Elo Hell is real

Our experiences are different then, or maybe we’re talking about different points in time, or different subreddits. There are strictly competitive subreddits, which aren’t going to feature (or even tolerate) conversations outside of tips, tricks, strategies etc. Then there’s more general subreddits where these kinds of conversations are allowed. Guess it comes down to which ones you’re talking about. In my experience, the general subreddits where progression conversations took place were not civil, mature, or open-minded. And how civil one thinks the FOX news or MSNBC youtube comments sections are, for example, is going to have a lot to do with your political ideologies.

Also, are you saying that anyone who thinks the game is rigged is part of a conspiracy theory? And if so, why?

Again, is the implication that anyone who disagrees is doing so on the basis of emotion? I’ve straddled the fence between both sides of this argument, and I think there’s valid points to be made on either side. But neither side has seen the code, and neither side has provided conclusive evidence. That skill “solves” climbing doesn’t invalidate any and every claim that the matchmaker has problems (I’m not saying that you’re saying it does, but this is not an “open and shut” case).

People rally around Taleswapper as the final say on this, but he isn’t: all he’s done is present generalities about how sorting algorithms can be written and then makes the highly questionable leap to “that’s how they’ve been written here.”

I retracted my post on this topic when I made the same (egregious) mistake of suggesting that general matchmaking theory best practices had found their way into into Overwatch – with absolutely no evidence that this was the case. This is like advocating that every doctor is a good one because SOME take the Hippocratic oath seriously. This is a category error.

To be clear, you’re suggesting that every criticism of the matchmaker is either flawed or emotional? How do you qualify an emotional argument? It would seem to me that most (if not all) of the arguments you’re referring to as emotional are just standard issue abductive reasoning. I’m not even sure what an “emotional argument” is now that I think about it. Whether an argument is effusive or not has no bearing on whether it’s right or wrong. Beyond expressing it emotionally or not, underlying claims are still being made, and those are judged on the basis of whether they’re valid/sound.

I haven’t posted there on the general Overwatch forums. I found the lack of moderation, immaturity, and tribalism there to be a problem; not to mention that it was mostly a breeding ground for bad Mercy fan art. And Competitive Overwatch and Overwatch University, et al, aren’t places where matchmaking can be discussed much without being off-topic.

Whether something is “incorrect” or not is not an excuse to be rude and disrespectful. And the category of what’s “objectively true” about matchmaking which goes above and beyond opinion or confidence or belief is small. There’s a lot of speculation taking place on both sides.

To me there is no question which side of this debate is more aggressive and rude. Typically it goes something like this:

Scenario 1:
Upset Person A experiences something in the game that tilts them. They come to the forum with complaints and theories about the situation. Those posts are swarmed upon by people that disagree with them. They’re often insulting and personally demeaning, and they’re placed into this category of generally delusional, lazy, conspiratorial or senseless. People attack these people not as a new person entering this arena, but as an extension of all the other people they’ve rolled around in the mud with. Person A is attacking the game, and then people attack THEM. Those people get defensive having been insulted, and they respond in kind.

In fact, Taleswapper is maybe the only person on these forums who seems to have a standard of behavior below which they will fall (and he deserves a ton of credit for that).

Scenario 2:
Someone who’s smurfed or hardcore grinded their way to rank X makes a post on the forums offering that this situation proves that anyone who criticizes the matchmaker is a hardstuck “copium” addict. The gauntlet having being thrown (which reminds, check out Ridley Scott’s “The Last Duel” – awesome movie) people get defensive and the predictable ensues.

I have never seen anyone rudely attack Taleswapper or anyone else like him who calmly and respectfully tries to explain how the matchmaker works in fact-based way.

