I stg this game has soft rigged matches

General Discussion
38

This is a good point, but I have some ideas (observations and thoughts) here which might be interesting.

Raw rating based matchmaking is a rare concept. You know pretty much all the considerations: playing with the same people, sniping, team compositions…

I have also observed a slight tendency to have counters - and things like mount or skin coordination. Not to mention the appearance of names in the most fitting times, such as HelpTheNoob or HappySpreader. Keeps me joking about living in a simulation.

A relatively simple solution to this all is to use something similar to the traveling salesman problem, using weights as a second dimension. Since MMR isn’t entirely accurate either, a weighed sum of these metrics can be used to actually find optimum. A lot of measures can be used with various weights, such as composition winrate, counters, matching mounts, waiting time for players, mirror composition bonus, meta composition bonus, the possibilities are endless. Some of these can be hard filters before the famous expanding search criteria state (which is per player). Then you sort the potential matches and pick the highest score (or a combination of).

(And it’s not even silly. Would you have a fair match by rating where one team hard counters the other, or rather a perfect 50-50% as far as compositions go, but one team is slightly higher rated? The mounts part is silly but I swear I’m seeing things :laughing:)

The rest is confirmation bias. You can have a counter in your first match, or after a few wins.

My Sylvanas streak yesterday, there was a Butcher in the first 9 or so. I’ve seen Li Li twice. Pretty much all matches had 4 AA per team (and Kael, likely because of Butcher). Pretty much all matches had stealth heroes. Notably, however, they were equally often teammates and opponents.

In short: it’s not magic actually.

ps.: Having one or two counters intentionally - as in, either logically defined or by winrate - as a weight can give a bit of replayability and fun. It’s like M+ affixes. We don’t draft, however we can still understand the setup given. Sure, it happens automatically, but it can be considered as “added fun scenario” in the above system.

2 Likes
39

The GM#1 comment contains a hint of sarcasm.

I’d imagine every player who feels they are lower ranked than they should be and the main reason they are not at that rank is the system is biased against themall differ in what rank they picture themselves.

Someone hardstuck silver could think I should be gold, another could think I should be plat.

Also with that mindset I feel there’s a good chance someone who was previously hardstuck silver and eventually climbs and stays in gold for a while will probably not be satisfied being hardstuck gold. Then it will be “I should be plat but blizz keeps putting trolls on my team.”

I’m not employed by blizzard/have an interest in defending hots

Based on what i’ve seen and my own experiences, to me there doesn’t appear to be a forced 50% winrate.

I do however see plenty of people who don’t seem as good as they think they are, make mistakes and blame other people for losing.

3 Likes
40

There are a number of studies into “illusionary superiority”, so it’s not uncommon to find someone that thinks themselves ‘better than average’ without actually knowing what an ‘average’ would be, let alone how to be better than that. Part of the issue of people not understanding how to be ‘good’ at something generally leaves them milling about at an unimpressive skillevel; since they don’t know they could be better, they don’t get better.

I can recall an incident with my nephews where they thought they had set a “world record” on Smash Bros. Ultimate’s homerun contest. They weren’t juggling the bag with combos; they weren’t using animation cancels to rack up damage points faster than usual; they weren’t throwing the bat to prevent the bag from launching prematurely; they weren’t getting more than 100% damage on the bag… They simply did a ground-pound on donkey kong and then used the bat and thought they had figured out something extraordinary. I’m honestly not sure why they thought so far as to think they had a ‘world record’ without looking up what the record was (at that time) let alone watch how it was attained.

A “typical” (assumption) player isn’t going to really know what is involved in “skill” but they are certain it was their “skill” when they win, and not a consequence of their lack of skill when they lose. It’s much easier for people to assume some system has to be out to get them if they aren’t satiated on their typical ‘false positives’ that reinforce their shallow expectations.

If a task does not hard-require people to have a specific skill to ‘rate’ their performance, they they will fill the gap with their ego instead, and damned be anyone that doesn’t agree with them on that :wink:

4 Likes
41

Many people overestimate their ability yes. Others observe patterns. Minky referenced above the devs did admit to putting 50% WR in the game. Not sure why it’s controversial, anyone is free to disagree the extent to which its prevalent.

