|
|
|
Topic: average opponents' rating...
| |
|
|
Author
| Message |
|---|
|
|
more often than not, it is difficult to guess the relative playing strength of a prospective opponent. sometimes, an opponent with a rating of 2000++ plays an uninspired chess while on the other hand, an opponent rated less than 1900, plays big-time chess!!
it's not a secret that a rating could be, intentionally or inadvertently, inflated by playing mostly against lowly-rated opponents. on the other hand, a player with a 1800 rating but with average opponents' rating of 1900++, speaks well of his TRUE STRENGTH.
in this context, i should suggest if it's possible and not burdensome on MIGUEL's part, that an average opponents' rating be reflected in each player's profile. SOME PLAYERS DESIGN THEIR STRATEGICAL PLANS BASED ON EVERY PARTICULAR OPPONENT'S REAL STRENGTH.
thanks MIGUEL and more power to you !!!
|
|
|
I agree with you. The other fact is that players that have considerable knowledge and good technique are somehow hiding because they tend to show their true strength at other websites where they play for money and where they devote themselves with heart and soul. Sometimes, I have seen people blunder here because of trying to play chess while babysitting.
I have run into three stupendous theoreticians here but ratings that do not speak about their true strength... I reserve for myself their names...
I proposed having a quick formula to calculate the winning percentage [ For example: Catalonian has played 10 games of which he has won 7, drawn 1 and lost 2. So that his winning percentage is 75% ] The formula was put into question because it didn't take into account the ratings of the opposition and so an idea of showing the average rating of the opposition was proposed.
|
|
|
hi,
I think must be very difficult to evaluate the rating in a competition whith characteristics postal chess. Level of players is very different in many games and it to cause rating change more.
For example: if I play whith Mandagarit and i win, my rating to ascend more or if I play with a player more under level than more, i to descend more. In other sistems how swiss mode , my rating to ascend o descend one point. And in a anual competition it is more real, than in one game. Perhaps that would be a solution: evaluate in a group of games, and no in one game, whith the swiss mode.
More longer ? 
|
|
|
To mandaragit: are you thinking in the opponent's rating average without taking into account games won or lost? Think of the following example: let's say I have played 30 games to players that are all rated over 2000 and lost all the games, my average opponent's rating will be let's say 2100 while my rating will probably be in the lower 1000s. Is this number still useful?
To Alopinto: I remember that old discussion about the average of wins and/or losses, and I still think it can be misleading. Also players who take risks and play stronger opponents are punished (if they lose as it would be expected). My main concert is that people will look at these percentages without knowing exactly how these numbers are calculated and what they mean.
To Catalonian: My guess is that people prefer instant rating updates. Also, I understand that the Swiss scores are applied within a single tournament and a big part of its workings is that the system tells players who to play against. After a win you face a stronger player, after a lose you play a weakest one. But here you can play anyone you like so I don't see how that could be implemented.
|
|
|
sir MIGUEL, the average opponents' average rating can be accurately assessed only if the win-loss-draw statistics are available.
even in other fields of interest beside chess, average statistics may prove significant to others while others may treat the same as useless.
let's say, player x challenged me to a game. he has an outstanding rating of 2200, but when looking on his past opponents' ratings, they're all below 1800. i accepted the challenge and he plays 1.e4. now, an experienced player, before replying, will assess first the relative strenght of player x. the defensive scheme to be applied against 1.e4 depends on opponent's strength. in my case as black, for example, if the average opponents' rating of player x is 1800, i will reply 1.--- c5. if it is 1900++, i certainly play 1.--- e6.
my proposition is, if ever it is to be implemented, to have a period of test of its usefulness. if majority of players agree, so be it.
sir, statistics are statistics. so far, nobody has reacted negatively except you.
thanks a lot.
|
|
|
Miguel, i didn't said method must be aplicated after each game. But it can possible. But it could to do a regularization after each week ?
Perhaps to many complicated, i understand it.
|
| 1 2 3 4 Next |
|
|
|
