Statistics on problems solved so far
Statistics on problems solved so far
Thank you for all the interesting statistics already shown.
I have three suggestions that I would like you to consider:
(a) on the statistics sheet it shows the count of members that has solved a specific number of problems. I would find it interesting to see how many of these members are currently "active". At the moment it shows more that 763 000 having solved one problem. I suspect that many of those members have not solved their single problem recently. A possible definition of an "active" member might be that a member must have solved at least one puzzle in the last 365 days. This exact definition was previously used for the ranking of members per country. This statistic does not have to be updated in real time, a update every week or so would still give a very good idea on the progress of active members.
(b) on my progress sheet, my average difficulty of problems solved is shown. At the moment it only shows this average figure rounded to the nearest one percent. Would it be possible to add at least one (even two) decimal places? I have solved around 200 puzzles and my average percentage (rounded to one percent) does not change very frequently. It would be interesting to see how your average percentage ticks up over time as you solve more and more difficult problems.
(c) It is intesting to know the highest difficulty level of a problem that you have solved. This statistics can remain unchanged for a long time (if you have solved one very difficult problem relative to the other problems you have solved). Is it possible to add another category something like "Number of problems solved with a difficulty level in excess of X% is Y."? The value of X might be fixed for all members say 40% or 50% or 60%. Alternatively it might be linked to the difficulty level of the most difficult puzzle solved (e.g. set X equal to the difficulty level of the most difficult problem solved less 25%). As a side note, it would be interesting to see (on the statistics sheet) the distribution of numbers of members based on the highest difficulty level that each member has solved so far  e.g. have more that 2% of members solved at least one puzzle with a difficulty level of 75% or higher?
I have three suggestions that I would like you to consider:
(a) on the statistics sheet it shows the count of members that has solved a specific number of problems. I would find it interesting to see how many of these members are currently "active". At the moment it shows more that 763 000 having solved one problem. I suspect that many of those members have not solved their single problem recently. A possible definition of an "active" member might be that a member must have solved at least one puzzle in the last 365 days. This exact definition was previously used for the ranking of members per country. This statistic does not have to be updated in real time, a update every week or so would still give a very good idea on the progress of active members.
(b) on my progress sheet, my average difficulty of problems solved is shown. At the moment it only shows this average figure rounded to the nearest one percent. Would it be possible to add at least one (even two) decimal places? I have solved around 200 puzzles and my average percentage (rounded to one percent) does not change very frequently. It would be interesting to see how your average percentage ticks up over time as you solve more and more difficult problems.
(c) It is intesting to know the highest difficulty level of a problem that you have solved. This statistics can remain unchanged for a long time (if you have solved one very difficult problem relative to the other problems you have solved). Is it possible to add another category something like "Number of problems solved with a difficulty level in excess of X% is Y."? The value of X might be fixed for all members say 40% or 50% or 60%. Alternatively it might be linked to the difficulty level of the most difficult puzzle solved (e.g. set X equal to the difficulty level of the most difficult problem solved less 25%). As a side note, it would be interesting to see (on the statistics sheet) the distribution of numbers of members based on the highest difficulty level that each member has solved so far  e.g. have more that 2% of members solved at least one puzzle with a difficulty level of 75% or higher?
 euler
 Administrator
 Posts: 3067
 Joined: Sun Mar 05, 2006 4:49 pm
 Location: Cheshire, England
 Contact:
Re: Statistics on problems solved so far
I've made some updates based on your suggestions:
1) The mean difficulty is now displayed to two decimal places.
2) The progress page now shows the number of problems at each difficulty rating from 5% to 100% (in 5% increments) and indicates your progress with each of those levels. I think that many users will find this helpful.
With reference to your other suggestion/request... At the moment, each time a member solves a problem their profile is updated with the total number of problems solved. Similarly, the total number who have solved a particular problem is also updated. However, knowing that A members have solved one 5% problem, B have solved another 5% problem, and so on, does not tell us how many people have solved a 5% problem in total. Similarly for other difficulty ratings. This would require going through every profile and through every problem each member has solved in turn, to determine the difficulty of each problem so that a boolean table could be obtained for each user: 5% true/false, 10% true/false, and so on. Then this could be used to update a master array. But it would be far too DB process intensive.
1) The mean difficulty is now displayed to two decimal places.
2) The progress page now shows the number of problems at each difficulty rating from 5% to 100% (in 5% increments) and indicates your progress with each of those levels. I think that many users will find this helpful.
With reference to your other suggestion/request... At the moment, each time a member solves a problem their profile is updated with the total number of problems solved. Similarly, the total number who have solved a particular problem is also updated. However, knowing that A members have solved one 5% problem, B have solved another 5% problem, and so on, does not tell us how many people have solved a 5% problem in total. Similarly for other difficulty ratings. This would require going through every profile and through every problem each member has solved in turn, to determine the difficulty of each problem so that a boolean table could be obtained for each user: 5% true/false, 10% true/false, and so on. Then this could be used to update a master array. But it would be far too DB process intensive.
impudens simia et macrologus profundus fabulae

