How is the score computed?

[duhmel]

I've done lots of puzzles (never beaten average) but I can't figure out the method used to compute the scores. No matter what I do I score between 100 and a50 even though my time goes from 1.5 to 4.0 times the average solution time.Is it a secret or is it available.

[briang]

The total possible points appear to be calculated strictly based on the number of clues, as you will notice if you look at puzzles with the same number of clues, and it is computed from there. The highest possible appears to be 1360 on a 26 clue crossword. Your score is calculated on the time it takes to complete. Whether the actual scores generated are curved based on average solve time, I haven't taken the time to try to figure out, but that would make sense.

[duhmel]

How about a response from the moderator?

[redpicker]

duhmel, you might want to check out this thread:

https://forum.puzzlebaron.com/forum/puzzle-baron/acrostic-puzzles/24566-new-scoring

Particularly post #11.

[duhmel]

duhmel, you might want to check out this thread:

https://forum.puzzlebaron.com/forum/puzzle-baron/acrostic-puzzles/24566-new-scoring

Particularly post #11.

Here's what i see from the thread

"But a very good performance compared to the median solve time will net you roughly 90% of the maximum points allotted"

Don't know what 'good' performance means but when I've been 1.5 times the average I gotten 30% -40% of the maximum and when I am twice the average i get 10% to 20% of the maximum. I don't understand why the actual algorithmn is such a secret.

[waldo]

Just asking for a friend, but if you were given the algorithm, how would it change the game for you? You still have to complete the puzzle quickly to score more points regardless of the number. Assuming all all scored on the same basis, the method is of no consequence.

[duhmel]

Just asking for a friend, but if you were given the algorithm, how would it change the game for you? You still have to complete the puzzle quickly to score more points regardless of the number. Assuming all all scored on the same basis, the method is of no consequence.

Just curious since I am spending a lot of time on these puzzles and I would like to understand the method of scoring. Right now the scores seem quite random. I don't understand why the algorithmn is a state secret.

[redpicker]

Just curious since I am spending a lot of time on these puzzles and I would like to understand the method of scoring. Right now the scores seem quite random. I don't understand why the algorithmn is a state secret.

To take a quote from the other thread, post #11, that you apparently didn't see

... The specific formula may still be tweaked some based on real world stats as we continue to grow them.

If the admin were to define the specific formula, what would be the process be to modify that formula? Would they have to post the proposed changes so that (non-paying) members could criticize them? IMHO, the admin has no obligation to define any specific scoring scheme and have the ability to change it whenever and however they want. Hopefully, these changes will increase the enjoyment of the players (and result in more players).

As stated in the other thread, the algorithm is intended to remove the advantage a previous scoring scheme gave to those solving only short puzzles, and, I suppose, to encourage players to solve longer ones. In my opinion, it has accomplished that, without specifically discouraging players from playing only shorter ones, if that is what they want. This new scheme is apparently based on statistics, which is, understandably, proprietary information to the owners of the site. There should not be an expectation that the players be privy to such information since this site is funded by ads, the revenue from which are based on the statistics.

[AirRandom]

To take a quote from the other thread, post #11, that you apparently didn't see

If the admin were to define the specific formula, what would be the process be to modify that formula? Would they have to post the proposed changes so that (non-paying) members could criticize them? IMHO, the admin has no obligation to define any specific scoring scheme and have the ability to change it whenever and however they want. Hopefully, these changes will increase the enjoyment of the players (and result in more players).

As stated in the other thread, the algorithm is intended to remove the advantage a previous scoring scheme gave to those solving only short puzzles, and, I suppose, to encourage players to solve longer ones. In my opinion, it has accomplished that, without specifically discouraging players from playing only shorter ones, if that is what they want. This new scheme is apparently based on statistics, which is, understandably, proprietary information to the owners of the site. There should not be an expectation that the players be privy to such information since this site is funded by ads, the revenue from which are based on the statistics.

While we're at it, can we get some insight as to how the monthly "Fastest Scorer" challenge values are computed? Is that average over all the puzzles they solve that month? If so, is that similarly weighted against the length of the puzzle? If it's not, that seems biased toward people who choose to only solve shorter puzzles.

[briang]

While we're at it, can we get some insight as to how the monthly "Fastest Scorer" challenge values are computed? Is that average over all the puzzles they solve that month? If so, is that similarly weighted against the length of the puzzle? If it's not, that seems biased toward people who choose to only solve shorter puzzles.

Yeah, it would make a lot more sense if it were computed curved based upon the number of clues solved rather than strictly by the number of puzzles.

[brosis88]

It seems to me that some of the fastest solvers must type continually to solve that fast. Frankly, it just doesn't seem possible to solve any puzzle in about two minutes.

[AirRandom]

It seems to me that some of the fastest solvers must type continually to solve that fast. Frankly, it just doesn't seem possible to solve any puzzle in about two minutes.

It's possible if you only solve short puzzles.

[duhmel]

It seems to me that some of the fastest solvers must type continually to solve that fast. Frankly, it just doesn't seem possible to solve any puzzle in about two minutes.

It's very possible if you have multiple accounts - solve first the normal way and then go to a different account and then just type in the answers. Why people would go to that effort to 'win' a non prize is beyond me.

[brosis88]

duhmei's answer seems the most logical. Some of these fastest solvers must have multiple accounts, solve normally, then encounter the same puzzle again and just type in the solution. These folks are making the rest of us look slower than we are, which is pretty despicable.

[AirRandom]

duhmei's answer seems the most logical. Some of these fastest solvers must have multiple accounts, solve normally, then encounter the same puzzle again and just type in the solution. These folks are making the rest of us look slower than we are, which is pretty despicable.

I haven't encountered the same puzzle twice here (although I haven't been playing as long as some of the top players have), and the puzzle number isn't exposed anywhere on the page... so this explanation would have to involve some serious page hacking in order for it to work. Seems unlikely that multiple people are doing it like this.