College Policy Debate Forums
November 24, 2017, 05:59:13 PM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: IF YOU EXPERIENCE PROBLEMS WITH THE SITE, INCLUDING LOGGING IN, PLEASE LET ME KNOW IMMEDIATELY.  EMAIL ME DIRECTLY OR USE THE CONTACT US LINK AT THE TOP.
 
   Home   Help Search Login Register CEDA caselist Debate Results Council of Tournament Directors Edebate Archive  
Pages: [1]
  Print  
Author Topic: statistical analysis -- power matching distorting win-loss records  (Read 1789 times)
coach_hanes
Newbie
*
Posts: 7


« on: November 02, 2015, 02:58:21 PM »

Hi everybody,

I analyzed the 2014-2015 debate season (varsity only). There's evidence that power matching substantially pushed up or pulled down teams' win-loss records, perhaps altering many teams' rank by 4 percentiles. Possible solutions include: (a) matching every preliminary round randomly; (b) using weighted records as the first method of ranking teams, because weighted records seem to be less distorted by power matching; or (c) using Elo or Glicko ratings to track teams throughout the year and to determine rankings at individual tournaments.

Please read the paper attached for a three page summary.

Russell

* The Problem with Power Matching.pdf (109.14 KB - downloaded 550 times.)
* debate data crunching.xlsx (1171.06 KB - downloaded 265 times.)
* methods.pdf (95.57 KB - downloaded 278 times.)
« Last Edit: November 03, 2015, 05:03:40 PM by coach_hanes » Logged
Adam Symonds
Sr. Member
****
Posts: 349


« Reply #1 on: November 03, 2015, 01:17:35 AM »

I'd like to see the entire argument laid out, rather than a 3 page summary. I have a number of unanswered questions about methodological assumptions.
Logged
coach_hanes
Newbie
*
Posts: 7


« Reply #2 on: November 03, 2015, 08:47:19 AM »

Sure. I will type it up. It should take me a couple of days.
Logged
coach_hanes
Newbie
*
Posts: 7


« Reply #3 on: November 03, 2015, 05:04:29 PM »

OK, I already had most of the methods typed up, so it was easy to finish. I attached it to the original post.

If you have questions, please feel free to ask.
Logged
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2013, Simple Machines
SMF customization services by 2by2host.com
Valid XHTML 1.0! Valid CSS!