if they ‘can’t rise’ then that is the limit of their ‘skill’ and that is the purpose of MMR: to find the skill range and match people accordingly. However, the number and metric are arbitrary and tend to cater to a poor understanding of what ‘skill’ entails and only serves to gratify people with something shiny. If they understand how to improve, then the number is mostly superficial.
People tend to think they’ll ‘grind out’ an mmr climb the say way they’d get gear from Wow/Diablo and hope time turns into stat boosts to compensate for limited strategy, performance, and cooperation. Part of the issue of the playerbase is that basic gameplay – soaking experience, not dying, and aiming skillshots – are effectively “cheating” to the average player. Simply playing one game after another in a “fair match” does not mean that they can, or will climb. Since the match tries to create a match where both sides have an “equal chance” to win, what commonly happens is that a player will start at the low end of an averaging block and get carried for a few games. When the MMR rises, they’re not on the upper portion of the same average and have players below their rating to create the same average. However, since their ‘skill’ didn’t actually increase, they do not fill the same role of increased responsibility, lose some games, and stay around the same area. Instead of reflecting on what they can do better, they look for something else to fault instead.
“oh, I’m a gold but I’m matched against a plat”
Outside of boosting smurfs, most of the skill divergence between non-GM ranks is all over the place: the game does not reward specific mechanical prowess, and it doesn’t teach people the ‘correct play’, so most players are in the same mess making the same mistakes and just looking for something to blame instead. Usual ‘climbs’ just have people party together so there’s more likely to be better cooperation on that side: two people moving together is that much harder to gank, and that much easier to contest an objective. However, it doesn’t take a party for a player to learn to coordinate with another.
Regardless of most individual ‘skills’, team cooperation tends to be more telling of game success as a typical player will either engage one-at-a-time, or run away while one ally takes all the damage.
2000 damage on 1 hero kills a lot of the roster, but that same 2000 split across 2-3 heroes becomes something that can be healed up and seem like damage wasn’t dealt at all; so a common reaction is that people’s perspectives oscillate between ‘no damage’ and “zomg op”, so having a buddy being able to step in and soak or two, stand ground and deal damage, tends to do more toward winning a game (climbing) then someone playing out hundreds of games at bronze, silver, or whereever.
So long as people don’t notice those little things they’ll do for a friend, and not do for an ally, then they’re not going to climb the ranks regardless of whatever color border or plaque is attached to their avatar. So long as a matching system does not enforce specific skill expectations on their metric, then there aren’t any expectations to have on what separates one rank from another.
What people can do is put in that extra bit of gameplay awareness and actually be the improvement they expect to actually climb the ranks, and not assume this is some Baal Run or Arthas raid and they’ll eventually grind out the loot/wins. The ‘random’ matching is just math, and the metrics are what they are. What is ‘broken’ is how people avoid doing the processes that get them better at the game, and then wonder why they didn’t climb.
While an mmr can be higher for one player, the ‘skill’ weighted by the system is spread so thin that any number of variables could be throw out the effective value of that number. However, if people are fixated on something being more shiny than their own, then they’re likely already coming up with excuses to not perform to the weight of their own mmr, so of course they’re going to ‘not climb’.