Hi, If you have missed it we have started a 14-day puzzle challenge here on Say Chess. The goal is simple to complete as many 5-min Puzzle rushes or Puzzle Storms (Lichess) as possible in order to improve our pattern recognition. For more info: https://saychess.substack.com/p/14-days-of-puzzle-rush-storm-challenge/
Oh man I hope no one gets offended but man is it impossible to please people ! It’s just a challenge to do some chess tactics, you can just write your scores on a sheet of paper (or google doc) and send them at the end of the day in bunch of 10. Yeah it’s not perfect but we are not looking at sending people on the moon here. I believe the earth won’t explode if someone enters the wrong date, no need to hire a programmer or I don’t know what. If it’s too complicated, you can just do the challenge, not report and just check if you get higher scores after 2 weeks I don’t know.
I did today 100 runs (10 attempts/batches for 10 runs). In two of these I put the wrong date (26th instead of 27). I hope you can fix that as I do not want to spoil the data.
You probably may know the reason of this mistake - 100 runs average 36-40 is close to 4K puzzles solved today. It may explain why I have difficulty to figure out what day is today ;)
May 27, 2023·edited May 27, 2023Liked by Martin B. Justesen
Well done Martin!
I love this challenge and the way you prepare that and realizing that :)
Possible improvements (of course in case you think they are needed and helpful)
1. Put the lichess and chesscom icon/logo with the specific reports data headers. This way we can immediately can see the ones we are most interested with (to me it is lichess). The icon/logo should be before the header [i.e. "icon/logo" - "Puzzle Storm / Puzzle Rush - Highest low score"]
2. At the end of the article - provide a short description (1-2 sentences) about the specific data group characteristics. For example: Puzzle Storm / Puzzle Rush - Highest low score - it means how much the worst performance is different from the best one. The bigger number - the better.
I guess as for now these improvements would be the most important (especially for readers who are not statisticians). In the meantime the batch of 10 scores is SUPER helpful - thank you very much for providing this! :)
What are your thoughts after day 1 of the challenge? Are you surprised by the numbers of participants? :)
Thanks! And thanks for the improvement ideas :) I will try to improve on it, and later also make some data visualisation. And I think we have a good amount of people starting out (some have yet to register anything). Will be interesting to see how many % manage to record a run for each day of the challenge
If we are talking about improvement ideas - it is my passion and I try to look for the ways to make it better and better: especially last 2,5 years I devoted to solving chess Puzzle Storm at lichess.
I want to see what are the necessary components to "get really good" at solving puzzles (tactics) - and I chose lichess and Puzzle Storm puzzles for tests (and in the meantime the Blitztactics.com).
My average score at Puzzle Storm is 35-40 and I cannot break 61 (I reach such score 3 or 4 times) even though I have solved so far 850 K puzzles (the details below).
1) LICHESS
a) Puzzle Storm = 9316 x 35 (average score per each run; the highest score 61) = 326 K
b) Puzzles (normal mode) = 9312 (each level of difficulty - mostly up to 2200-2300 lichess)
c) Puzzle Racer = 3789 x 40 (average score about 40; the highest score 84) = 151,5 K
d) Puzzle Streak = 105 x 40 (average score about 40; the highest score 72) = 4,2 K
Maybe get someone with some programming experience to grab the data. People self reporting could have errors of omission easily too that make auto grabbing data much more reliable.
I am very confused! I thought I signed up, but then I did 5-10 puzzle storms today and then my info didn't show up in the list today. And I am pretty sure I did them before you sent the email. Would you confirm that 'AngelOfMusic' is signed up for lichess?
Oh man I hope no one gets offended but man is it impossible to please people ! It’s just a challenge to do some chess tactics, you can just write your scores on a sheet of paper (or google doc) and send them at the end of the day in bunch of 10. Yeah it’s not perfect but we are not looking at sending people on the moon here. I believe the earth won’t explode if someone enters the wrong date, no need to hire a programmer or I don’t know what. If it’s too complicated, you can just do the challenge, not report and just check if you get higher scores after 2 weeks I don’t know.
