Why is my rating so low?

Bro this thing is Fr rigged.

uh okay cool? How old r y'all anyways?

I am 11 right now hoping to be better some day

I mean 12 years old sorry

How many paragraphs do you want???

Calm down. You dropped your title as if that explained anything. All you said was "I'm a candidate master but I can't rationalize my position. I'm a CM tho."

I commented that because I wanted a candidate master to rationalize better as to why playing bots is such a bad idea. In my opinion, bots perform more consistently so I see how playing against them could help me improve.

I'd appreciate it if you helped me understand instead because I really wonder which is better too so I wanted to spark a healthy discussion but you opted to have it that way. Ok.

If your learning new things you will do worse at first and lose more games so it looks like you don't improve even if you are then when you get he right idea you start winning more (more towards middle games and end games openings are easier to understand even for 600s sometimes)

Watching other people helps you realize how you should move the pieces but it doesn't teach you how or why or the ideas

So focus on trying to come up with new strategies or ways you can beat your opponents if something's not working you either need more practice or your doing something wrong and need to gain more understanding or figure out what your doing wrong

A good way is to analyze your games and try to figure out how/why your doing things wrong

That's the problem consisistancy is more bad you want to try to predict your opponents moves even if they are different

Tbh I wouldn't recommend chess.com bots if you want to do bots you should try to play against stockfish. Bots arnt a great idea because ethe blunders are completely random humans aren't so random which makes them harder to play against (up to a certain point )

Your confusion is normal, I would also be confused if I was in your shoes.

I was lucky to begin chess competition out there, in the real world, with real chess boards, chess sets, chess clocks, score sheets, arbiters in the flesh, you name it.

So, the oddities on chess websites are to me just chess websites oddities, and I don't pay attention to the whole commercial online garbage trash can online chess tends to carry around anymore.

For one example, this chess website, chess.com, comes and go tell you that this bot has that rating. Yet, it's no rating at all, for ratings are the product of games actually played, between rated humans. What chess.com insist on calling a "rating" for their bots, is what is called an EVALUATION. And, as such, has as much value as the judgement of the coder or the staff member (who gave it) over chess strenght is accurate. Which will be more than often, something close to the absolute zero.

Chess websites do implement such things with one idea in mind: commercial success, or if you want, money income. Every dumb feature that is supposed to produce cheap thrills among the guilible newcomers, and encourage them to pay for something, like a pay account, has priority over you understanding anything about anything.

Therefore, I do invite you to go to the FIDE website, or the USCF website, and take a good read about the rules of chess, and how the ratings do work. So, you'll be less confused about it all and know how function real chess competition and rated games.

I see. So it's really just the random blunders that make it unadvisable. It makes sense. Because even though they play "consistently", they do make blunders out of nowhere. That's also the reason why people could cheese them with the hippo. The idea is, if you play solid chess, they'll make an exploitable blunder at some point. Alright. This point was made by the other guy as well but I was wishing that he'd elaborate about it more.

Your ideas about bots are incorrect, there's no evaluation of any sort in their ratings. Bots (engines) and every configuration (Skill or UCI_Elo) are calibrated through engine vs engine tournaments. You ask what is the baseline? FIDE-rated Grandmasters who competed with engines once gave that baseline. Engines have obtained FIDE-rating thanks to them. You would argue that FIDE ratings inflated since then, but if you look up to what extend you will find that at that level it would be maybe 50 Elo, nothing significant. Since software version, configuration file, executables, source codes are not changing, not evolving or degrading, all results are 100% reproducible. Once rated forever rated. And some new unrated yet engines can be rated against a rated engine. That's how we know that Stockfish 17 is 3600 something. You can do it yourself on your computer. Download some engines, run tournaments between them and observe that calibration is correct. If it is not correct - sure, message developers. But prove it first before making such claims.

Chess.com bot engine (Komodo) internally has no Elo setting, only Skill setting but developers (GM Larry Kaufman and others) provided approximate Elo <--> Skill mapping formulas. These developers know what they're doing, they know about ratings more than we do, they aren't doing any cheap guesses.

Chess.com bot engine (Komodo) runs locally in your browser or on your mobile device and uses local resources. It might have time limit for thinking configured AND THEN depending on background processes on your device and processing power, bots may act weirdly. Therefore I don't recommend playing bots on chess.com if you're looking for a stable performance. You can download Komodo and use it locally on your computer without time-limit per thinking, only with Skill and other settings adjusted.

Even so, this is still no rating such as it is for a human player.

You do claim chess.com is doing the job to pair chess engines against each other like in the days where Mephisto was winning every year the chess softwares world championship back in the 90s, well, I do doubt you actually know what anyone is doing in their job.

Why spend the money to see what happens when "depleted engine 1" plays "depleted engine 2" and so forth, when these depleted engines can't reproduce human errors but do play perfectly where a human would fail, then blunder their Queen for no good reason in order to obey to the command "play 2° best move now instead of 1° best move"?

So, coders try to weaken them so strong chess engines such as we all know (komodo, stockfish, etc) by asking them to do this and to do that instead of just playing the best move they find. But it never ever produces the play of a human 1300 nor anything close to it.

I do understand what I wrote was some of a shortcut, and that reality is some more complicated than that. Yet, the truth is the same in the end: 1300 "rated" bots don"t play like 1300 rated humans, and probably never will. So, you can't accept the lie, and the fact chess.com is not making it clear that these ratings are no actual ratings.