 Posts: 351
 Joined: Sun Mar 22, 2015 2:30 pm
 Location: India
 Contact:
Re: Statistics on problems solved so far
Seems 85% difficulty rating is the most difficult for me.. Solved just 1 out of the 10 problems available which makes Problem 544 (View Problem) even more sweet for me...
It is not knowledge, but the act of learning, not possession but the act of getting there, which grants the greatest enjoyment.
Re: Statistics on problems solved so far
This is a really nice addition.
Technically, everyone is full of himself.

 Posts: 6
 Joined: Tue Mar 21, 2017 2:09 pm
Re: Summer 2017 Website Update
On the progress page, problem 617 has the color of not solved problems even if you have solved it.
Is there a better place to post things like this in the future?
Is there a better place to post things like this in the future?
 euler
 Administrator
 Posts: 3067
 Joined: Sun Mar 05, 2006 4:49 pm
 Location: Cheshire, England
 Contact:
Re: Summer 2017 Website Update
The Errors/Warnings/Bugs thread is probably the best for any bugs/glitches you have found, but I tend to check most threads in this area (News, Suggestions, and FAQ) regularly.
Thanks for letting me know about the bug. It was an unintended consequence of another change I made this morning. It should be fixed now and, hopefully, I've not introduced another bug in trying to fix the bug introduced by fixing the previous bug.
Thanks for letting me know about the bug. It was an unintended consequence of another change I made this morning. It should be fixed now and, hopefully, I've not introduced another bug in trying to fix the bug introduced by fixing the previous bug.
impudens simia et macrologus profundus fabulae
Change of Behavior of the new Problem Statics in Progress tab
Yesterday I found a table was added to the Progress tab, showing the numbers of problems solved for each difficulty category. However, I noticed that my number of problems solved for 100% decreased from 18 to 14 just now, I have also observed this change in some friends' profiles. Moving the cursor on the blue info icon no longer shows problem 566, 579, 585 and 597, and the average difficulty is affected too. Is this change intended or a bug? If it's intended, why the total count of is still 18? (Those four problems are the only ones whose difficulties are not finalized)
 euler
 Administrator
 Posts: 3067
 Joined: Sun Mar 05, 2006 4:49 pm
 Location: Cheshire, England
 Contact:
Re: Change of Behavior of the new Problem Statics in Progress tab
I'm struggling to keep up with all the threads this is appearing in at the moment. To quote me from earlier...
Thanks for reporting it. That bug should be fixed, but who knows what else I've broken?!
So you've found "another bug in trying to fix the bug introduced by fixing the previous bug".
Thanks for reporting it. That bug should be fixed, but who knows what else I've broken?!
impudens simia et macrologus profundus fabulae

