Unless there’s evidence for the claim, it has no place in the argument and doesn’t need to be considered one way or the other. It doesn’t matter how objective he’s being, what matters is his conclusions. If it’s a flawed argument, his level of objectivity, be it high or low, can’t salvage the argument so it’s a non-factor.
2 Likes
Remember when people were concise and got to the point?
Yeah Me neither.
3 Likes
I would need hard proof Blizzard is forcing wins/losses on people and the “50% winrate” isn’t just a byproduct of being placed with/against people of equal skill to your own.
If no hard proof can be provided (And frankly it hasnt) this thread basically just falls onto conspiratorial thinking and not keeping your ego in check.
1 Like
This is my concern right. If you get a fair game it’s still a 50/50 chance of winning, especially with 12 people being involved. I honestly don’t imagine it would look much different to a lot of people if it went one way or the other.
1 Like
It wouldn’t. People might win more when they’re climbing but you’ll peak eventually. I used to be a big Brawlhalla player and I was in the “top 500” by rating. Once I peaked I basically just entered a cycle of winning a few games and then losing a few. The same thing would happen in OW. You’ll reach the SR you deserve to be and you’ll just enter 50/50 matches anyways.
3 Likes
And for example, what would qualify as proof in this situation? Also Blizzard has already stated that they’re going for 50% odds that either team could win.
False. You’re implying that anything for which there is no “hard proof” is “basically conspiratorial” thinking. That’s just objectively false and your reasoning is on that is flawed.
And if you’ve got no “hard proof” that his criticisms are ego-driven, you’re just “falling into conspiratorial thinking.” …Your logic…
2 Likes
I mean I appreciate the effort here but he specifically mentioned there not being a meta if the game was balanced. So while you put a lot into that it doesn’t have much merit here
Well, I’m a bronze ranked player, and I’ll tell you right now that if MMR disappeared and I had to get slaughtered by play against gold+ ranked players almost every game, I’d simply stop playing.
I also don’t think my gold+ rated teammates would appreciate having me on their team dragging them down.
It’ll never happen. Big tech are some of the largest donors to politicians, who then put policies into law that favor big tech. It’s a whole club, and you aren’t in it.
It’s called corruption, and it runs much deeper in our governments than most people realize. It’s disgusting, both/all political parties do it, and there’s nothing you can do about it.
As for “forced 50% winrate”, you aren’t forced to win 50% of the time. The matchmaker makes matches that it thinks will be even and both teams have a 50% chance of winning. In order to climb (or fall), you have to prove it wrong. That’s how it works, you’re not going to be able to stay at the same skill level and rise or fall drastically.
2 Likes
Lol, ignore my response. I got confused about who you were responding to and therefore what you were asking.
Just something to think about. How does everyone feel about the content of this video?
3 Likes
I didn’t watch the video but I can give my opinion.
I think the algorithmic matchmaker that’s designed to make certain matches harder and easier is good…if the game wasn’t too team dependent to where your impact is at the mercy of your teammates doing well. If it was other games like Valorant, this system would work quite well, provided they don’t make the game unwinnable or unlosable, but only slightly harder to win or lose depending on your performance to keep you more engaged.
So on paper this system makes sense and I’m not against it. But when it’s implemented in a team based game that’s overly team dependent like OW1, I just don’t see it being a good idea. That video you’re referring to is talking about League of Legeneds as well and from my knowledge MOBA games in general are also heavily team dependent as well similarly to Overwatch, so I don’t see it working in those genre of games either.
This system would only make sense on games that’s more skill-based and less team dependent, like StarCraft 2, CS:GO, and Valorant. The more team dependent the game is, the more they should tone down on this handicapping system IMO. In summary, I think it’s okay to have it in Overwatch, but it might need to be toned down a notch as long as the game remains heavily team dependent.
2 Likes
Basically EA published a paper on “Engagement Optimized Matchmaking Frameworks” . EA only cares that you keep playing. If you lose too much , or even win too easily you will “churn” away from playing. They want matchmaking that has the lowest chance of doing that. They found exact combinations of W/L/D and the associated a specific “Churn Risk” percentage. And the lowest “Risk” was associated with a higher number of loses and draws.
- 3W/4L/5D = 2.6% Risk
- 3W/0L/0D = 3.7%Risk
- 2W/6L/2D =4.7% Risk
- 0W/3L/0D = 5.1% Risk
etc etcetc Whole chart.
Therefore…if the game predicts you will quit, it adjusts match odds to keep you playing. Giving you a near guaranteed win, draw, or even a loss. This prediction also takes other stats into account when calculating
- Install Date
- Play Frequency
- MMR
Their conclusion being that “matchmaking based on fairness is not optimal for engagement” And you know how much corpos love “engagement” these days 
5 Likes
There are some variables that can’t put into the W/L/D though, like the quality of individual matches. If I get stomped, I’m less likely to queue for another game because I would lose trust in the matchmaker being able to give me high quality games. But if the game was a really close one and I lose, I would be more keen to queue for another game because it was almost winnable.
