for example, my puzzle rating is 3000+ but im not even 2000 OTB yet. the specified times are around 30 - 40 seconds and i typically take 4 minutes. This is why my OTB rating is not 3000
Chess.com's puzzle rating system is trash
Sort:
i agree
Oct 10, 2020
0
#25
The OP´s bullet rating is 1100. His puzzle rating is 2400. OP moans his puzzle rating is too low.
Oct 10, 2020
0
#26
horselover123 wrote:
pinkblueecho wrote:
The OP´s bullet rating is 1100. His puzzle rating is 2400. OP moans his puzzle rating is too low.
OP did not say his puzzle rating was too low, he only said that the system is not good and there should be more time per puzzle
and in real games, the winner can win with 0.1 sec left on their clock or 20 min left on their clock, but they still win regardless of how much time is on the clock as long as they don't flag. why would they give such a small amount of time for each puzzle, when in real games no one is timing you to do this puzzle/tactic in this short amount of time? i get why they would time it, to help improve your speed and stuff like that, but it doesn't have to be 20 sec. In classical, your moves should not take less than 15 sec to make, unless it's theory and the first five moves or whatever. giving such a short amount of time for each puzzle encourages quick thinking over precise calculation.
You can take as much time as you want. The target time is to get 100% of the available points. The tactics are rated based on other players failing to solve or solving correctly, or at least they used to be. So, to get to 2800, means enough players within 400 points or so of the puzzle missed it enough to raise it's rating.
As mentioned, the example times in the image shared were for a 2500 rated player to solve a puzzle 200-400 points lower rated, which seems correct. Points would still be gained by solving it close to the average time as well.
Oct 10, 2020
0
#27
Puzzles are all about pattern recognition. The more you do them the faster you can solve similar puzzles, or at least have a candidate move that is the correct move.
i bet I could solve minor piece endgames quickly if I did a hundred of them, but I can’t. That’s because I skip every single one of them, they’re so tedious lol
Oct 10, 2020
0
#29
krazeechess wrote:
SNUDOO wrote:
Puzzles are all about pattern recognition. The more you do them the faster you can solve similar puzzles, or at least have a candidate move that is the correct move.
i bet I could solve minor piece endgames quickly if I did a hundred of them, but I can’t. That’s because I skip every single one of them, they’re so tedious lol
true, but it is also calculation. i think there should be at least a small increase in the expected time. you can't just solely go off of pattern recognition. some puzzles are not obvious moves, and you actually have to calculate between a few different moves. that's why there should be a increase. or, instead of a linear increase from the lower rated puzzles to the higher rated ones (i think it is linear), it should be exponential.
Puzzle ratings are already inflated, though. Your request would only inflate it more. And I’m not sure why you didn’t post something like this in the lower rating ranges? Because it was easy to solve?
Oct 10, 2020
0
#30
krazeechess wrote:
as you can see, 2 points for getting a puzzle right is nothing. this is why we need more time
Not really. If your rating is 2000, for example, you should be able to spot a smothered mate tactic in like 30 seconds. Yes, if you play classical time won't make much of a difference, but I guess they want you to be able to spot puzzles quickly.
Oct 10, 2020
0
#31
horselover123 wrote:
23Dragons wrote:
krazeechess wrote:
as you can see, 2 points for getting a puzzle right is nothing. this is why we need more time
Not really. If your rating is 2000, for example, you should be able to spot a smothered mate tactic in like 30 seconds. Yes, if you play classical time won't make much of a difference, but I guess they want you to be able to spot puzzles quickly.
ya, but not all puzzles are smothered mates and all puzzles aren't as ez as smothered mates
it's just an example.
Oct 10, 2020
0
#32
krazeechess wrote:
23Dragons wrote:
krazeechess wrote:
as you can see, 2 points for getting a puzzle right is nothing. this is why we need more time
Not really. If your rating is 2000, for example, you should be able to spot a smothered mate tactic in like 30 seconds. Yes, if you play classical time won't make much of a difference, but I guess they want you to be able to spot puzzles quickly.
spotting puzzles quickly is supposed to be for puzzle rush. not rated tactics. rated tactics are about getting the puzzle correct, yet they urge us to solve it quickly
some of it is recognizing patterns quickly. If I spent two hours mulling over some weird endgame tactic in contrast to 15 seconds trying to calculate it and I got the same rating gain, how would that be fair?
