How can I get out of the 1800 range rating in chess?

@ThunderClap said in #19:

STUDY is what he is asking about' NOT ONLY PLAYING ... So at least try to name RESOURCES like Chess Videos Chess Mags Chess Books Chjess Websites that you can praxtice on ... At Least Try Everyone I will have to see which Resources this 1800 has used already & he seems to be doing well anyway with his Lichess Rating but is asking which Materials to Use Like a new Chess book That just came out on It's A Matter Of Technique is the title a nice large book on a good subject matter @Rey0825

Keeping in line with that thought, one of the most useful books I know is: Silman's Complete Endgame Course. Much recommended.

@Molurus said in #8:

Glicko and elo are different methods of rating calculation, but what you're saying here is still complete nonsense.

You can compare chess.com and FIDE elo ratings with Lichess ratings and there is almost consistently across the board a +300 / +400 difference in rating. Lichess ratings are inflated in that way.

I think the notion that player pools are incomparable is inherently flawed.

My hypothesis is you only need a certain % of shared players throughout player pools before you can start building correlations, which there is certainly a high % of players shared between Lichess, Chess.com, and FIDE.

These players who are shared between player pools establish baselines for comparisons across rating systems, since you can compare these same players' ratings across the different pools / rating systems.

While my rating and playing strength has substantially improved thanks to bullet, I'd probably be a stronger player if I played longer games as well.

I believe in playing a variety of time controls if you want to improve sincerely. Bullet teaches rapid tactical and pattern recognition. Longer time controls, deeper thinking and calculation of novel lines.

People act as though bullet makes you worse or something. I don't believe this is strictly true. It only makes you worse if you let it do so.

@thefrickouttaherelol said in #22:

You can compare chess.com and FIDE elo ratings with Lichess ratings and there is almost consistently across the board a +300 / +400 difference in rating. Lichess ratings are inflated in that way.

This difference between Lichess and chess com has nothing to do with the rating system used. Heck, both use Glicko.

It's also a little bit strange to call this 'inflation'. You might as well call chess com ratings 'deflated' and it would make equally little sense. Average ratings haven't actually changed on either site, and there is no standard for what is 'normal'.

I think the notion that player pools are incomparable is inherently flawed.

It's statistical comparisons only. You can't compare them in the way you can compare (say) meters to feet. For that they should be measuring the same thing, and they don't. (Heck, not even the rating categories are the same. 'Blitz' means something different here than it does on chess com. But that's really besides the point, something that makes the comparison even stranger.)

If you say that you are 6 foot tall, I could infer from that that you are 1.8288 meters tall, and I would be completely accurate. And that's because they are different measures of exactly the same thing: length.

If you say that your chess com rating is 1600, infering from that that your Lichess rating is 1950 is complete nonsense. That might be off as much as 200-300 points, which is really worse than guessing. Even if you disregard that you neglected to specify which rating category. (These diferences have NOTHING to do with the sites. They're differences between specific rating pools. There are huge differences between, say, Lichess blitz and Lichess rapid as well. By your reasoning, they should be the same across the board.)

And that's because, unlike what people believe, these ratings don't measure the same thing. The comparison is valid on average only. It makes no sense in individual cases.

That's the problem with statistics. People tend to apply them to individual cases, and that definitely is flawed. Heck, even scientists make this mistake way too often.

The easiest way to get out of the 1800 range is to lose a heap of games :)