I think there’s some justification for spending time here to write and debate, so it’s not all just an egregious waste of time (I’m honestly way more bitter about all the hours spent playing Overwatch). I credit this place with me A) doing a hell of a lot more writing and proofreading than I otherwise would have and B) getting me into a fairly common routine of analyzing my own arguments and reasoning and that of others. It’s clear you do the same, so cut yourself some slack. :slight_smile:

I honestly think you’ve become a better writer and communicator since first knowing you, and I think I have too. Not that either of us were bad before, but there have been a lot exchanges that forced me to think through my arguments (and yours and others) and there’s no way that all that time and practice doesn’t come with improvements. I think you, Basil, Tale, Cuth, Rigged (I know you’ll disagree on this one), myself, and others have elevated the level of discourse on this topic, or at least we routinely present arguments that take time and effort to respond to and unpack.

We all do it. We have a strong internal sense of what’s true and what’s right and wrong, and we seek out information that confirms it, all while being less “moved” and impressed with information that doesn’t fit our worldview. No one likes to be wrong.

Most subreddits just aren’t a good place for open discussion unless it’s a fairly neutral one, or debate and discussion is the reason it exists. People come to Overwatch related forums wanting to express their opinion, and what they get is immediate, withering confrontation. And it has a chilling effect on other people who see that. And like I said, the way Reddit is run makes it very very easy (if not inevitable) for tribes, groupthink, and hive minds to form. Then those places form an immune system around any dissenting opinion that passes through.

There’s no question that shutting up, and “getting after it” is the superior option when it comes to accomplishing the goal. Life isn’t fair and at the end of the day, people content to merely complain don’t get very far. Overwatch, like so much else, has a tipping point: a critical moment where enough small changes eventually produce a large or irreversible effect. I think anyone with a given mindset about the game (or life in general) is going to, commendably, meet that tipping point, or that line in the sand wherever it is (and in this context, it’s how good should you have to be to get out of, say, gold, and again, keep in mind that people are comparing OW to their other life experiences, ones where there is no Blizzard designing the landscape, and those real-life experiences strongly informs their opinions on what’s fair and what isn’t).

But there is a conversation about where that line/tipping point should be. You combine this with the – indisputable, objective fact – that Blizzard does have some say over where that line falls, and there’s certainly a valid conversation to be had about whether the placement is correct. That’s just the nature of life on this planet: any decision is going to be open to question and criticism.

Group A: “meet the line wherever you find it.”
Group B: “the placement of this line is ridiculous.”

Like you said, it’s going to come down to some reasonable middle ground.

You’re right, and apparently the community doesn’t much agree, lol. But it’s weird because it’s like asking people who are out of work and suffering vs. those who are employed and comfortable what their views on unemployment benefits are (aka identity politics).

As for developers (or any business really) and what they’re trying to do: It’s a spectrum really. Some are more than happy to exploit people any way they can, and it becomes a very “us” vs “them” mentality. Others see business as more of a symbiotic relationship where the goal is for both parties to flourish as much as possible.

I don’t think the gaming industry is generally on the correct side of that spectrum (which is often the case when you have shareholders, because exploitation is profitable in the near term, and people often don’t care much about the long term – they’ve already cashed out their chips by then).

p.s.

I can’t believe how absurdly stupidly long this. I only really see how long the post is after I hit “save.” It doesn’t seem like much while I’m writing it. Ridiculous.

1 Like

Delusional once again. People complain about having bad teammmates and luck in every game. Why do you think that league of legends is universally known as the most toxic game? Surely a perfect matchmaker would leave no reason for people to be angry? Not the case. Your teammates doing bad matter so much more in that game because the enemy scales in power based off of your teams deaths. Yet in LoL good players will climb just as Overwatch. Bad players are always going to blame something else. That’s why they are bad. It’s really not difficult to understand.

3 Likes

I think this is the most useful bit for me to respond to.

It seems that your current claim, BrightTitan, is that the matchmaker is tuned such that players struggle to advance, because the OW devs want to increase the grind of the competitive ladder in order to make more money.

And that is a different claim than many others are making. It is a specific claim though. And it can be interrogated.

The main question to ask is this: does this version of the matchmaker actually increase Blizzard’s revenue? That is, does it get more players to stick around longer and play more games (and presumably buy loot boxes?)