42

That’s not what I said.

As you insist on twisting the words of everyone posting here, I will be lazy and just quote AltRightFrog to clarify further.

Ask someone here for the link to the Blue post if you’re really interested in exploring this further.

*Edit, OMG, sorry for calling you AltRightFrong! I’ve corrected that.

4 Likes
43

because you’re effectually illiterate. When people don’t read things, then they don’t understand things related to the reading material.

this is the topic likely to be cited for “admitting” anything but you are the sort to not actually ever reference anything, so people don’t actually know if you’re full of ‘it’, or actually informed, or just posting contrary things to ‘troll’ the boards. Not that the options matter, this is already too long for you. And it’s not like this is the only game that people have a dev trying to respond to “forced 50” to quote as ‘admitting’ something they didn’t read.

The statement of being ‘true’ does not mean they ‘admitted’ to programming the game for it, it is ‘true’ in that it is what people “experience” as a consequence of matchmaking existing, much the same as it is ‘true’ that people “experience” seeing a black and blue dress appear to be gold and white. However, just because the ‘sensation’ is what happened, does not actually mean that that is the “truth”.

However, it is far easier for people to keep claiming that something is rigged than it is to read 12 words past their imposed limit and consider they might be “wrong” about something.

6 Likes
44

Neither is your attitude for discussion. Thus, the point of calling you out.

Those respectfully replying to you have a plethora of knowledge for the game. They should’ve stopped wasting their time on you 30 posts ago.

You made a wild claim and still have not expressed or provided any logics or proof to those claims, even when questioned.

:roll_eyes:

1 Like
45

I’m always happy to discuss stuff with ya, but when I start getting sleepy reading it, it might be a bit long. Maybe we could take a philosophy course together someday. I looked at some of your blue text there. I guess if you guys always believe what you’re told, that’s all well and good. Some of us reserve a healthy does of skepticism which typically holds true.

46

If I give any analogies, I’ll be accused of going off topic and Xeny will write a . Generally people don’t run out and tell you their business practices. Anyway, I’ll continue noticing patterns, even if people don’t appreciate it. No point in people getting so stressed, it’s fine to disagree. Maybe someone can start a new thread.

47

write a novella correction*

48

There is a functional difference in being able to tell when people are being honest about what they’re talking about, and refusing to believe anything that doesn’t agree with an ignorant opinion.

The people that tend to refuse to differentiate their opinions are also the ones least likely to actually be informed about it, but they’ll act like they know better regardless of the apparent contradiction.

What sort of effort would it take in addition to the existing game to ‘rig’ the matches as people suppose to be the case? If those sort of features were implemented, why haven’t the code-reading dataminers revealed them?
By the way, there are fans that datamine the game and have posted fixes to particular bugs, so while they can be ‘in on it’ to suit your ‘skepticism’, the bottom line is people that agree with the chronic conspiracy complaints do not actually know what sort of effort it takes to actually realize what they claim, and they haven’t thought about the implications of people not acting providing the ‘proof’ they claim is evident…

That’s why there is little corroborative evidence, constant assurance that it’s ‘easy to see’ without actually demonstrating any specifics, and it has to be a flaw of someone else to not ‘clearly’ agree, or rather, it takes some tremendous ‘class’ or unbeknownst field in some unrealized nether to possibly consider that sometimes the simplest explanation is actually the correct one, and the simpler explanation is that creators are often more ‘honest’ than the crying critic.

For stuff to be so ‘obvious’, it’s pretty damning how lazy these things have to be to not put in that iota they claim that sets apart the ‘obvious’ they can’t spell out, and yet, somehow not notice the connection between their laziness in one area to the next as being problematic in discerning sincerity. Then again, if they’re not going to be honest, it falls to reason they expect others to not be either, but it’s not like they’re going to be ‘punished’ for that sort of lying.

2 Likes
49

4 Likes
50

Either that, or you need to improve your attention span.

I’m sure it does.

But nobody is more certain their skepticism is healthy than a person with an unhealthy amount.

