« Collective Choice: Rating Systems | Main | On Being an Angel »

January 03, 2006


Feed You can follow this conversation by subscribing to the comment feed for this post.

F. Randall Farmer

A data point on ELO cheating for you: Yahoo! Games uses ELO rankings for several their two-player games. Before recent abuse mitigation changes, some people used robot to accumulate scores in excess of 6,000,000 points. The abuse-the-ranking game had become a totally seperate competition.

For now, Yahoo! has capped the ELO scores at 3,000 (I think.) This removed most of the cheating incentive.

Christopher Allen

In the slashdot discussion of this blog entry http://games.slashdot.org/comments.pl?sid=173988&threshold=0&mode=flat there was an interesting link to beatpaths.com, which is sort of a tourney system for when you can't complete a full set of round-robin or double-elimination competitions, as what happens in NFL during the fall season.

What is also interesting to me is that it introduces a goal for ranking systems that I'd not thought of before -- prediction. The purpose of the beatpath systems is actually focused on predicting the outcome the next set of the weekend games.

Stephen Waits

No mention of Glicko and Glicko2?

Edward Vielmetti

There's a ranking scheme that Mark Newman from the U of Michigan did for college football, which Yahoo Search turned up a pointer to here:



Physicists Juyong Park and M.E.J. Newman aimed for a ranking system that's fast and easily understood by fans (in contrast to the cumbersome, opaque BCS formula). They based their ranking method on the notion that, if A beats B, and B beats C, then A also beats C, even if it may not actually play C. Hence, the method counts both direct wins by beating a team and indirect wins by beating a team that beat another team.

(end quote)

You end up with a directed graph of teams as nodes and games as edges, and compute a function of that graph that is related to a centrality metric.

Matthew Weigel

This article is a great survey of the subject! I'm working on the PvP ranking system for another game with some unique problems, because the teams in a PvP match can be of disparate size and widely differing power levels, which are then balanced so that wildly mismatched teams (ignoring relative rank at PvP for the moment) are actually reasonably matched.

It starts here: http://ncanson.livejournal.com/3938.html

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

My Photo