Are lower rated players getting better?

The system in place gives players some say in how they are paired by limiting the range of ratings. Adding even more logic on top of that, to massage the pairings, will create fewer potential matches and likely increase paring waits while the system tries to find some perfect fit.

Add to that, one person's idea of a perfect fit won't end up matching another's. 

When you're losing, your rating drops, so the restrictions on rating range will keep you from playing much higher players. If you're on tilt, you'll drop down enough to get to a point where you're more likely to start winning some games again.

The same is true when you're winning, your range will include some higher rated players. If you have steaks of wins, you get to where most players will likely be above the rating you started the session with. 

Quite frankly, it doesn’t matter what you want. If you don’t want to face higher rated opponents when losing because you’re on a losing streak, well then, too bad. The current seek limitations are fair.

Bold of you to assume I drive 

I would say that players in the 1300-1500 range have definitely got better in the last couple of years.........

BUT also there are almost 10 million rapid rated players now, there used to be well under one million. In 2019 a 1300 rating would likely put you in the top 100,000.  I suspect that has more to do with including shorter games in rapid than covid or anything else.

I figured with practice, playing people close to my range, and analyzing my losses, I'd get better. I do not always want to only play people lower (maybe now I do - not that it would matter). I don't have high expectations, just maybe get progressively, slowly, better as I play more. 

The opposite is happening. It seems I was able to get in slightly more wins than losses, so I experienced a very gradual increase in score. Now I am on a negative trajectory, regardless of who I play.

I'll try tightening the matchup parameters. I currently have then set as +/- 100. At this point though, I go into games with a negative mindset, and it usually results in defeat and anger.

I like playing the game, but it is making me increasingly miserable. 

Yes. Most lower rated players will improve in some way. 

Is this different when you play on a mobile device?  I notice that almost all the time,i get pair up with an opponent whose elo is less than mine, sometimes lower than the 150 limit i had set.

That’s literally what the rating ranges in the settings are for but ok… if you just wanted to complain, go ahead

If you set a custom seek in the app, it should try pairing you in your chosen range, since it's the same live server process doing the pairings as far as I'm aware.

better players are getting lower rated.

You may not but a lot of other players would be upset if they had longer waits to get paired. There are a lot of competing interests and the site can't meet them all. Some players are going to be playing events, some only certain pools, and some in the random pool. That latter pool already has some restrictions when it comes to rating ranges, and creating more sub pools does not seem like it will improve things.

You can always make a suggestion to the site to add and optional pairing method, but I'm going to guess it's unlikely to happen. You can do that under Help > Make a Suggestion

i thin yes

think*

If you are on a losing streak just turn the rating cap down, and you will be paired up with lower rated people. 

Not really at your range of rating. Anyone below 1500 (which includes me) loses mostly because of hanging pieces or making serious tactical mistakes.

heyyy

Well, after each win you could set your seek range to the minimum less-than rating and the maximum more-than rating. That will maximize your chances of playing a higher rated player. And vice-versa after each loss.

But why are you so focused on the rating of your next opponent after one game? You don't become better just because you win a game. After one win or one loss, your chess skill stays the same. That is, although rating change over the long-term is skill related, there is a significant random driver on the short term (e.g., being especially alert, being on tilt, happening to play someone significantly under or over rated, you or your opponent dropping the internet connection and so on).

Also, within a certain range, rating difference is essentially meaningless for predicting a single game. For example, at a 50 point rating difference the expected score over the long run is 57-43. Over the span of a few games that rating difference won't be apparent from the results.