranked MM) is something that has really turned me off from the game. It's just one of the most poorly done systems I've had the displeasure of being forced to use in quite some time. They just make so many poor assumptions, draw so many bone-headed lines in the sand, and it just reeks of a team that has no clue what they are doing. Which is surprising coming from a company that has had MM in their pedigree for quite some time (with SC, WoW, HoTS, HS, etc) where you would think they would have learned some lessons and applied them here. For instance:

1) Having a difference of just 1 point, 1 point in team average ELO makes one team an "underdog". That is just insanse IMO; I mean the balls on the devs to assume their rating system is so infallible that 1 point means the team has a low enough chance of winning to give them a "boost" for it. Just. Boggles. My. Mind.

2) How can we still have ranked games where a whole team is penalized for 1 person rage quitting. Especially when all maps have an automatic "reset" (switching sides, etc) point where you could easily insert a new person to keep the game competitive and fun.

3) ELO works best in 1v1 games, which is why it is used in chess; but the more players you allow to "group" the worse your system is going to be at judging all of them. Not even LoL lets you group with more than 1 other person because it skews the MM too far. No clue why blizzard thought they could do better and just allow any number of players to group together.

4) Smurfing is something very common, but judging from other players streams (and my own experiences) blizzard seems to have completely forgotten about it when dealing with their ELO systems in this game.

5) And just all the other small things that can be done to abuse the system (like a player leaving early causing the game to end and no one gets anything).

And here's the thing I disagree with the OP on; MM is not a zero-sum game. You can have plenty of fun in a losing game so long as it is competitive and close. Where MM fails is when you have one team get steamrolled by the other team, or teams that are totally imbalanced w/in the team but not compared to each other (so each team has a "godly" player that does most of the carrying and a "newb" who is constantly dying and dragging the team down). There is plenty of room between those two areas where plenty of fun and competition can be had w/o someone getting to have a good time at another players expense.

Yay, a good guest article!

Looking at matchmaking from an ethics angle, games with manipulative matchmaking like Clash Royale are pretty atrocious.

This article is so true. Thanks for posting this.

I find their claims of competitive play mode being skill based rather laughable to begin with. An exceptionally good player is still only 1/6 or 16% of the team. When your individual "skill" rating goes down despite getting several gold medals, they're being punitive in their rating degradation to compensate for algorithm failings that necessitate some outcome. So something you had no control over to begin with determines your "skill" rating, not so much individual skill.

'They got a President, ain't they? They got somebody who knows what a shotgun's for, ain't they?'

Really enjoyed reading this. Thanks!