suggest provisional rating marked as such

Sort:
Avatar of normajeanyates

I suggest provisional rating be displayed as eg 1914[P] or something. 

for example i am rated 1914 or so as i write this but ive played only 11 games [won all] -- i didnt know where i stood so i started out so midgame i realised i was playing weak opps.

So 1914 is  misleading. 1914(P) would make ppl look  - and maybe search faqs to see what (P) means, if implemented.

[ i know - glicko and all - it should eventually even out - but at corespondence it may take years plus no site i know including this one uses unmodified glicko - there are adhoc modifications to glicko not backed up by analysis.]

 


Avatar of PawnFork
Until then, visit a tentative opponent's page and check out his current games.
Avatar of Baseballfan
I do know of at least one other site that uses unmodified Glicko, perhaps we should do what they do. They put the players RD as part of their rating, so normajeanyates, if your RD is 160, it would display your rating as 1914 [160]. That way, I could look at your rating a know that you have played a few games, but that you rating is still likely to change considerably for your next few games, win or lose.
Avatar of normajeanyates
Baseballfan wrote: I do know of at least one other site that uses unmodified Glicko, perhaps we should do what they do. They put the players RD as part of their rating, so normajeanyates, if your RD is 160, it would display your rating as 1914 [160]. That way, I could look at your rating a know that you have played a few games, but that you rating is still likely to change considerably for your next few games, win or lose.

 Excellent idea!

 


Avatar of erik
yes, a good idea. unfortunately i think that would cause more confusion than answers :) not sure what to do here... we'll think on it!
Avatar of normajeanyates

confusion can be cleared by faq on:

1. why RD, why not SD?

2. What is the big deal about this bayesian thing? Rev. Bayes's  theorem is that trivial a posteriori probablility theorem, isn't it? We did that at junior high school :D

3. Why do some bayesians get on the defensive and call their probabilities 'likelihoods'? 

Or just post a link to late Ed Jaynes's unfinished public-domain book :)

 


Avatar of erik
lol