Haha!! Not embarrassing at all actually 
it’s been a while since I’ve seen you on here I think 
It’s also possible I’m just not observant. 
Anyways hope you’re doing well!! 
2 Likes
Yeah, I just don’t find the forums particularly valuable anymore. It feels like I’ve had every discussion before, like 90% of posts are hate on my mains, my role, rants or weird conspiracy theories like this one. And I’ve just found better use for my time.
Thank you, you too.
3 Likes
I can appreciate that, I’ve found the Reddit community to be much healthier and less likely to blame their teammates and the boogie man.
While Overwatch is definitely in a tumultuous state, the competitive community that has a genuine interest in enjoying the game and improving as players is very much alive.
Just not here on the forums unfortunately.
1 Like
Reddit is better, and I do spend some time there, but there’s still a lot of “hardstuck” rants, albeit disguised better, and unhelpful advice. I’ve mostly moved on to Discord’s of coaches, if at all. That can be a circlej erk for sure, but it’s better than rants and bad advice. And, as I said, it takes up way less time than it used to anyway.
3 Likes
I see them on the Valorant Reddit too.
smurfs need to be banned
cant climb because system is rigged
can’t climb because match maker
Blah blah blah blah
So many entitled children in competitive gaming these days.
Everyone thinks they’re Super, JJonak, TenZ, and BabyBay….
Kinda unbelievable.
But it’s what it is.
My org still does some Gold/Plat coaching but I’ve stepped away from Overwatch for the most part and am doing more with our Valorant team since we hired a high level coach and manager for our GM team.
I’d like to hit Diamond 2 in Valorant and probably move on to coaching there unless Overwatch 2 is as good as we’re all hoping.
Yee Yee!!
Super glad to see you around again!!
2 Likes
If you’re rigging the contest odds in advace, there is nothing to compete on. It’sa rigged contest. Competition is about qualifying into a particular pool, and trying your luck against random samples from that pool. If you rise against a particular average representation of that pool, you move to a new pool. There is no need for state or memory or adaptation. Real sport doesn’t work like that.
1 Like
Fair odds is the opposite of a “rigged” outcome.
Yes, that is what competition is about–playing against opponents in your pool. How can we tell if two people are in the same pool?
2 Likes
It’s rigging for fairness. Which is a subset of rigging. As per definition.
The point of fair competition is to not force the odds in advance. Your coins analogy is basically arguing in favour of random matchmaking.
Placements? Win the first random match, win the second random match for/against winners of the first random match, etc. Quite quickly the higher skill players rise up and the lower ones drop down and you continuously refine the range or bracket from which you draw your random matches from.
I showed you all these already and Receipts has entire books on the subject if you dig around. I suggest you come up to speed on the topic and questions that aren’t already answered, or refute key areas of the framework on the basis of better math if you can find it.
1 Like
Is a GM chess player participating in a FIDE tournament 25 elo from her rank “a rigged tournament?”
The point of fair competition is to play against people at your skill level.
2 Likes
By statistical analysis which results in an individual MMR of course!! 
I suppose we could use an individual number like SR but, wouldn’t that mean it’s easier to disguise your true performance and make identification of smurfing much more difficult?
Hmm….
2 Likes
Is a regular season game of overwatch league like Shanghai dragons versus London Spitfire unfair?
Are you asking if hand-picking professional players for a team following OWL regulations is fair? Sure.
Is hand-picking a professional team analogous to ladder match making? No.
3 Likes
Hey, just leaving this here if you guys want to experiment with simulating open queue-esque matchmaking! Hope you guys find it useful!
2 Likes
Not even a little bit…
Hey Nano!! 
I’ve been following your conversation on that post a little bit. I really don’t want to support his post though because the game isn’t rigged.
I am interested in your program though, and as an engineer would love to try it out, and discuss the observed results.
Can you tell me a little bit about it?
2 Likes
It’s a fairly simple and straightforward tool that simulates matching teammates for an OW game.
