|
|
|
Topic: average opponents' rating...
| |
|
|
Author
| Message |
|---|
|
|
i have to agree with mandaragit,yes there are some strong players here but if you look at the games of some top rated players they are poorley playd .Why?? the answer is simple they have a high provisional rating and defend their ratings by playing not the strongest players here.If intrested IM's or GM's visit the site and they see this games they gonne ask how this is possibel.If an average opponent rating is availebel they can see where this high rating comes from and look at other games of strong players who have had tougher opposition and also better played games.In my oppinion we have a greater chance like this to have new strong members.
gamelover
|
|
|
Let us be forgiving with ratings and such but consider gamelover's observation:
| if you look at the games of some top rated players they are poorley playd .Why?? the answer is simple they have a high provisional rating and defend their ratings by playing not the strongest players here. |
How right is he! Yes, I have seen some games here where tactical blunders (overlooking a hanging piece) abound. Manipulating ratings as gamelover states is very, very possible. In fact, one of the best players here sports a rating of about 1500. Now how can that be? I honestly don't know but he is very good with openings theory and his games are full of content... Even worse, we have a Fide Master here whose rating doesn't represent accurately his true strength!
Of course the website will depurate itself in a Darwinian way to eventually gather the people that are actually strong and the hackers (I count myself in the latter group but I will give anybody a game that is worth his/her time... Yes, anybody can beat me but he/she will have to sweat blood...) but that will take a long time and lots of games!
Ratings do not represent who is who here! In the meantime, seeing the average rating of the opposition and the winning percentage would lead us to conclude who is who. One example:
Rating = 1950
Played = 60
Won = 20
Drawn = 5
Lost = 35
How good is this guy? Out of 60 possible points he made 22.5 and lost the rest for a winning percentage of 37.5% Now, what sort of opposition did he face? Let's say that the average was 1780
Now, that's a respectable average but two things are glaring at me: He played much weaker opponents and his winning percentage is extremely low. Am I to conclude that this guy is a good player? I honestly would reconsider playing this fictitous opponent. In fact, a guy with a rating lower than his/her opposition's average rating is a red flag for me unless you are Kasparov and as far as I know he doesn't play here.
In any case, this supporting information would be usefull for players aspiring to have more fun (Isn't it fun knowing that you are about to play a good player and that you will get back your time's worth?)
By the way, play higher rated players! It is a joy knowing that you will face trouble and learn in the process as you earn your rating points from people that made you sweat blood...
|
|
|
If you re-read this thread from the beginning you will find no place where I say "no, I do not want to implement this feature".
I actually started thinking in how to compute this average and posted a question to all of you that probably got lost between the many other posts because it wasn't answered. I'm going to copy my question/suggestion in a new thread.
If I can get this feature to a level of simplicity (i.e. no more than a couple of hours of work) I would be more than happy to bump it up in my priority list. If it is more complex than that I will prefer to leave it for later.
Thanks for your comments to all of you. I really appreciate your suggestions, even though sometimes I have to be very critic and make sure it is something worth doing. I also have to understand the feature with 100% detail, for me saying "average opponent's rating" is not enough. I need to think exactly what that means in computation terms, so I have to ask (maybe too many) questions to clarify it.
Thanks,
Miguel
|
|
|
I would also like to make a comment on the rating system. I am rated 1900 Canadian Chess Federation, yet I have a 2100 rating here. Being a new website, most people here have to be patient and let people establish themselves first... Have fun!!!! :-)
|
|
|
A lot of the problems with false ratings being too high will go away after many games are played. Over time, the ratings will become closer and closer to a true rating. A 2000 ranked player with 100+ games under his belt can be assumed to be a good player, but a 2000 ranked player with 15 games may not be. There's not enough statistical information to know. I'm not sure where the threshold is, but I'd guess that at around 30-50 completed games the ranking would begin to be an accurate assessment of their strength.
Of course there's the case where someone with a 2000+ rating only plays much lower ranked people, but so what? That person isn't going accept a game from you anyway. Let them beat up on us not so good players if that's what it takes to make them feel better about themselves.
The only reason I would like to see an average opponent's rating (in whatever fashion you guys think best) is for myself. To see maybe how I do against variously strengthed players.
--p
|
| Previous 1 2 3 4 |
|
|
|