Martin
I did today 100 runs (10 attempts/batches for 10 runs). In two of these I put the wrong date (26th instead of 27). I hope you can fix that as I do not want to spoil the data.
You probably may know the reason of this mistake - 100 runs average 36-40 is close to 4K puzzles solved today. It may explain why I have difficulty to figure out what day is today ;)
Amazing effort 📈👏🏻👏🏻👏🏻 I will correct it for you!
Well done Martin!
I love this challenge and the way you prepare that and realizing that :)
Possible improvements (of course in case you think they are needed and helpful)
1. Put the lichess and chesscom icon/logo with the specific reports data headers. This way we can immediately can see the ones we are most interested with (to me it is lichess). The icon/logo should be before the header [i.e. "icon/logo" - "Puzzle Storm / Puzzle Rush - Highest low score"]
2. At the end of the article - provide a short description (1-2 sentences) about the specific data group characteristics. For example: Puzzle Storm / Puzzle Rush - Highest low score - it means how much the worst performance is different from the best one. The bigger number - the better.
I guess as for now these improvements would be the most important (especially for readers who are not statisticians). In the meantime the batch of 10 scores is SUPER helpful - thank you very much for providing this! :)
What are your thoughts after day 1 of the challenge? Are you surprised by the numbers of participants? :)
Thanks! And thanks for the improvement ideas :) I will try to improve on it, and later also make some data visualisation. And I think we have a good amount of people starting out (some have yet to register anything). Will be interesting to see how many % manage to record a run for each day of the challenge
If we are talking about improvement ideas - it is my passion and I try to look for the ways to make it better and better: especially last 2,5 years I devoted to solving chess Puzzle Storm at lichess.
I want to see what are the necessary components to "get really good" at solving puzzles (tactics) - and I chose lichess and Puzzle Storm puzzles for tests (and in the meantime the Blitztactics.com).
My average score at Puzzle Storm is 35-40 and I cannot break 61 (I reach such score 3 or 4 times) even though I have solved so far 850 K puzzles (the details below).
1) LICHESS
a) Puzzle Storm = 9316 x 35 (average score per each run; the highest score 61) = 326 K
b) Puzzles (normal mode) = 9312 (each level of difficulty - mostly up to 2200-2300 lichess)
c) Puzzle Racer = 3789 x 40 (average score about 40; the highest score 84) = 151,5 K
d) Puzzle Streak = 105 x 40 (average score about 40; the highest score 72) = 4,2 K
LICHESS TOTAL = 490 K
2) BLITZTACTICS
a) Infinity puzzles solved = 23546 [ https://blitztactics.com/chessteachertom ]
b) Speedruns completed = 9509 x 30 (the number of puzzles each run) = 318,3 K
c). Countdowns completed = 405
BLITZTACTICS TOTAL = 352 K
3) OTHER sources (solving puzzles at streams with lichess studies, the puzzles with PGN format from my database/computer, etc.)
I estimate I have solved about 6-8 K puzzles (during last 6 years), mostly simple puzzles like mate in 1, 2, 3 or tactics up to 1700-1800.
ALL THE SOURCES (A+B+C) in total = 850 K (not counted puzzles solved from paper books and workbooks - only digital puzzles for the last 5 years).
Let me know if now you believe I am really obssessed about solving chess online puzzles or should I provide more evidence :)
How do you recover from a bad 3rd move
Hi Jan.. not sure I understand your question?
I was supposed to play Nc6 but played Nf6
I signed up before hand, earlier this week roughly. I don't want to manually enter every run I do.
You can enter 10 at a time. If you don’t want to be free to not participate. I do not have the capacity to look up
Results for 100 people 😬
Maybe get someone with some programming experience to grab the data. People self reporting could have errors of omission easily too that make auto grabbing data much more reliable.
I would need access to the chesscom api, which I’m unsure how to do. But I think self reporting works fine even though errors can happen
I am very confused! I thought I signed up, but then I did 5-10 puzzle storms today and then my info didn't show up in the list today. And I am pretty sure I did them before you sent the email. Would you confirm that 'AngelOfMusic' is signed up for lichess?
Hi Joe, have you entered the runs into the form?