Besides, same stupid commercial lies do go on with "puzzle ratings". This is where you can see even better the spirit of fraud that do prevail in the misleading use of terms that do belong to the noble art and sport of chess.

Again, any real, significant and actual chess rating must be the product of games played rated human vs rated human. Anything named "chess rating" obtained otherwise, is a fraud when it coimes to compare to a human rated player same strenght, for all the reasons I mentioned, and more.

I think , chess is all about practice with new position and dare to make some risk not stupid .. Not like I am saying to do blunders but risk that you think are good ... I recently realised these and personally i also have hard times in chess like I think my highest will not be more than 1300 currently but I think believing in yourself and doing constant practice will help you!!! Consistency is needed in everything not in just real life but in chess also!!!

Chess com literally broadcasts computer tournaments every second. Not only Komodo.

GM Larry Kaufman made Komodo before it was acquired by chess.com. GM Larry Kaufman did some deep investigation of FIDE vs chess.com rating differences. So yes, he knows how ratings work.

Late 90s, early 00s - yes, that's when engines got their rating from humans.

Myth that weakened engines play top-moves then blunder a queen is just a myth, you clearly have never written an engine and don't know how easy it is to make sure such pattern never ever happens. In fact, in low-level human play such pattern is more prevalent. BUT as I mentioned, chess.com bots run locally and may not be reliable and may indeed display such pattern.

Stockfish is not very humanized, true. But we have Komodo with Human personality, we have Maia, that's 1, and we have a big issue: humans actually cannot easily distinguish human from engine. If it was so easy, ch3at1ng wouldn't be an issue!

Also, humans have good and bad days. Humans can relax or concentrate. Engines are more stable in that regard.

If an engine beats some players with certain score you can rate that engine, there's no fraud because that rating then perfectly predicts future performances of this engine against other humans and against other engines rated in same way. And you can exclude humans from this process. Maybe some inaccuracy will appear, but again, you didn't measure, you didn't do any experiments, so you can't claim anything about their accuracy.

You do exaggerate big time. Playing every now and then 2° best move instead of playing 1° best move is simply the first attempt to deplete a chess engine, make it play weaker, and maybe look "human". Of course more elaborated commands have been invented since then, but the problem remains the same: no matter what, engines don't play like humans.

So, when it comes to bots, letting newcomers nelieve that one bot will play like a 1700, is simply a fraud. And the purpose of it is making money by generating cheap thrills and illusions.

Last but not least, I do notice you skipped the part where the same BS is implemented in puzzles, where some sub 1000 player can make themself belkive they'd have a GM potential for achieving mindblowing "ratings" out of solving lotsa puzzles. Like "hey I've got that potential, let's not waste it and now invest money in chess BS trainings and lessons online".

If there is a myth, that myth is that online chess can replace real chess. And an other myth is that AIs can play chess, or even understand a one single word of English. But I won't bother elaborate on that: others have done it already and bad faith do bore the poop out of me.

To anyone who is sure that it is easy to tell engine play from human play I already offered a challenge. 12 game, tell which is which, options:

A-White is human, black is bot
B-White is bot, black is human
C-Both are humans
D-Both are bots

Just list of letters, feel free to add your reasoning.

Answers are here:

https://pastebin.com/HCxwkAnf

once posted always there, I cannot change them but they are under password, I will reveal the password later as well as info about each game.

But challenge is to avoid using an engine. Because you don't use engine to analyze your opponent while you are playing do you. Or you can separately analyze each game with an engine after you've done analyzing manually, to make two sets of answers.

If someone gets 100% right answers I am deleting my account and never coming back.

1)

2)

3)

4)

5)

6)

7)

8)

9)

10)

11)

12)

Kaeldorn (and maybe some others too), you make some very good points, and I hadn't thought about the money-making aspect. I can definitely see how making newbies feel like they are higher rated / beating a high-rated player (for me, the 1700 bot) would help the site's profits. That is absolutely true. Everyone likes to feel like they're succeeding, and people will pay for that feeling.

Thanks, everyone. Bottom line is that I know I'm certainly not a master chess player, and at this point I should just focus on learning the game. Since my goal isn't to become a master competitor (I'm too old for that!), I don't think it matters hugely to my progress if bots are or aren't as good as a human (but I understand why it may matter to more experienced players).

I wonder if there's a generation / age gap in the way people think. I'm probably older than many people here (I'm in my 40s), and I'm inclined to believe those who say that OTB is the best way to learn (not based on stats or research, so I'm not claiming that this is really true, it's just what sounds right to me). I wonder if that's my old-school way of thinking, or if younger players believe this as well.

There are benefits in usage of real-world board&pieces (big, I mean properly sized, weighted, shaped and colored), more neurons activate when you look at real 3-dimensional shapes and when you touch them with hands, when you hear sounds, all that helps with learning because brain can make more connections. However you can still play with bots while using real chessboard. DGT or other electronic chessboard or just a regular chessboard and manually entering your moves into a program.

And I'll say again, that's it's not that easy to tell human play from bot play, if you think it is, solve my above challenge. For sure playing against humans is fun but you don't always have them nearby. Online chess is too chaotic, ratings do not correspond to the play well and level of concentration varies and many distractions are possible while OTB, players tend to do their best, OTB ratings are usually more accurate.

I don't believe in conspiracy theories, so I think bots on chess.com act weirdly just because they have time-limit on thinking and it is device-dependent (chess.com bots run locally using your local resources, not server resources).