Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix freestyle fall count calculation #115

Open
scotthue opened this issue Aug 14, 2014 · 5 comments
Open

Fix freestyle fall count calculation #115

scotthue opened this issue Aug 14, 2014 · 5 comments

Comments

@scotthue
Copy link
Member

From Robin:

Example:
Individual: with 2 major, 1 minor:

  • Mistake Score = 10 - 0.5* major -0.25*minor
  • Mistake Score = 10 - 0.5_2 - 0.25_1 = 10 - 1 - .25 = 8.75
  • Final Dismount Score = 10·(1− mistake score/number of riders)
  • Final Dismount Score = 10*(1 - 8.75/1)
  • Final Dismount Score = 10*(-7.75)
  • Final Dismount Score = -77.5

Group with: 5 major, 4 minor, 10 riders:

  • Mistake Score = 10 - 0.5_5 - 0.25_4 = 6.5
  • Final Dismount Score = 10*(1 - 6.5/10)
  • Final Dismount Score = 10*(1 - 0.65) = 3.5

(this appears to work....but)

Group with 0 major, 1 minor, 10 riders:

  • Mistake Score: 10 - 0.5*0 - .25 = 9.75
  • Final Dismount Score = 10*(1 - 9.75/10) = .25

Why would they have such a low score if they had only a single minor?

Group with 8 major, 0 minor, 4 riders:

  • Mistake Score: 10 - 0.5*8 - 0 = 6
  • Final Dismount Score = 10_(1 - 6/4) = 10_-.5 = -5

This is why this calculation is broken.

I propose: "max(0, 10 *(1 - ((# major falls * .25) + (# minor falls * .125)))/num_members)"

With this, a single competitor would get 10 if they have 0 major falls, 0 if they have 4 major falls. If there are pairs, they get 8 major falls before they get to 0. If there are 3 competitors, it goes to 12 major falls....etc etc.

I look forward to hearing what we actually want to use for the calculation.

@scotthue
Copy link
Member Author

I don't understand why the weighting was changed from the previous version to the update suggestion by @rdunlop.

Minor Falls Major Falls
Old Weighting 0.25 0.5
New Weighting 0.125 02.5

My suggestion is just to correct the mistake in the formula and get the following:

Mistake Score = 0.5*[#major] + 0.25[#minor]
Final Dismount Score = 10*(1- [Mistake Score]/[#riders])

@rdunlop
Copy link

rdunlop commented Aug 14, 2014

@scotthue , the weighting was changed because it felt like giving a score of 0 for 2-major falls was a little harsh. But, I think I'm starting to agree with the corrected formula. (though I would put "max(0, score)" into the calculation instead of only in the text).

If I run my examples again:

Example:
Individual: with 2 major, 1 minor:

Mistake Score = 0.5* major +0.25_minor
Mistake Score = 0.5_2 + 0.25_1 = 1 + .25 = 1.25
Final Dismount Score = 10·(1− mistake score/number of riders)
Final Dismount Score = 10_(1 - 1.25/1)
Final Dismount Score = 10*(-0.25)
Final Dismount Score = -2.5
Final Dismount Score = 0 (cannot be below 0)

Group with: 5 major, 4 minor, 10 riders:

Mistake Score = 0.5_5 + 0.25_4 = 3.5
Final Dismount Score = 10_(1 - 3.5/10)
Final Dismount Score = 10_(1 - 0.35) = 6.5

Group with 0 major, 1 minor, 10 riders:

Mistake Score: 0.5_0 + 0.25 = 0.25
Final Dismount Score = 10_(1 - 0.25/10) = 9.75

Group with 8 major, 0 minor, 4 riders:

Mistake Score: 0.5_8 + 0 = 4
Final Dismount Score = 10_(1 - 4/4) = 10*0 = 0

@gossi
Copy link
Member

gossi commented Aug 27, 2014

There was a formula in the last rulebook committee discussion. I thought this was going into the rulebook, too. Where is the last rulebook committee software running?

@scotthue
Copy link
Member Author

@gossi, the rulebook committee website doesn't work anymore because it was a subdomain of iufinc not unicycling.org. Perhaps you could fix that.

@gossi
Copy link
Member

gossi commented Aug 27, 2014

Ah, we transferred that domain the last night. @olarf is now in charge of that. I hope to move the IUF to its new server by the end of the week. So we should look into this after the migration.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants