ADVERTISEMENT

Wrestlestats methodology

CP84

HB Heisman
Sep 10, 2013
5,869
9,502
113
How does the individual ranking methodology function? I can understand trying to rank wrestlers with a limited sample size this early in the season is difficult but a few things I see defy reason.

Nick Suriano is currently ranked 15th with an 8-0 record and a 7-0 win over #7 Cruz. Dalton Marci at 0-2 is ranked ahead of him.

Jered Cortez is rated #5 with no top 20 wins and losses to the 18th and 19th ranked wrestlers.

Topher Carton is ranked well below Vince Turk.


I like a lot about this site, specifically the head to head comparisons, so I don't want to be too dismissive. I just don't understand how this methodology works and creates such odd rankings.
 
It uses a modified version of the ELO rating system used by some online gaming system (like Yahoo Games) and most notably, Chess. It's supposed to give a rating for a "persons" whole body of work so it takes awhile for someone to move up in the rankings, despite beating other opponents that are ranked ahead of them.

For an example, in Chess. If someone rated a 1200 beats someone that is an 1800, they would probably just move up about 40 points, while the 1800 guy would move down around 40 points.

Whereas, in a real-world ranking system (NOT rating system) people will say, since the 1200 guy beat the 1800 guy, then the 1200 guy should now be ranked ahead of the 1800 guy.

Here's more info on ELO: https://en.wikipedia.org/wiki/Elo_rating_system
 
I agree w the OP. Great site except for the severely downplaying early season results. You really need to change your algorithm calcualtions or nobody is going to pay much attention to the site. Matt Kolodzik at 141 is another glaring example of hurting the credibility of the rankings. Kolodzik has beaten Ashnault (2 out of 3 times in 12 months) and Cruz of Lehigh, yet he's ranked 12, Cruz 11 and Ashnualt 5. Every other site has Kolodzik at 3 or 4.
 
ADVERTISEMENT

Latest posts

ADVERTISEMENT