Checking if Elo system is oppressive [With proofs]

It seems there’s been a misunderstanding. I wasn’t suggesting that a player's country of origin determines their inherent skill—skill is developed individually and isn’t tied to nationality. What I’m referring to is localized rating inflation, which can happen due to factors like time zones and regional player pools.

When you play on a platform like Chess.com, you’re often matched with players based on who’s online at the same time as you. For instance, if you're consistently playing during certain hours, you might predominantly face players from specific regions or countries that share your time zone or general availability. These regional player pools can develop their own "ecosystem" of ratings, particularly if players in that pool don’t regularly compete with people outside of it. As a result, there can be rating inflations or deflations relative to the global player base.

For example, you might notice that in some time zones or regions, players appear stronger or weaker on average within a particular rating band. This doesn’t mean people from that country are inherently better or worse, but it does suggest localized rating distortions due to the players you’re more likely to be matched with.

So, to clarify, I’m not making a generalization about a country’s skill level. Instead, I’m pointing out that the patterns of match frequency across time zones can influence the rating system in ways that make it appear some regions are stronger or weaker than others. It’s an observation about how the player base interacts across the system, not a judgment on skill.

Is anyone then able to imagine what chess playing elo "strength" would look like if all chess players actually played for the checkmate win result more than say for elo rating, or chess titles, etc....?Or am I apparently the only one here that plays chess games for the checkmate win result more than anything else when it comes to playing chess?

IndianCamels would disagree. But you aren't wrong. The environment in which someone learns and plays chess can have a big impact on their development. In countries where chess is more popular or where there are strong chess traditions, players often have better access to resources like chess clubs, tournaments, coaches, and even stronger competition. Regular exposure to high-quality practice and strong opponents can naturally lead to faster improvement.

People do play for the win result mostly. But to track your own progress and have fair matchmaking it's important to have a fair rating system.

But heres the thing: After 400 ratings match up by skill in that specific time control.

Also note that the med. rating is 620, so youre making your case for a small minority

Minority? There are 3.3 million Blitz players rated between 100 and 300, with 1 million players stuck at the minimum rating of 100. Do you really believe that every single player rated 100 has exactly the same skill level? The system clearly struggles to differentiate between players at the lower end, leading to an artificial bottleneck. What’s the point of defending this? Instead of justifying a flawed system, we should acknowledge its limitations and work towards a more accurate way of reflecting player skill, especially in the lower ranges. Are you arguing just for the sake of it, or do you believe this system really works as intended?

And I was taught growing up lessons like, "sometimes you just have to go back to the drawing board" quotes and such, but I know and understand that it is extremely difficult for the chess.com staff to perform their professional responsibilities and take on the "HERCULEAN" task of attempting to do their best to provide a service and actually take on the "ULTIMATE WARRIOR" like task of trying to please the chess.com chess playing community and all, but people weren't taught other important life lessons that sometimes as a community the chaff needs to separated from the wheat and be thrown into a huge pile and set aflame. Proverbially speaking that is.... meaning that a community needs to perform the task of caring fore their community while at the same time, a community also polices their own as well. I also understand that all of us aren't perfect, but it is still not a good reason to strive to better everyday than how you were the day before either.

When there a 1billion accounts, 0.3 percents is a minority

Lying isn't nice.

Players: 29,470,758

And you suggest it's ok to oppress a minority just because it's a minority?

Ok still. I was talking about rapid pool.

Lie again. Rapid pool has 78 mil players with 9.8 mil rated 100-300. 2.1 mil of them are 100.

A minority is any group under 50%

The median rating is 400

You could say that everyone sucks. But differently. While unfair rating system erases and randomly shuffles those differences. It’s important to be constructive when discussing skill and refrain from dismissive terms like "suck", as we all start somewhere, and a fair system should aim to represent every player’s actual ability as accurately as possible.

@IndianCamels, calling millions of players a "minority" to dismiss their experience is both silly and disrespectful. Stop repeating this.

What makes you think that? The fact that millions of players with different skill levels are squeezed into a narrow, chaotic range of ratings suggests otherwise. It’s worth considering that while the system probably works fine at higher levels, there are significant issues lower down.

100 is 100 anywhere. In Elo difference in 100 means 35.9935% winning chance for the weaker player. At any range of ratings. 400vs500, 1000vs1100, 3000vs3100.
Here, at low Elos, ratings are meaningless (some 100s could beat some 200s) but not because it's impossible to measure but because of the flaws of the system such as low starting rating, rating floor, uncertainty etc. System basically refuses to measure them.

I think it would be beneficial for everyone to do their homework on the math and mechanics behind rating systems like Elo before posting opinions here. Let’s base our opinions on a solid understanding of the topic to make the conversation more productive.

Your model isn't normal...