Is the ranking system really that accurate

I have started playing around 1800 SR which is pretty bad indeed. Over the course of a few months I have managed to climb up to around 2480.

Then I have bought a brand new account and it rated me 2710 for the first time I was playing on it. Next season has came I got placed 2830. Since then I am swinging back and forth between the 2600s and 2800s.

It really feels like the game is trying to force you and keep you in the rank you have been first placed in. It does not really let you go anywhere unless you are overwhelmingly good or bad.
If you got better over the time the only way to get your SR to somewhat represent your current skill is to buy a new account.

Why did it place me far higher on the new account?
Let’s say I have got boosted (mercy main btw.) then why am I able to maintain my SR between 2600-2800? It’s 2724 in this very moment.
So which rank do I really belong to? Because either I am worse or better but I do not think 2724 is that accurate representation of how good I am at the game.

Its not. Way too many factors destabilize it. Group mechanics allow golds to play in diamond games. Smurfing allow GM players to play in bronze. You’ll rarely get a fair match and even then it turns out the match is not fair because the aforementioned problems skewed the rating of the participants.

A good working ranking system should have let me go all the way up to 2800 on my very first account yet it told me I am only gold.

If I am only gold, well then again a good ranking system should force me down on the new account.

Either way it’s messed up.

Exactly. +that’s actually shows how random it at least is.

The more matches you play, the more certain the system will be of your MMR. If you stay in the same range, it’s because you’re not getting better.

One common misconception is that being “good” should allow you to climb. The fact is you need to get BETTER to climb. If your skill level doesn’t increase over time, the only thing that happens is that the game’s certainty of your MMR increases.

If the system was so certain of me being a gold player, then why did it rate me 2700+ on my new account?

Speaking of certainty, isn’t it that it just:

You got me wrong I think. It’s not that I want to rank up. It’s that it feels like it’s putting me in random ranks and tries to keep me in it.
I have never managed to get out of gold on my first account but I have done it several times on my alts and my main got placed higher initially.

So which rank do I really belong to if I have a 2400 an 1800 and a 2600-2800 rated account at the same time? I myself do not really know where should I be playing and the whole thing feels really inconsistent.

1 Like

Your SR as newly placed is based on 10 matches with large changes in opposing team SR. It’s highly uncertain.

If you have played a lot of matches on your 2400 account, that SR is likely closer to your true SR.

Nothing is trying to keep you in place. What happens is that the game tries to match you against opponents with an SR it believes to be the same as what it believes yours is. If you beat that opponent, it thinks it had underestimated your skill, and overestimated the opponents. So it will increase yours, and decrease the opponent’s. And vice versa, of course. As you keep playing, this will continue happening, and as long as you beat opponents it thougt were at your skill level, it will adjust what it thinks you are at. When you start having a 50% winrate, your actual skill (SR) has converged with your actual skill, and in order to increase your SR (which is just a reflection of your skill), you need to improve your actual skill.

This is what has happened. Over the long term, you are at a 50% winrate, meaning your skill isn’t improving. To change this, you need to structure your practice and not just keep playing the same way. Few people actually do this, which is why a lot of people believe the match maker is “holding them down.” They simply aren’t improving.

As to which account has the most accurate SR; which one have you played on the most? Also keep in mind that the skills required are different in the different ranks. In bronze and silver your individual performance is the most important, as people there hasn’t learned teamwork. In plat and up, teamwork starts getting very important, and once you hit diamond you will need to master positioning and awareness, on top of teamwork and mechanical skill, to keep climbing.

It’s not the game trying to keep you back, it’s simply that you are not improving enough to climb, in the right areas.

The best way to make the ranking system more accurate is to remove seasons and make the matchmaker random. Not truly random mind you, it’d still match people of similar SR but it wont try to balance it. It’ll be a bit of a chaos but eventually it’ll balance out.

Trying to make a 50-50 is a truly doomed endeavor.

50% chance of winning isn’t a hard requirement, and if you get a match with an estimated 45% chance, your SR will increase more if you win.

This how it supposed to work. I have around 400-500 hrs on all my accounts. They are all pretty close to 50% winrates. The only thing that varies is that one is placed at 1800, one is around 2400, and this one is swinging back and forth between 2600 and 2800. I believe my true skill has to be around this range.
I mostly play the same heroes on all of them.

It just doesn’t make much sense when I am silver on an account but plat on an other.

Do you have 400-500 hours of competitive for each account? Or all game modes? Only competitive will affect your MMR.

It is possible that your personal ability is worse than your teamwork, which on certain heroes (several supports, rein, etc) would keep you from climbing.

The ranking system is broken as an empty egg head. After competting the last 3 rounds, 1 win and 1 lost, the third and last game for ranking position the player on the team left and after 2 min the round was over but it indicate a unfair lost. How is this system working with this kind of leavers/bad players among us all players? Its like to burn your game code on this game

Next time the leaver will be on the enemy team, and it’ll even out. That’s the point everyone fails to grasp; statistics require a large number of samples. You can’t just look at one game, or three, and say “but I lose 66%, and have a leaver in a third of my matches.”

Play 30-50 matches, and let me know what your win% and leavers/match for both teams are. You’ll see that the leaver stat is equal, and that the win/loss trend is is fairly consistent.

A 2400 sr and a 2700 sr match are pretty much the same

Yes, as the difference in team SR is small, and therefore the average actual skill is basically the same in both matches.

2400 and 2600 are basically the same thing