I think the anecdotal evidence suggests that it does the opposite. Players actually leave rather than sticking around. And this happens for all sorts of reasons, but one of those reasons is the sense that the matchmaker is unfair- that the ranked mode unfairly holds people back from progressing.

And that brings me to my larger point- there are established principles of engagement based game design. Devs know how to do this. And the hypothetical matchmaker that prevents people (or even significantly hinders people) from ranking up ain’t it. Engagement based game design is the opposite of a matchmaker that holds people back.

There are various versions of this, but the shorthand is that gamers must feel like they are continuing to make meaningful progress and have agency if you want them to continue to feel engaged in your game.

That is, they need to progress. That’s point one. And they need to feel like they are progressing because of their skill and their choices. That’s point two. There’s a third component that sometimes gets brought up relating to meaningful social interactions in game. But the first two points are key.

So the whole theory that the matchmaker is purposefully calibrated to hold people back from meaningful progress in order to increase Blizzard’s revenue through engagement based game design does not even make sense on a fundamental level.

If they were doing that in a way that worked- it would require them to let players progress as their skill increased. In fact, if they wanted to skew things- they would skew them in order to increase progression.

Stymieing player progression is like the antithesis of engagement based game design. But the problem is that we are not trying to make a meaningful investigation of what is actually happening in order to better understand it.

We are simply trying to back-fit some theory (any theory, when one is shot down we’ll simply replace it with another) in order to explain why our lack of progress does not relate to our actual skill. And as long as that is our methodology, we will continue to misunderstand what is actually happening in game, we will continue to hinder our progress, and we will continue to undermine a good in-game experience.

4 Likes

Actually one more bit on this point. One of the ways the “rigged matchmaker” theory has been suggested to work is that players are either white-listed or black-listed, and only white-listed players are allowed to move up the ladder. And one way it has been suggested that players might get white-listed is by buying lootboxes.

That is, at least, a coherent theory if one were looking for a ‘Blizzard is doing things that undermine the accuracy of the ladder in order to increase their revenue’ theory. I don’t think there’s any evidence of it occurring, but it at least makes sense on a fundamental level.

The big problem with it is that if Blizzard wanted to implement a pay-to-win scheme, they would need to let the players know about it. How can they expect us to pay-to-win, if we don’t know we need to buy lootboxes in order to progress?

3 Likes

I think that’s a fair summary, I would agree.

I used to agree with you on this (remember our ‘dial’ conversations), but the more I think about it, the more I am inclined to believe it’s not true for Overwatch.

A few weeks ago, I gave the following example to demonstrate that being only marginally better than your peers results in an insane amount of games:

We have established that a 50% wr means you are accurately placed on the ladder, i.e., your actual skill corresponds with your SR. I think it is reasonable to assume that 52% wr corresponds to being marginally better than your peers.

Let’s, then, assume a hypothetical player who is placed at 2500 SR on the ladder and to which the above property of being marginally better than others at 2500 SR applies. Now to rank up 100 SR, said player would have to play 100 games. This is an insane amount of games that most players don’t even play in one season, and that’s the requirement to climb a small amount. You can play with the numbers a little if your definition of ‘marginally better’ is different to mine, but you’ll find the ballpark of the number of games is very similar.

Here is the crucial logical conclusion: the above observation holds true for any matchmaker, be it ELO, TrueSkill, or any other variant. In any matchmaking version, 50% will mean accurately placed and being marginally better will result in a ton of games to climb. The only scenario for which you could logically assume ‘artificial grind’ is one in which the matchmaker treats some players of the population differently than others, i.e., preventing some but not all to climb. And I know you don’t assume that.

I agree on this and I think they are taking a step in the right direction by changing to a 5v5 design. The game will still be more team reliant than the other games mentioned, but it’ll definitely reduce the feeling of ‘being held back by team mates’, I think.

I have a feeling that we will be talking about this again once OW2 releases with a battle pass and or the pve part of the game where this will be of great relevance, hah.

