Flaw in the rating system

Chess.com Uses the glicko-1 rating system, and don't get me wrong - It's great. It accounts for Your rating, your opponent’s rating, and rating volatility.

Rating volatility refers to how ‘reliable’ your rating is. For example, If it is a new account, or an account that has not played in a long time, their rating would be less reliable. Hence, the rating swings are higher so that the player’s rating adapts to their current strength. However, there is something that rating volatility doesn't account for. Current form.

I have played multiple accounts that are :

A. Much higher rated in another time control

B. On a tilt , but are back to their current strength

C. In very good form (A win streak for example)

D. A new account that it's rating hasn't stabilized.

This means that your opponent's playing strength is currently higher than their rating, and as we said, The glicko-1 rating system accounts for A. Your rating B.Your opponent's rating - This accounts for their rating, not their strength. That means if you're playing someone equal to your level in terms of rating, not strength, you will lose the same amount of rating as if you're playing someone who is your level, but actually is playing like 200 elo above you. Now, that obviously isn't fair.

So, a better way to do it would be to account for:

A. Ratings In other time controls but IF AND ONLY IF they haven't played the time control they are currently playing in a long time

B.move accuracy

C. Current form

You just want Glicko-2?

Yes, I feel that rating volatility is very important

A - you are rated at your level for a reason. If you play someone in blitz who is 400 points higher in rapid, assuming no cheating or sandbagging, there's nothing wrong with that. He's rated the same as you in blitz for a reason.

B - that's too bad man

C - that's too bad man

D - what?

Your current form is always going to vary. Some people are streaky while others are more consistent with results. I do not understand why you feel like this should be accounted for.

If your form is good, your rating should go up. The goal isn't to keep everyone at the same rating.

Furthermore, there's no way of predicting current form. You can try, by taking data such as most recent win streak, how long each game went on for, and so on. But we end up taking past data to make current matchmaking decisions ... Wait, that sounds familiar. Isn't that how the rating system works? I mean, you play games, then your rating changes based on the results... Your past games influence how your current self will be matched ...

You may open the superb support bot, type "I want to suggest a feature" and describe your feeling