
	T³ - TableTop Tournaments Welcome Guest. Please log in or register. Login with nickname/e-mail and password (Lost password?). Follow us: \| Contact

Choose your country:

Tournaments:
* Home

* Overview

* Calendar

* Map

* ETC

* Share

* Search

Players:
* Login

* Hitchhiker

* List of Players

* Clubs

Organizers:
* Register

* GoePP

* FAQ

NTR/ETR:
* Information

* Ranking

* Clubranking

* Search

* Add Results

* Army ranking

* Hall'o'Fame

Information:
* Tabletops

* Newsletter

* News Archive

* Feeds

* FAQ

About T³:
* What is T³?

* Memberships

* Terms

* Privacy

* Cookies

* Media

* Contacts

News - Upcoming modification of the NTR computation

(Permalink)T³ - Upcoming modification of the NTR computation

2012-02-22 14:57

Based on the running poll about "Team tournaments in the NTR" we're planning to add some modifications to the math behind the rankings.

The current standing of the poll is

* No change: 15%
* No team tournaments: 20%
* Reduced weight: 58%
* No interest: 7%

based on rouhly 350 votes.

So we'll keep the team tournaments in the NTR, but reduce their impact. There are a couple of approaces to do this and we're going to do some math to see how each turns out. Many thanks to everybody who gave some feedback on this.

Working on the math we're (probably) going to simplify the computation a bit and modifiy the weighting factor (WF) so that all 5 results will count as 20% of the final NTR points. This has to be checked with some calculations first.

Written by Blackhawk

Comment...

Comments
#12012-02-22 16:49 YeN In order to reduce the weigh of team tournament, an easy way is to take de number of team (for the tounament size) instead of the number of player. Written by YeN (Contact)
#22012-02-22 16:52 Blackhawk That would be easy, but way too much. Written by Blackhawk (Contact)
#32012-02-27 11:59 Ravajaxe " Working on the math we're (probably) going to simplify the computation a bit and modifiy the weighting factor (WF) so that all 5 results will count as 20% of the final NTR points. This has to be checked with some calculations first. " I would like to make a suggestion on this point. I suggest to take into account the 10 best tournament results, instead of the 5 best. This would dampen the big ups / downs in rankings we see today, as soon as a significant tournament is added to the statistics or gets too old. Furthermore, it will promote a player consistently having good tournament results, instead of players managing a few exceptional ones. Regarding the high number of tournaments saved in the database for people being in top 200 list of their country (WHFB and W40k at least) , it would be appropriate to expand the calculation. NTR includes a regression percentage over time that I feel well designed, unfortunately for a lot of players attending to tournaments regularly, picking only 5 scores to NTR means that even good results get "forgotten" as soon as they get slightly old. Taking the 10 best would fix that too. Written by Ravajaxe (Contact)
#42012-02-29 19:37 Drako What about make an average of places taken by each team considering a non-team tournament? For example in a team tournament with 3 guys team, each member of the first team will have (1st place's point+2nd place's point+3rd place's point) /3 Member of the 2nd team will have (4th place's point+5th place's point+6th place's point)/3... Maybe it had been proposed or maybe it's not possible? Written by Drako (Contact)
#52012-03-01 17:59 Blackhawk >Taking the 10 best would fix that too. On the other side a newcomer would need 10 tournaments to get anywhere near a good ranking the old ones would generate too much points. >What about make an average of places taken by each team considering a non-team tournament? Yeah... I'm working on something like this. At the moment a team tournament is generating too much points if you're looking at the average over all its placements. Written by Blackhawk (Contact)

Comment
You have to be registered at T³ and logged in to write a comment. More info...