If anything you have inspired a few people to pick up on it. You know it is very important for me to not misrepresent the arguments made on both sides of the debate. I take this very seriously, and I have only ever seen you make this point. Although, I obviously can’t claim to have read every single post and comment made on this topic on here.

3 Likes

I wrote about this while ago. The devs used to have a bit more progression built in, in that they artificially started you lower after placements so you could climb. This was about as artificial as you can get, even though I normally hate calling anything in a video game artificial since it’s all artificial.

There really isn’t progression in OW and I really don’t think it was ever intended. Not that you can’t improve, but it takes massive amounts of improvement, as you noticed, to do so.

I hear people say that it’s marketed otherwise, but I never saw the marketing where it said you’ll get more SR on time played. I could have missed it, or likely just interpreted it differently though.

Like, BT is taking above about “grinding skill”. That’s just not how skill works. No amount of grinding will raise your SR.

2 Likes

This is a good point and one I’ve brought up as well. If we want to see what engagement based game design looks like in an OW context, we only have to wait for OW2 (and particularly once the PVE comes in). One thing that I expect to see is one or more options for player progression that are not tied to winning competitive matches.

Players will be unlocking resources with which to progress their PVE skill trees, for instance. And while I am sure there will be ways to do this solely through playing the game, there very well may be ways to increase the rate of gain by spending currency.

But the core element will remain- engagement based systems require a sense of progression. The whole impetus behind the ‘rigged matchmaker’ theories is to find a way to explain lack of progression.

That is the opposite of engagement based game design. Players need to be able to at least pay to gain access to player progression- and that’s true even in nefarious engagement based design.

1 Like

That’s very interesting, thanks for that! My point stands though, something like this would only ever be possible for placement matches. Other than that, artificial grind could only exist in a system that treats different players differently, which no matchmaker of the games under consideration does.

This is what I had in mind also.

1 Like

Yeah. I don’t know what the whole system will look like, but I do know they’ve already said they want hero missions to be replayable and there will be end game content. So whatever form it takes, it’s clearly being designed with long term engagement in mind.

3 Likes

I was agreeing with you.

I think it’s funny that someone has a complaint about artificial grind in a game that I don’t even think has grind, and if you do choose to progress then it’s very much not artificial.

The fact that people think they should progress at all is the problem here. “Should”, being a key word, rather than “could”.

1 Like

Oh, I know. I was just reiterating it for people who were too lazy to read my long-ish comment, hah.

2 Likes

I have in fact watched these videos. Yes.

I have no idea what you’ve been saying for the past four years. I don’t read every member’s post on here.

Lettuce simply mentioned complaints about the matchmaker. If he wishes to then he may now backtrack that statement and move the goalposts. I simply responded to that claim.

You evidently haven’t watched the second video. The matchmaker is literally described as being [a vulgar term beginning with S.]

Matchmaking is what he said. Not the system, not the matchmaker, but the fact that people camp their ranks, too many smurfs, and game design choices around groups. Again, player mentality and game design.

You, on the other hand, appear to believe that the matchmaker puts people on your team that it KNOWS are bad in an effort to ensure your win rate stays at 50% .

No one with any sense is making the complaint you guys make here. If you are actually making a complaint about how player behavior drives a poor experience given OW team based design, them please be more clear so we can get this silly argument over with.

2 Likes

I’m just more putting it forward as a competing theory. I’m pretty much agnostic on all of it. I’m not making any claims one way or the other. I’m pretty particular about not making any declarations (most of the time) because I honestly don’t know. Don’t think I made any claims about it being artificially difficult, but there’s evidence that there is and there’s evidence that there isn’t too. I think all of these conversations have to be viewed more as “there’s strong” or “there’s weak evidence” for this or that rather than “this IS the case.” I fully accept that you or RHA or Tales are totally right on this and I’m totally wrong, but we have never, and probably never will get a definitive silver bullet from Blizzard. People might disagree with me and that’s fine, and we are probably weighting various bits of evidence differently.

