Something to remember about Weera and NASPA's rankings

[andysaunders]

You don't get a ranking unless you're active.

You're not active unless you've played 28 fully-rated games in 24 months.

Weera's not going to be "top 10" or "top anything" until after he's completed another multi-day, or a couple of one-days.

Comments

[quinquennia]

For the sake of argument, Weera's initial rating could have been >2300 if he had played a 1900s opponent and won that game. John chew estimated his chance of beating a 1900s opponent would have been 40%.

28 games isn't going to straighten out a 2300+ rating with the lower ratings multiplier, especially if Weera plays at a 1800-1900 level.

Another fun one to think about: When Ross Brown was new to the Scrabble scene, his first tournament was the Toronto Open, and his first opponent was 1700s. He won the first game, and had an initial rating of 2387 or something ridiculous (700 points above his opponent rating) after just one game. What would have happened if he called in sick for the rest of his games? Or if he had to leave because of a family emergency? I assume he'd be rated 2387. He could play several tournaments slowly frittering away those rating points and become the top player after 28 games quite handily by playing his cards right.

I'm thinking for a new player, the first 20 games or so, no matter the number of tournaments participated in, should be counted as one tournament. A five game tournament (or a one game tournament) is too short to assign an initial rating. Something to discuss with the ratings committee or NASPA, whomever does that now.

[jigsawn]

Isn't there a higher ratings multiplier when you're still in your first X games? I think X is about 50.

[jigsawn]

If it wasn't clear, this is because ratings are supposed to be more volatile in the early going. It sometimes takes a while for a player to have played enough games for the rating to settle at the level it should be. Weera's case is not an uncommon occurrence in Scrabble or chess, nor probably any other game with an Elo-like rating system.

[quinquennia]

yes there is, but it's been awhile since I've seen the chart. I think it was posted on the old NSA site, but I can't find it on the NASPAWiki. This is all I get:

"The K value, or sometimes called the Multiplier, is usually 20. However players with less than 50 tournament games will have a higher K value, the idea being this helps accelerate them to their “proper” rating. Also, once a player's rating goes above 1800 or 2000, their K value drops to 16 or 10, respectively."

Needs clarification.

I believe the chart with the old ratings system clarified this:
something like
Under 51 games and rating over 1800: K=32
Under 51 games and rating over 2000: K=20

[andysaunders]

http://www.scrabbleassociation.com/ratings/how.html

Your multiplier is looked up in the following table:
Games Played
1-49 50+
pre- 1-1799 30 20
tourney 1800-1999 24 16
rating 2000+ 15 10

[quinquennia]

Thanks, I just found it here:
http://www.math.utoronto.ca/jjchew/scrabble/ratings/how.html