 Posts: 4
 Joined: Fri Jul 15, 2016 6:24 pm
Re: Statistics on problems solved so far
Whow, the new table about solved problems per difficulty level is awesome, thanks a lot for creating this! What a nice surprise!
However, as of today, the count concerning the one 100% problem I have solved so far has disappeared (it was still there yesterday).
Could it be that the way nonfinalised difficulties are handled may have changed since yesterday? Or perhaps it is a bug? Yesterday, the table reported 18 available problems at difficulty level 100%, while today only 14 of them show up. I think 18 was the correct value.
While I am it, I could imagine the table could be transposed, so that it would need only three rows and 21 columns, which would still fit and save some space. Clearly not important, but perhaps worth trying.
Thank you so much for creating this wonderful site! I cannot imagine how I was ever able to live without it...
And best wishes for the new year!
Andreas
However, as of today, the count concerning the one 100% problem I have solved so far has disappeared (it was still there yesterday).
Could it be that the way nonfinalised difficulties are handled may have changed since yesterday? Or perhaps it is a bug? Yesterday, the table reported 18 available problems at difficulty level 100%, while today only 14 of them show up. I think 18 was the correct value.
While I am it, I could imagine the table could be transposed, so that it would need only three rows and 21 columns, which would still fit and save some space. Clearly not important, but perhaps worth trying.
Thank you so much for creating this wonderful site! I cannot imagine how I was ever able to live without it...
And best wishes for the new year!
Andreas
 sjhillier
 Administrator
 Posts: 507
 Joined: Sun Aug 17, 2014 3:59 pm
 Location: Birmingham, UK
 Contact:
Re: Statistics on problems solved so far
To try to help Euler cope with all the messages coming in on different threads, I think this question is answered by his reply here.Obergscheidle wrote: ↑Sun Dec 31, 2017 3:38 pm
However, as of today, the count concerning the one 100% problem I have solved so far has disappeared (it was still there yesterday).
Could it be that the way nonfinalised difficulties are handled may have changed since yesterday? Or perhaps it is a bug? Yesterday, the table reported 18 available problems at difficulty level 100%, while today only 14 of them show up. I think 18 was the correct value.
 euler
 Administrator
 Posts: 3067
 Joined: Sun Mar 05, 2006 4:49 pm
 Location: Cheshire, England
 Contact:
Re: Statistics on problems solved so far
Thanks, Steve.
I've moved all the related posts into this thread to try to keep it in one place.
I've moved all the related posts into this thread to try to keep it in one place.
impudens simia et macrologus profundus fabulae
Re: Statistics on problems solved so far
I have just noticed the toggle on the statistics sheet that allows you to switch between statistics for all registered members and active members only. I do not know how long the toggle has been available but I find it very interesting (and that deals with my post at the start of this thread.)
Thanks again for a brilliant site!
Thanks again for a brilliant site!
 euler
 Administrator
 Posts: 3067
 Joined: Sun Mar 05, 2006 4:49 pm
 Location: Cheshire, England
 Contact:
Re: Statistics on problems solved so far
You clearly don't read the News page on the main site.
You're very welcome. It was a splendid idea you had. Worthy of taking the time to implement.
You're very welcome. It was a splendid idea you had. Worthy of taking the time to implement.
impudens simia et macrologus profundus fabulae

 Posts: 54
 Joined: Thu Nov 03, 2016 4:32 pm
Re: Statistics on problems solved so far
Didnt feel like making another thread so I will just post this here.
I think it would be a nice addition that the problem numbers would appear when you hover over the bar at some difficulty.
Right now for example it says "Completed 11% of problems with difficulty rating 100%" but it can be quite time consuming to find out which problems of that difficulty were solved.
I think an addition of something like "Problems : 439, 502" would be great.
I think it would be a nice addition that the problem numbers would appear when you hover over the bar at some difficulty.
Right now for example it says "Completed 11% of problems with difficulty rating 100%" but it can be quite time consuming to find out which problems of that difficulty were solved.
I think an addition of something like "Problems : 439, 502" would be great.
 euler
 Administrator
 Posts: 3067
 Joined: Sun Mar 05, 2006 4:49 pm
 Location: Cheshire, England
 Contact:
Re: Statistics on problems solved so far
Hovering and showing which problems would result in too much text for those who have solved many problems. Instead what I have done is changed the colouring system when a problem is solved. The shading indicates the difficulty of the problem.
impudens simia et macrologus profundus fabulae
Re: Statistics on problems solved so far
Looks great.
Re: Statistics on problems solved so far
Great.
Orange is the color of our royal family.
It gives me the real orange feeling.
 euler
 Administrator
 Posts: 3067
 Joined: Sun Mar 05, 2006 4:49 pm
 Location: Cheshire, England
 Contact:
Re: Statistics on problems solved so far
Glad you guys like it. Along with the tribute to all of our valued members of the community from the Netherlands, it does seem to fit better with the colour scheme of the level award badges too.
I'll make a brief news announcement to let people know that their page is not broken. Most of the work I do is hidden or subtle, but this one is rather inyourface.
I'll make a brief news announcement to let people know that their page is not broken. Most of the work I do is hidden or subtle, but this one is rather inyourface.
impudens simia et macrologus profundus fabulae

 Posts: 54
 Joined: Thu Nov 03, 2016 4:32 pm
Re: Statistics on problems solved so far
It looks great! Thank you.
Re: Statistics on problems solved so far
I have a suggestion for an addition to the detail shown of the "fastest 100 solvers" thread. Could there be an additional column that shows how many puzzle those top 100 puzzlers have solved? Lets take puzzle 668 as an example: John Smith had solved 295 puzzles of the 668 puzzles so far. The new column will then show 295.
I am interested to see how the number puzzlers that has solved less than 100 puzzles compare to those that have solved more than 600. The number of puzzles solved shown for a person should not change as it would be the number of puzzles that person had solved at that time  no need to update the value if that person solves subsequent puzzles.
I am interested to see how the number puzzlers that has solved less than 100 puzzles compare to those that have solved more than 600. The number of puzzles solved shown for a person should not change as it would be the number of puzzles that person had solved at that time  no need to update the value if that person solves subsequent puzzles.