2 Likes
51

I wish it were possible to have an unhealthy amount at this point in history, regarding this topic. But, a person needs to have good pattern recognition and be willing to unlearn years of misinformation.

Hey Xeny, I see your reply too. :slight_smile:

52

The pattern to be recognized, of course, is people who claim algorithms can be rigged and then fail to provide a self-consistent model of such an algorithm and how it actually accomplishes its intended goal. Most famous and long-standing on this forum is to trumpet some silly idea of a forced 50% win rate that somehow only applies to one person and not the other nine people playing in the match, which the devs ostensibly do in order to make money in a manner that’s never clearly explained.

There’s a similar pattern, actually, of people claiming that they were banned for absolutely no reason, despite their forum posts showing that they have absolutely no sense of common courtesy.

The similarities in the two are left to the determination of the reader.

5 Likes
53

So what patterns are you recognizing that we aren’t?

There’s not a forced 50% WR due to a win streak. The game isn’t putting idiots on your team just because you won 10 in a row. The game isn’t putting blinds on your team when you pick an AA hero to drag your WR down.

The purpose of the matchmaker is to find you matches that are fair and balanced. How does it do this? Well that’s pretty straight forward. When you win, your MMR goes higher yes? When you lose it goes lower. Makes sense right? If you win 5 games, your MMR increased 5 times and so naturally you’ll be facing players with a higher MMR because its matching you with players around your own (new) skill level. If your true skill isn’t on par with these players, or higher, you’ll most likely lose. Its not complex. If you played chess with a person of higher ELO, most likely they’ll win. Now inversely, if you were to lose 5 games in a row your MMR will drop 5 times, and then in the 6th game, if you are truly better then those players you’ll win again. This is of course assuming the playerbase is large enough.

It isn’t, so the MM has to compromise a lot more. There’s also the issue with smurfs, trolls, or complete idiots. The MMR can’t counter these things because there’s no way to detect whether or not the player behind the screen is an idiot, troll or smurf.

The purpose of a matchmaker is to find balanced matches, which only can truly work when people are at a MMR that represents their actual skill level. You’ll know when you’re at that point when you are more or less around 50% WR. Give or take a couple % points.

If you think the game and its devs are lying to you, then you just don’t understand ELO’s or matchmakers.

1 Like
54

citing the post of a developer who had to leave because he was incompetent is the formula for failure.
the win or loss of games does not depend on the mmr.
It depends on the synergy between the heroes of your team depending on the map and that they can play a coordinated game.
this is not always obvious.
matchmaking failure is a product of the inability to order chaos.
because you think lol has only one map.

55

That’s a variable, but I’d argue player skill is a bigger factor.

1 Like
56

People blame MM for putting trolls and feeders on his team but people need to understand MM cant control human behavior.

Also if you look at people’s mmr on HP they are divided into catagories. Someone with high account mmr can still be trash with a hero he is bad with and that way drag his team down.

And then you get the usual ‘‘mmr is rigged’’ threads.

People also forgets that MM cant make balanced games all the time. It can only make games out of what is available to it.

Thats why you get exstended que alot of times because MM have to throw alot of rulings away and soften up its search. And when it does that rainbow games tend to pop up alot more. Then you ofc can end up being the highest winrate player on your team while it gives you 2x 35% winrate trolls that ruin the game.

Nothing rigged about this game. My account got 16000 solo games and I have been able to maintain a positiv winrate for a long time.

Only time you will see people with extreme high winrate over 1000 of games is people that abuse QM for free wins with special comps opponent stand no chance againts. The MMR of those people are usually 600-1000 mmr higher than your team cause MM cant win a balanced game againts those people.

2 Likes
57

Yeah, some players will rig matches grouping up. Wintraders are probably very rare but also a thing.

You could argue that they should have a seperate solo queue and team queue. However that would probably make queue times unbearable with the small player base.

I used to like playing with my bro and there will be odd player that invites me to a group after a game.

It can be fun playing with people but i find it weird when people smurf, group up as a 4 + stack, voice chat and flex on the lower ranks.

Plenty of people seem to like boosting their ego with wins and artificially high winrates, which is weird and toxic.

1 Like