Hmmm, “marginally better” works nicely in a system where skill distribution is good and even, and poorly where it isn’t. I’m not a game theory expert, but randomness probably hurts better players more than it does weaker ones.

I think if this was established there wouldn’t be a “great debate.” :slight_smile: The other side of the argument says that there’s a “gravity” at around a 50% winrate. Such that when you try to break away from it, randomness or or intent or both pull you back down. If you’re a better player, there’s a much better chance that you’ll encounter random players who are worse than better. Also, factor in that tanks, which are in demand are shipped all over the place creating large variances in skill. You and I already established that as the player pool reduces, randomness goes up. As stacks queue up, randomness goes up (consider how it’s mathematically more difficult to match 12 people of equal skill in OW, versus two in chess). Also apathetic players who’ve allowed themselves to drop into lower ranks through a lack of consistent effort also inject randomness. Alts and smurfs = more randomness. Queue time limits = more randomness. There’s so much randomness in Overwatch that it’s tough to say when players are fairly placed, or how often they play in good fair matches, the result of which provides accurate information about the players who won and lost. If a single player throws, everyone else on the team loses. And I’m pretty sure the better you are, the higher the odds you have of being placed with a worse player (more randomness). So if a good player loses multiples times in a row and has bad stats because of randomness I think it creates more fuzziness. The closer a player is to mean, the more a 50% chance of winning is a non-issue, and the more standard deviations away from the mean, the more 50/50 matchmaking otherwise alters that player’s trajectory.

Engagement based matchmaking (EBMM) prioritizes, well, engagement over Skill (it doesn’t ignore it). This means the way players respond to wins and losses (churn) affects their placement in subsequent matches. I think your theory holds if EBMM isn’t a thing in OW. I’ll be blatantly unscientific here, but I have noticed in my sessions, that my first match when powering up the game (when I actually still played) was much more often than not a win. Now, if that has something to do with something besides my skill, that’s EBMM. There are proven benefits to EBMM (which is why it’s a thing). Does it exist in OW? I’d say probably. Why? Because it’s profitable, and it’s how I’d expect a company to operate if it prioritized engagement over competitive integrity.

Agree, and still have zero desire to play/purchase the game. It’s too similar to what I’m utterly burnt out on at this point. OW feels like the feeling you get when you’ve eaten leftovers 4 days in a row. And OW2 is just too similar to what we’ve had. I don’t equate more running around the map for off angle pokes to be a “different” experience, but it’s sad that developers do.

It’s basically integrating deathmatch into comp, but if you’re sick of comp and sick of deathmatch from crazy amounts of sameness over the last 6 years, I don’t know how you get excited for OW2. I’m glad for those that are, but if you’re sick of OW, you’re probably going to quickly be sick of OW2 as well.

Given how long players have waited for this, given how long they’ve done the same blasted things over and over. I personally would’ve taken OW2 in a different direction and made the experience way less derivative.

I’ve only agreed with XQC once and never will again probably, and that’s when he looked at the gameplay for a bit, put up his hand in confusion and said “It’s-a the same-a game-a.”

I’ll be curious to get yours and others take on the game, particularly with you being a high level player. I mostly want to see what the carry potential is like. And how make the game less welcoming and “protective” fairs in light of the profit motive. The harder the game, the less people tend to stick around.

You can literally search for every instance of “engagement.” And yes I do know you are a fair and honest actor in these discussions, though I do see you get tilted every now and then :slight_smile: but don’t we all from time to time?

1 Like

elo hell is a low MMR boogeyman cope for people who aren’t good enough to win enough games.

2 Likes

These are all fair points. But I’d ask, A) does grind exist in any form, in any games? And B) if yes, why? Why would something exist in a game if it’s an element the consumer doesn’t like? Not sure if you watch gaming channels at all, but bellyaching over grind and progression are topics I see covered, literally, every week .

Repetition exists in any game, and when players like it, we call it part of the gameplay loop. When players dislike it or find it excessive or onerous we call it grind. Grind is something, I think, that naturally seeks its own level. The more a developer can put in without annoying the **** out of players, the more they’re going to.

