Elo Hell is real

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

Tale, if you doubt everything hard enough, anything becomes possible!

1 Like

I agree that ELO hell exists, but only in a specific circumstance. Because on this game you need to be BETTER than your current rank to level up to the next rank, there is a big ‘carry’ requirement in this game because it’s quite grindy.

But I DO NOT believe that individually, the game is holding people back, AKA rigged.

You said you were GM and now you’re in Plat, I don’t know how long it’s been since you’ve played but the ‘skill’ is clearly there to climb.

If you wanna group up with like minded players give me an add on Discord - RonSwanson#8107 - we have a lot of masters/GM players who will be happy to queue up with you. We also have coaching from people who do it it their spare time (not saying you need coaching), but if you’ve been out for a long time then, maybe :wink:

Either way dude, don’t get stressed out this game is awesome, you just need to right condidtions to climb.

3 Likes

You are a sym lucio main with 100+ hours in quickplay.

I’m starting to see a pattern of the kinds of people saying elo hell doesn’t exist.

You are the kind of player that make elo hell a thing.

1 Like

you are level 38. Please do not speak on competitive matters as it is clear you are unexperienced

1 Like

The answer to ‘what is the best way to keep people engaged’ is absolutely intermittent reinforcement, which is exactly how casinos operate, and to some degree how Overwatch operates. Studies have found that intermittent reinforcement patterns keep players playing longer compared to predictable rewards. Therefore, there’s a great argument that randomness (grind) isn’t a bug, it’s a feature. See the excerpt below:

"Reinforcement theory is an area of behavioral psychology that explores the ways that people and other animals respond to rewards delivered according to various schedules. Rewards may be delivered continuously, at regular or irregular intervals. In rat studies, researchers found that the way to keep the subjects working longest was to reinforce the behavior at unpredictable intervals, rather than regularly.

Intermittent reinforcement and other principles of reinforcement theory are applied in areas of technology and business including [human resource management] marketing and [machine learning]. In designing mobile apps and social media, intermittent reinforcement might take the form of rewards delivered on a schedule that seems random to the users but is designed to keep them active for longer. For example, “likes” might be reported when an algorithm has determined that the user is likely to leave Instagram. Through its algorithms, Facebook can detect when a teenager is likely to feel insecure and deliver a confidence boost.

Online gambling and internet addictions exhibit the same principle: The individual receives just enough rewards (which might be in the form of wins or entertainment) at sufficiently sustainable intervals, to encourage them to continue."

Not according to studies, as “easier” and “harder” seem ancillary to reward schedules. Intermittent reinforcement, which has a high failure rate but features unpredictable rewards leads to more persistence and higher total effort. Which is more exciting to you, finding a $100 bill on the ground or receiving your paycheck?

Managing wins and losses and their schedule, as is routinely done in casinos, is indisputably and objectively the best way to encourage continued behavior in such a setting. So it seems to me that keeping players grinding for as long as possible with intermittent reinforcement is arguably the best available business model.

If this were true, anyone experiencing a net gambling loss would stop. But they’re not operating according to reasoned analysis, they’re operating according to the vagaries of dopamine while being subject to intermittent reinforcement.

This is not how casinos work. And if there were a more addictive and enticing approach than intermittent reinforcement, they’d be doing it.

There are problems with your premises as I’ve shown above, so this conclusion does not follow.

1 Like

But what are the rewards for playing comp- what the proponents of the various “rigged matchmaker” theories report is that they get stuck and get no rewards for playing comp. Their rank either stagnates or drops.

If they wanted to provide intermittent rewards they would skew the results such that players ranked up intermittently. What people who gravitate toward “rigged matchmaker” theories report is that they do not rank up.

That’s the whole issue people are trying to explain. And this:

Is not what we have here. Again, people quit playing OW. People periodically post threads wherein they ascribe their decision to quit the game to the “rigged matchmaker.”

Again, in order for this theory to hold water, we would have to suppose that a matchmaker which holds people back from meaningful progress and that many players decry as “rigged” actually encourages people to engage with it.

It’s just not a tenable theory, given what we can observe of the game, it’s declining (prior to OW2) popularity, and the discourse of the player base.

You could get more players to stay and get them to play more games, and that would increase revenue, but there’s no way that was happening here. They were demonstrably making choices that were not predicated on keeping more players.

And I’ve never heard anyone describe one of these hypothetical rigged matchmakers as desirable in any way. What people are trying to do is find a reason to believe that their rank is disconnected to their skill. And they are willing to believe highly implausible things as long as those highly implausible things confirm their pre-existing belief that their rank in OW competitive is not tied (in a meaningful way) to their skill.

3 Likes

Unlucky I can’t post links or screenshots here. Wanted to post screenshots of how I get called hacker and smurf in games, but somehow can’t climb.

2 Likes

Yeah man, what frustrates me is the fact that the longer i play this game, the lower rank i am in. I started 4 years ago and easily climbed to silver/gold, since then i have learned a lot about the game and now… i am dropping like a rock… and can’t do anything

2 Likes

I think it’s the opposite, people are looking for reasons to believe their rank is connected to their skill, when all of the evidence suggests it is disconnected. How can you know where you belong when there are often several skill levels playing together in the same game? How can you belong in a rank if some players in your rank are much better and others are much worse than you? If the ranks meant anything that would not be such a common occurrence.

1 Like

I don’t want to chase the high sr number to feel good about myself. I just want to have good matches of overwatch where people care about things like the objective, hero picks, grouping.

That’s the thing though- all of the evidence does not suggest that rank is disconnected from skill.

How many top 500 players have you seen that resemble bronze players? How many unranked to GM or bronze to GM streams have you seen during which you said to yourself, “these games look the same as the streamer progresses up the ladder.” I don’t know what ranks you’ve seen, but how often have you been in a silver match and thought it was like a plat match? etc.

So the gameplay that we see on the ladder suggests that rank is connected to skill. The dev statements and the in-game communication suggests that rank is connected to skill.

The claim that rank is disconnected from skill is the bold claim. It, therefore, needs to be supported by strong evidence.

4 Likes

You are not playing comp if you think the players in matches are all of similar skill matches. You have plat players playing like golds and golds playing like plats.

Even when I played in higher ranks, there were boosted players that just didn’t fit the SR.

Or in what rank would you for example put a winston in, that ults and doesn’t contest the payload that is being pushed into checkpoints? Or people that don’t group or don’t switch heroes when hard countered? Players that can’t aim or even play their role. Those are the kinds of people I play with in plat. I am not one of them. You are extremely wrong with your mental gymnastics.

2 Likes

Now your stats are finally available it is quite evident what the problem is. You are dying way too much. For instance on Widow and Hanzo you die more often than 99% of Widow’s and Hanzo’s. In fact, on all of your characters you die more than the average person. Not one exception.

You can obviously aim and are doing a lot of work but if it’s not winning you games you should focus more on staying alive than getting that marginal kill. Alternatively, you can keep blaming teammates.

3 Likes