Oct 10, 2020
0
#33
krazeechess wrote:
no that's not what i mean. they should increase it by 20 sec. in higher rated tactics, calculation is necessary.
And how fast you calculate is key.
Oct 10, 2020
0
#34
horselover123 wrote:
Martin_Stahl wrote:
You can take as much time as you want. The target time is to get 100% of the available points. The tactics are rated based on other players failing to solve or solving correctly, or at least they used to be. So, to get to 2800, means enough players within 400 points or so of the puzzle missed it enough to raise it's rating.
As mentioned, the example times in the image shared were for a 2500 rated player to solve a puzzle 200-400 points lower rated, which seems correct. Points would still be gained by solving it close to the average time as well.
i've often got like +2 points on a puzzle simply from going overtime, the last puzzle on my puzzle log even shows +1 points even tho i got it correct, cuz i went overtime one min
speaking of this (i know i should probably make a new topic) 99% of the puzzles i get are not even my rating, so i am a 2500 built off of >2300 tactics (i hope i am competent enough at math and that the symbol i used means less than 2300, not greater than 2300). i lose too many points whenever i fail one of these and barely gain any points, my tactics could he higher if i got puzzles actually correlating with my rating since i would gain more points/lose less points per puzzle, and yes i can solve them when i get them, i just dont get any
proof: out of 16 puzzles, only 2 of them are even 2500/close to 2500
i have one 2419 puzzle and one 2530, which i were both able to solve except for the 2530 which i got like 3/4 out of, but whatever if i spent more time i could've solved it 4/4
but 2/16 (or 1/8, for math people) of my puzzles are my rating, shouldn't it be more like the majority of them are my rating instead of just a ton of 2300 and below puzzles?
edit: i did more puzzles and i went back to check my puzzle log, and i've only had one new 2400/2500 puzzle out of like twenty (i failed it by doing the right moves in the wrong sequence, oh well)
I'm guessing there are very few Puzzles 2400+ so that's one reason you don't see too many of them. If the system gave you a higher percentage, you would likely get them repeatedly.
Oct 10, 2020
0
#35
krazeechess wrote:
i have a concept. what if there is two types of rated puzzles. one which is for speedsolving and one which is for solving slowly. speedsolving will have a time limit for each puzzle, based on the rating of it. slow solving will have a much greater time limit, and harder puzzles.
+1
Oct 10, 2020
0
#36
krazeechess wrote:
23Dragons wrote:
krazeechess wrote:
no that's not what i mean. they should increase it by 20 sec. in higher rated tactics, calculation is necessary.
And how fast you calculate is key.
true, but rated puzzle are supposed to prepare you for classical games. you won't have to calculate so fast in a classical game.
Right, but tactics are part of everyday calculation. Even when you play a classical game, you have to be able to realize in 5 minutes or less(you have to calculate a lot) whether a line is good or bad. If you take too much time, you might not have as much time to calculate in the endgame. If you take too little time, you might walk into some sequence that you looked over during your scan of the board.
Oct 10, 2020
0
#37
krazeechess wrote:
i have a concept. what if there is two types of rated puzzles. one which is for speedsolving and one which is for solving slowly. speedsolving will have a time limit for each puzzle, based on the rating of it. slow solving will have a much greater time limit, and harder puzzles.
That's an interesting idea. +1
Oct 11, 2020
0
#38
krazeechess wrote:
23Dragons wrote:
krazeechess wrote:
no that's not what i mean. they should increase it by 20 sec. in higher rated tactics, calculation is necessary.
And how fast you calculate is key.
true, but rated puzzle are supposed to prepare you for classical games. you won't have to calculate so fast in a classical game.
Correct is good, but quickly correct is better.
why settle for something second rate? And you haven’t responded to my question, why didn’t you raise this topic back when you were like a 1500 puzzle rating?
Oct 11, 2020
0
#39
krazeechess wrote:
i have a concept. what if there is two types of rated puzzles. one which is for speedsolving and one which is for solving slowly. speedsolving will have a time limit for each puzzle, based on the rating of it. slow solving will have a much greater time limit, and harder puzzles.
That’s called training mode lol
exactly, which is why his argument doesn't make sense, chess com puzzles are good the way it is, if u can't do it within the amount of time, then that's not ur OTB rating, that's how it works