It isn't exciting to win agaist weeker players
I like to do the same, but sometimes I feel like I'm insulting them. Some of those players had been playing longer than I had, so I feel almost arrogant when I offer to go over the game or to study the game.
If I'm playing someone a bit lower I do like to win in a stylish/cool way. Lots of folk say it isn't fun to play lower rated folk, this is true if that is all your doing. However, everybody has had that lovely ego stroke of cruising a game.
When I'm playing football in the park with my little nephews, all I have to do is kick the ball as high into the air as possible. They then tell me I'm the coolest adult in the world. But it does get boring, and my leg hurts after a while :)
Hi all again! I see a buch of new posts in few hours.
You all discuss about rating at the first place. However, my topic was about players weeker and stronger then you, not rating alone! My first goal is to improve my skill - and I need strong oponents for it.
I also think about programs, I read careffully discusions and am included into it. Maybe, one day, in nxt 6 months or a year, if I stiil be interested to improve my skill in chess, I will obrain Rybka or another program, and use it to improve my skills and to play CC games, when it's legal to use programs. They has some weekneses, as people say, and a "centaur" (man + machine) is still stronger then machine allone! It can be a next step for me.
Rating is more or less acurate measure of your success. I know only a bit about probability and statistics and can't understand what normajeanyates say , but I understand the basic principle. A difference in 100 poinst could be by chance, 200 probably indicate a real difference in power (or, success, if u insist).
There are various habits, of course. @Vance917: you have 74 games in progress - I can't imagine it! I am nearer to Norma Jean, for I concluded 10-12 rated game are optimal. (However, recently I started to play unrated welcome games against newcomers at chess.com.) It's unevitable you made mistakes and blunders sometimes. I play more serious in every game. It make me more sucesfull, unevitable, and my rating higher, but surely not "inflated"; we can say, I suppose, your is deflated.
My level of skil, measured in rating points, is 1900-2100. My result is 25% against people up of my level, 44% against my level, and 91% against people above my level. You can see a point of my topic here: it's not exciting to try to push the last number to 95% or so - but to improve 44% to 60% or so!
lydia, one point i made is : i will give an example:
rating diffence between 1500 and 1700 here is 200 point
rating difference between 1900 and 2100 here is 200 pint (same)
But i think that the probability that 1700 will defeat 1500 (i will guess 90%) is much more than the probability that 2100 will defeat 1900 (I will guess 65%).
About 2100 defeating 1900 i am guessing - i will never reach even 1900-level in my life probably. So what do you think?
A rating difference in the lower ranks is much easier to overcome than in the higher ranks. Once a player gets up to 1900+, their fundamentals should be firmly rooted enough that they are not giving up large upsets. While a 1700 rated player is no slouch, their skill set is not as polished and they are more susceptable to lapses that allow the upset special.
All experience; not always nice though.
ok the link is as follows.
http://citeseer.ist.psu.edu/336524.html
the paper is called:
Parameter Estimation in Large Dynamic Paired Comparison Experiments (1999)
download the Cached PS. (the non-cached doesn't work - it says download stopped [ of course you use freedoenloadmanager for download - no? then i dont know what it'll say :) ]
i havent tried cached PS.gz or cached pdf)
After you download it you'll see that the file is called
glickman99parameter.psOpen the ps (postscript file) from the freeware ghostview (you need to install ghostscript AND ghostview) - it won't open with [at least my] adobe reader.
Cheers!
It starts with what is *wrong* with the Bradley-Terry model :)
UPDATE: the cached PDF also works, but the ps.gz seems to be corrupted. [My other .gz files are opening - with the GPL'ed[*] 7-zip [much more convenient than gunzip.exe], but for this one 7-zip is showing a corrupted [all non-alphanumeric] ps file, which wont open.
[*] GPL= gnu public licence = one of the two popular "copyleft"s - see http://www.gnu.org