I wrote a lot on it in the quoted post above, such as how it approaches matchmaking, counter-picking, and how it models the selection of players with the simulated results being the assumed best possible matchup. This means that the results “would be” the game you queue into.
So far, it simulates in the fashion of open queue. So there’s no function of roles in the sim, however I do think it does its job well enough!
The motivation for making this tool was that there were far too many nuances and ideas on these forums to discuss adequately without triggering someone or causing them to troll post. So I decided to bundle up my understandings into code, which you can also tinker around with!
For now, it exists as a spreadsheet and graph. But later I do intend to move the code over to github.io
1 Like
I dig that, and I can most certainly appreciate the effort and I’ll gladly take a look once I get some better internet (stuck in a hotel rn 
).
Does your program attempt to create a 50/50 win chance or does it force 50% win rate?
As you mentioned in a different thread these are vastly different concepts and implementation would (clearly) be very different.
Also, why did you decide to write an open queue model? I feel like a Role Queue model would be simpler.
Definitely less variation in player performance anyways. 
At least that’s my initial reaction without attempting to write one myself.
1 Like
It doesn’t attempt to do anything outside of creating a match with players of as similar skill as possible, with as even odds as possible. As a result, the odds are reflective of each team’s overall estimated skill. It’s based on local MMR, which is the numerical value which represents player skill (which is an abstracted concept of what may be used in actual MMR calculations) which is used to create “even” teams.
Most of these questions I believe I provided some answer to in my original post. 
My intention there is to distill the core question from the fat. The question then becomes a matter of whether players are deliberately placed into matches that algorithmically favor a win or a loss, dependent upon a player’s subsequent win or loss streak or other performant stat (if otherwise), which can then officially be responded to with a simple “yes” or “no.”
The model I created answers a very different question, however the answer it provides allows us to see the numerical variance that exists within any matchmaking system. Which is also why I question why we don’t have more direct numbers in the interface.
Likely for very similar reasons that open queue was initially developed, and role queue was added in post.
Adding more sorting criteria, such as pairing 4 players in the same role, into a “fair” game with 8 other players who may or may not (ambiguity because of the lack of transparency in the current ranking system) be at the same skill as those 4 players in one role. Then to do this 2 more times for each role, to then find 3 groups of 4 players who can be counter-matched to create a game with 50/50 odds.
This is both algorithmically and computationally far more complex than the simple open queue model.
In addition, this will also require a different presentation model which requires more logistics data that needs to be exposed, but I think it’ll get to the point it’s just a wall of numbers and stats.
For example, one such scenario which shows a list of hundreds of potential players, marking which ones were selected and for which role, then again for each “pool” of roles, and then to demonstrate why any particular selection of players from each role would create a fair match with 50/50 odds.
All of this, and in the end the results would look basically same as the open queue model, possibly with more variance between the role groups. That’s just my perspective on it though.
It’s derived from a knapsack algo, but I implore you to explore the concept of matchmaking yourself!
Numerically, the tool demonstrates what I think are a great deal of player experiences. It shows why “smurfs” appear, and why some players underperform.
The best common ground we can possibly have between the two camps is to look at the numbers, and so far I think this is the best (attempt) we currently have.
At the very least, those numbers are easy to read!
Haha, well I think it’s possible these forums consume more resources than the spreadsheet!!
Probably….
But….
A. I’m relaxing
B. I had some general questions
C. I really don’t support adding comments to cutberts thread as I have no desire to draw attention to a thread that could potentially misguide new players to the forum.
Edit: internet seems to be working better now, I may be inclined to peruse 
1 Like
Ehhh, I can almost guarantee it’s easier to check it out!
Fair enough, but I do think I provided the answer to some of your questions in my original post.
I’m really impartial, because the issue as of yet has not been officially responded to.
All I want is for us to have as productive of conversations about it as possible!
Much of the infighting here is from a lack of understanding of core concepts. Hopefully I can help alleviate some confusions by presenting impartial data.
1 Like
Toxic players in almost every game. I mute now 90% of the games I play.