If the engagement based matchmaker is accounting for the quality of individual matches, such as how long each match lasts for, then I could see it working. But if it’s just guessing the player’s retention based on win, loss and draw numbers, I’m not sure if that’s going to be that accurate in understanding the player’s needs.
1 Like
I’ve not seen the paper myself, only quotations from it. But seeing as install dates are among metrics, I wouldn’t be surprised if game length could be one of them. Me personally? I don’t want any of that. You have to fight an enemy, you have to fight your own team (toxic or otherwise) , and then you have to fight the system.
>https://imgur.com/eZKyAyy<
Now OW2 has a major PVE component. I would not be surprised if algorithms could be used here too. Therefore there might be safety as a 5 man squad.
3 Likes
Here’s the thing though (especially in lower ranks). People think that, because their team got ROFLSTOMP’d, its indicative of poor match making.
But when your team is running Hog/Ball as tanks, Junkrate/Symmetra for DPS, and Mercy/Lucio as supports, while the enemy team is running Rien/Zar for tanks, Cree/Mei for DPS, and Ana/Lucio for Supports, its pretty clear this isn’t a discussion of match making. At all, in fact.
Now, thinking of the above example, can a Team Composition comprised entirely of comfort pick heroes outplay a team composition built around internal synergies (that have been programmed into the game mind you)?
Yes!! They most certainly can.
But lets discuss the likelihood that players, on any given team in any random match who play comfort picks, are really going to get together and work to make their team composition work.
That is VERY unlikely. And we’ve ALL seen this fail time and time again.
I’d argue that most games (more than 70%) are actually fair and balanced. The determining factor ends up being incompatible/uncooperative teammates/team work.
And there is SIGNIFICANTLY more proof to suggest this purely based on how the game of Overwatch works.
- No conspiracy theories… Just game sense and good game play.
1 Like
They literally haven’t. I’ve been down OPs patent loophole before. The cited patents are either completely irrelevant to Overwatch or are completely irrelevant to competitive MMR - sometimes both.
3 Likes
Conspiracy theories have always existed but they are not 100% wrong either. Specially in F2P games, having people playing to have a good baseline to pick for the MM is a must so the EOMM is kind of a thing, but its not so prevalent and certainly not “evil”.
The issue with Solo Queue is real though. You are more “vulnerable” to the MM toying with you and putting you in places where it needs you to be, but you dont necessarily belong- The more RNG you are subjected to, the more you feel you dont control the outcome of the match and the more you suffer. Its not a secret that team games are incredibly different, experience and enjoyment wise, when you have a 3-4 stack to play with consistently.
2 Likes
There’s an issue with team queueing as well though that people looking for the “one weird trick” to fix their play experience often overlook- If you group with players who are lower skilled than you and/or lower skilled than their current rank, it will put downward pressure on your SR. There are two reasons this happens- if players are significantly lower skilled than you, they will decrease the average SR of the lobby you queue into, which means that you will get much less SR for a win and lose much more SR on a loss. Secondly, if you queue with players who are less skilled than their current rank, they will contribute less to your team than the matchmaker expects them too. This makes it more likely that your team will lose. Essentially, they are in the process of ranking down to their true rank and they will suffer losses as they do so, which means you will likely rank down as you queue with them.
So grouping can be a benefit in two main cases. If everyone in the group is roughly evenly skilled and there is good team synergy within the group then everyone in the group can benefit (although, even then, you will get harder matches in a group queue than in a solo queue as the matchmaker expects teams to play better than random groups). The second case is if someone of lower skill groups with someone of higher skill who can carry them. But in the third case where someone of higher skill groups with someone of lower skill, they will need to carry in order to win and even if they do carry they will get very little reward.
This can all be gamed (the higher skilled player can group with the lower skilled player suffering some SR loss even while the lower skilled player is able to rank up, for instance, and then the higher skilled player can solo queue to repair the damage to their SR), but it is not true that the grouped experience is beneficial while the solo queue experience is terrible.
I found it much easier to rank up from Silver to high Plat in solo queue, for instance, while I initially ranked down to Silver from high Gold playing in a group. Sometimes the people you are queueing with are detrimental to your play experience/SR. It all depends upon the specifics of the group and how comfortable someone is adjusting to random players in solo queue.
2 Likes
The cons of stacking are heavily lessened compared to solo queue dude. Even if you queue with people that are of less SR than you, you know their weaknesses, and if you are on discord, the comms are way better.
1 Like