There wasn’t anything hidden… it was right out there, from the very first. You just chose to ignore it an attempt to subvert the point I made…the example I gave was perfectly clear, based on facts, and simply illustrated what the consequences of having hidden factors weighting the results of matches can be over time - it gets much worse as the number of matches run increase.
Good to see you acknowledge that my point was clear in the first place, it would’ve been much simpler to just do that rather than claiming I was “disingenuous”. At any rate… if they really wanted it to be as you suggested just a weighting based on the relative ranks of the two teams, they could easily just show a simple number ‘Team A has +3 point advantage over Team B’ - instead of having a hidden complex metric. But they don’t, because that isn’t how it works.
And no - I’m not trolling. Trolling would be saying something like this:
The best mmr system for a video game developer would be one that adjusts rewards to maximize playtime and money spent by the individual player. It would try to place players in ranks where they will spend the most cash on the game…and then hide the details from the player so they don’t really know that’s what they’re doing.
They dont want us to know cuz its either basic or stupid and they dont want us to criticize them for it
1 Like
Well it’s called hidden MMR for a reason. I believe Jeff put it in layman’s terms that “everyone is assigned a hidden number” and that’s how you get matched with other players. Players with similar numbers.
MMR is your current ranking in the system. The number you can see. That is determined by match stats, who you play against, win or lose, etc
I’ve written some here regarding the algorithm itself (link actually has two versions of the explanation, 4 years apart from different people, as they’re both true they’re just saying the same thing in different ways).
Kaawumba has a good explanation of it here. This should be your primary source.
Don’t listen to most people. The discussions have gone so far off the rails that even people that are ok with how the system works still don’t really understand how it works.
The most common mistake is the false belief that SR is used in matchmaking at all, but they’ve said repeatedly and unequivocally that it is not used and never has been used.
They’ve been very clear from the outset that it does not use SR in matchmaking. Link in the above post.
Though, to be fair, I think some of the confusion lies in how “THE EVIL OF MMR” is generally portrayed on these forums, because I see a lot of cases where people would be 100% accurate if they just swapped the terms SR for MMR, but they think MMR is something else entirely that doesn’t even exist.
kill participation and kdr are the most important factors it seems.
Deaths are especially highly penalised
winning is also very influential
1 Like
It makes sense. If you have good game sense or good mechanical skills or both, you are most likely killing many more than you dying.
Especially high solo kill number would skyrocket you out of an elo in no time. That s how they combat smurfs.
1 Like
Want to know my hidden MMR is for QP because always getting paired with and against Mid Diamond- Mid Masters players recently. It’s just pain
Not sure about death and KDR. I’ve already seen smurfs, chain killed but positioned in frontline and all the time in fire.
I had a decent KDR for seasons, around 5 or 6, but still stuck. Another player on the forum was even at 9-10, it was the same.
Solo kills make probaby more sens. It hard to have a good amount of them, if your team is effective, even a Moira can deny you to get them. i’ve a low ratio of them. I’m more a finisher than a killer.
The game doesnt track where we stand, “bad positioning” is standing somewhere that gets you killed before you can complete your mission. For smurfs, frontline is good because they can kill before their team gets killed.
If you have high KDR, maybe play a bit more aggressive? Your MMR may go up by itself but you still need to win to move SR.
I know a lot of people here are guessing but the honest answer is that we do not know the formula that goes into MMR. We know that it compares you against people of similar MMR and adjusts based on your performance and we know some small adjustments that have been made over time (for example the early version of MMR had a bad habit of not giving healers enough MMR for dominant wins because you simply had less healing to do).
The reason though that we do not know can go back to the early seasons of Overwatch and everyone’s favorite hippocratic pacifist (props to those who remember that reference) Mercy. Back in season 4 people figured out part of the MMR calculation for her and so what people would do is get her ult and then hide and wait for a 4 or 5 man mass res and only use it for that. Despite the fact that these players were winning maybe 40% of the time they were gaining SR because the system saw their resurrections and thought that they were outperforming their rank. This actually caused a lot of problems once fixed as you had a bunch of players in Masters+ who were nowhere near that good and they fell like a brick once that was fixed but it hurt game quality for a time.
nobody really knows except blizzard. and its in blizzard best interest that nobody knows so theres no conscious exploitation of the system
I play very safe indeed.
During a rare mirror game, i could compare my ult charging rate with the one of the ennemy Ashe. She was playing agressively, i’ve personnaly killed her 3 or 4 times while i stayed alive but we got our Bob, in the same moment, three times in a row. The last one, she got him even earlier… She was charging her ult faster than me, even dying a lot. She was just spamming her shots in frontline. No positioning…
I’ve stopped to record my Sr gains a long time ago, it was depressing, but i’m actually really winning more than i loose. For the last 5 games, 3L-2W and -13 sr. It’s really slow to climb even with a good WR and time consuming. I play 2-3 ranked games per day after 10 minutes of warm up with aimlab, a custom game and a quick play.
If you are killing their Ashe but she got Bob at the same rate, your Tanks are just feeding way too hard. There s hardly anything you can do about that.
Back then the game purposedly put me into lower ranks, like 500 SR below my average. It was still very hard and slow to climb back up for me. At some point i stopped worrying too much and just treat it as QP so that i could have fun again. The climb was just a bonus.