I also don’t think consumers are rational, or that a lot of what they do is driven by conscious choice-making, and this can be used against them. I don’t think for Blizzard it’s a question of how happy can you make people while they’re playing, I think it’s more about how much you can keep them coming back.

If people will toil longer in a system that keeps them artificially engaged not by choice but by manipulation then padding the process has a business case. If a player has a goal to get plat, and you the developer find that players will, on average, devote 300 hours to getting there before washing out, some might ask: if it only takes them 50 hours to reach plat without adulteration, but they’ll toil for 300, why the hell WOULDN’T we pad the process to take between 275-300?

Bad for the consumer is not necessarily bad for business (exhibit A, nicotine).

Your logic here assumes that people are rational (they aren’t), that people can’t be coaxed into doing things that are counter-productive (they can), and that they can’t be sold a bad bill of goods or be manipulated (they can and they frequently are). Additionally, by your reasoning, intentional grind should not exist as it is “purposefully calibrated to hold people back from meaningful progress,” and yet its rampant within the gaming industry. Either grind doesn’t exist, or some of your premises are incorrect.

No. They would only have to gather data, see how players react in split testing environments, and form a business case around it.

Here’s a hypothetical

Premise: players care more about reaching in game goals than the opportunity costs
Premise: players desire to reach plat and on average will play 300 hours before washing out on that goal
Premise: Under conditions X, it takes them 100 hours to get to plat
Premise: Under conditions Y, it takes them 250-300 hours to get to plat
Premise: holy #$(@ under another set of circumstances, people will keep grinding to plat more than we ever imagined because friends, boredom, new heroes, OWL league, streamers, addiction, on-again/off-again motivation, etc

Conclusion: do whatever the market will bear.

Ad hominem. You may believe unflattering and negative things about the people with whom you’re debating, but trying to undermine their arguments by attacking them or their motivations is a logical fallacy.

1 Like

The things children say…

1 Like

It’s actually not based on the idea that people are rational. It’s based on a lot of studies that have been run in order to determine how best to maximize engagement. There’s a whole branch of educational theory based on this- gamification of education. And the entire thing began with asking the question “how do we get players to continue to engage in behavior/systems that we want them to engage with?”

What we learned is that progress is crucial. In fact, you can get people to engage in a system as long as there is a score, and that score goes up when they do something (and I would argue this is orthogonal to rationality- it is neither rational, nor irrational). Cookie clicker is the quintessential example here, and this is why mobile gaming is so freaking profitable.

If, then, Blizzard wanted to maximize player’s engagement in the competitive mode, they would need to skew things such that they made it easier, not harder, but easier for players to progress up the ladder.

To your point, the goal for gaming companies who want to maximize profit above all else, is to create a system that players can rank up in indefinitely (I think this speaks most directly to your grind point- it’s not that grind prohibits progress, it’s that it provides progress over time- but the progress is still crucial).

If OW were designed that way, the comp mode would not have an SR cap, and players would progress in proportion to how much time they engaged with the comp mode.

The current system is the antithesis of that. It’s not an engagement based system.

3 Likes

First off, I don’t find this whole loot box argument compelling or sensible. It just seems too ridiculous to take seriously.

But I don’t think they’d have to let players know about it. They could do what’s known as painting the bullseye around the arrow: seeing who their most profitable players are, and tailoring their experience accordingly.

If they wanted to incentivize players to spend money on the game, they would need to let players know what the reward for buying loot boxes was. Otherwise, the pay-to-win scheme fails.

But even if your idea were the case, we’d have to believe that players who buy lootboxes want a matchmaker that makes it more difficult for them to rank up. And the strong claim is that the matchmaker prevents meaningful progress for all but the smallest percentage of god gamers.

Why would folks who buy lootboxes want that? There is no credible incentive structure for either the devs or players who buy lootboxes to want the matchmaker that people imagine exists.

1 Like