My friends, I have seen that Overwatch has enabled players to create their very own game mode. I like this idea, it’s a good step in the right direction, however now we need the ability to make our own matchmaker system. Our very own ELO system that correctly pairs teams up, a real ladder system that is not so confusing to people, one that is transparent so every person is to be knowing exactly what is happening.
I will be the first to design a correct matchmaker that accurately ranks people and all of this code can be open source and I will do it for free and ask for nothing.
It’s time we fix this game ourself my friends.
10 Likes
I’ve said it before and i’ll say it again
soo much of the game’s core design is about obfuscate
transparency is the only thing that will save this game
no more medal system players argue over this a lot citing it gives people reasons to be toxic
the truth to that is players are usually only toxic because of 2 reasons
a they don’t understand something and it frustrates them
b there just a standard troll on the web
i’d argue that overwatch’s core design does a lot to confuse, frustrate and absolutely nothing to encourage or educate
like the “TIPS” could be used to tell someone when they are hard countered or not being effective etc
just soo much work was obviously put into
purposley not telling teams the whole story
and it seems they wanted teammates to talk to each other about who had what etc
but that has just made others distrustful on ladder in general
and most don’t believe others when they say they have x medal
when a composition isn’t working etc
and then toxicity erupts and the game is lost
they need to have a team accessible scoreboard with atleast k/d, damage blocked, assists, healing, etc
like what pursuit.gg had
11 Likes
I agree my friend, we must have full transparency. There should be nothing to hide with these matchmaking mysteries. I am fearing that they have dug their hole too deep and now it is too late for them to be climbing out.
5 Likes
This would be fantastic. The game is too good to be hobbled by a bad matchmaker.
Related: Opaque Matchmaker: Don't legitimate electronic sports need transparency?
3 Likes
The one thing that would be interesting is how many people are currently playing while you’re playing. SR will be harder to obtain in certain elos with less people on. You end up facing the same teams and 6 stacks end up facing non 6 stack teams. Tho, doing so would also cause problems as well. You’ll notice that during double XP weekend, ranked games were easier for us that continuously play, and more difficult after it was over. Same when new events start, new heroes are released, or new maps.
Basically, any information you gain to help you go up in SR, aside from learning from your own mistakes, will cause problems with the game itself. You’ll be able to pin point the person doing a crappy job, cause them to quit playing as an example by being toxic to them.
That said, I would love to know more about my own stats. Damage taken, healing received, etc… after the game is done so I can improve.
1 Like
it’s very important that participants are aware of the rules
Nerds sitting at home thinking they can design a better system, forgetting literally every comp game on the planet uses the same sort of setup. You guys are better than every gaming companys employees on the planet?
HAHAHAHAHAHAHAHA
haha
ha
ahhhhhhhhh
Quit your job and go work in the gaming industry. You will make a killing if you are that good.
7 Likes
Exactly how does it follow that if you’re great at making matchmakers that you’d be great at making games?
You don’t think they have people specifically hired to create the matchmakers?
lewl.
1 Like
Oh my, what a good point you have made. Your logic blows my whole theory out of the water, thank you for pointing that out.
I like the idea of more transparent data. Maybe for example, metrics of who is doing the most damage to you in a match, as well as universally. We all know Phara’s counters are hitscans, but if we see that soldier is 45% of the damage we take on that char, mcree 25%, snipers 25%, etc., then we have better data to understand more of what’s going on.
TBH man people can’t group up, let alone use data to improve in real time. All it will do is cause people to be toxic towards said Mcree who did 25% and before you know it you have rage quitters.
I do however think they could implement a post match data screen for you so you can properly see more data on how you did personally. That’s all that matters really, your performance in the role you are playing. What the team does doesn’t really matter.
1 Like
Great idea. Seams like they wanted us to fix it by our selfs because they didnt fix the problem they just gave us the tools and said code it ya dumb miners
Care to elaborate? how would it work ?
the matchmaker is flawed because it is all based of the bell curve assumption of skill distribution. if they instead built off a power law distribution curve assumption it would create a much more accurate matchmaker, however they are also constrained by queue times.
basically this game will not have a decent matchmaker because its built off an inferior distribution theory and they prioritise short queue times over fair matches.
1 Like
Interesting idea, but no distribution will be accurate when AI is judging skill based on contrived performance metrics. And why is any kind of assumption of distribution needed at all?
1 Like
Yes my friends the bell curve distribution is not making sense here, there is no need for any type of distribution whatsoever. This is what I have been trying to tell these people at blizzard is that the matchmaker must use LESS ASSUMPTIONS of skill and distribution. The players using the ladder system will create a distribution naturally but this distribution should not be used to create a ladder.
With game playing matchmaking less intervention is more. I think this matchmaker could be better at forecasting burger sales for McDonald, it is not good at estimating player skill.
3 Likes
They do not obviously. There is a reason ESEA and faceit are making bank on simply providing a MM system, so much they even afford to host their own lan-tournaments and hand out cash prizes, quite insane really. All that is possible ONLY cause the service provided by the owners of the IP;s AKA game developers are so crappy. Think about that for a second, when your service is so bad a competitor can take up more than 50% of the market shares and make bank.
Provide the source code, upload it on github or something.
I’d love to take a look at that, rofl.
I don’t believe that one single individual will be able to develop a matchmaking of this magnitude on his/her own.
I have already started to do this myself and have invested a lot of time to run simulations and possibly learn from them.
There are an incredible number of things that need to be considered when creating matches.
The matchmaker is required to do significantly more than bringing together players with the same rating.
It would not be so trivial to develop a player interface for matchmaking. Similar to the workshop. I don’t see any real benefit in it either.
The game developer should set the rules and make them as fair as possible. BUT I don’t think these rules should be kept secret under any circumstances. 