Actualizing the categories scoring rules

Case number:671071-995827
Topic:Game: Other
Opened by:Bruno Kestemont
Status:Open
Type:Suggestion
Opened on:Tuesday, August 27, 2013 - 12:53
Last modified:Monday, December 28, 2020 - 15:02

There is a big difference in categories scoring rules, sensitive to the frequency of puzzle posts.
Some categories rankings last for months, other last only 2 weeks.

Prediction:
-"Your best 8 out of the last 10 Puzzles";
-latest 10 puzzle: 08/22 to 08/07 => last 2 weeks only !
-Suggestion 1/8: "Your best 16 out of the last 20 Puzzles"

Beginners:
-"Your best 2 out of the last 2 Puzzles."
-latest 2 puzzles: 08/15 and 08/15 (actually the same puzzle)=> In practice, rank 1 is limited to 100 => equivalent to "Your last puzzle".
-Suggestion 2/8: "Your best 2 out of the last 4 Puzzles"

Design:
-"Your best 8 out of the last 10 Puzzles";
-latest 10 puzzle: 05/31 to 08/27 => last 3 months
-Suggestion 3/8: -

Symmetry:
-"Your best 5 out of the last 10 Puzzles";
-latest (25) puzzle: 03/14 to 08/27 => last 6 months (5 allows to be ranked 1 within 3 months)
-Suggestion 4/8: -

Electron Density:
-"Your best 8 out of the last 10 Puzzles";
-latest (25) puzzle: 08/15 to 01/06 => last 7 months
-Suggestion 5/8: -"Your best 5 out of the last 10 Puzzles";

Hand-Folding:
-"Your best 8 out of the last 10 Puzzles";
-latest (25) puzzle: 04/27 to 08/12 => last 4 months
-Suggestion 6/8: -

CASP ROLL:
-"Your best 8 out of the last 10 Puzzles";
-latest (25) puzzle: 02/27 to 08/22 => last 6 months
-Suggestion 7/8: -"Your best 5 out of the last 10 Puzzles";

CASP:
-"Your best 100 out of the last 100 Puzzles"; "every 2 years"
-latest (25) puzzle: 07/13/12 to 04/05/13 => last 9 months or more ? this is puzzling
-Suggestion 8/8: -"Your best 5 out of the last 10 Puzzles"

(Tue, 08/27/2013 - 12:53  |  6 comments)


Joined: 09/21/2011
Groups: Void Crushers

I whould rather like to see something like:
The best 80% of a category played in the last 3 months with an automatic extend of that period when a minimum number of puzzles is not played in the last 3 months. That would prevent the big jumping effects like we see for instance on the 5 of the last 10 rule or now in the prediction category.
And for the beginner category something like you proposed.

brow42's picture
User offline. Last seen 43 weeks 1 day ago. Offline
Joined: 09/19/2011
Groups: None

I believe CASP should count for the entire competition. Yes, this does mean that you have to do every puzzle. But it's a competition with a well-defined start and end. Earlier scores should count for the duration of the competition. Also, CASP puzzles can be very different. I did well on the knots, not so well on the giant helix accordions that I could barely wiggle, and down-right poorly on the refinements. My score shouldn't be based on the puzzle type that I'm best at.

Also, I now wonder what happens if puzzles have different point values. If I get #1 on 10 10 point puzzles I have 100 points. If I get #5 on 3 100 point puzzles, I'll be way past that.

spmm's picture
User offline. Last seen 3 weeks 1 day ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

When the categories were introduced IIRC the idea was to give us more puzzles - now there are six beginner puzzles which can only be played by people with <150 points, and only 3 ordinary puzzles. We seem to have less rather than more puzzles.

spmm's picture
User offline. Last seen 3 weeks 1 day ago. Offline
Joined: 08/05/2010
Groups: Void Crushers
Joined: 09/24/2012
Groups: Go Science
Status: Open » Closed

after 2 years, this suggestion did not convince a lot of people. I close it.

Joined: 09/24/2012
Groups: Go Science

Just to mention that it would be fine to re-activate categories ranking (on player's page). It was motivating + it was useful to find experts in each category.

It was introduced here on Thu, 06/07/2012:

https://fold.it/portal/node/992875

Then it disappeared.

Sitemap

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons