Monday, August 1, 2011

Higher rating is stronger

Generally speaking the higher your rating the stronger you are. Ratings are an indicator of strength.. Many of us considered Kasparov the strongest player in the history of chess. Kasparov achieved the highest rating of all time. The two highest rated player currently is Anand and Carlsen, one is the current world champion and the other considered to be a future world champion and one of the strongest active players in the world.

Of course you will lose the odd game against someone rated much lower than you. But over a larger number of games, you will score more.

If ratings are not reliable, then Mas cannot claim to be the best Malaysian player. Going by this logic, the number one Malaysian  player should be the national champion and that is Lim Zhuo Ren. See how ridiculous it is if we adopt this logic?

Of course if the ratings are not in your favour, discount them as not relevant, not important, not indicative and so on. However the whole chess world uses the FIDE rating as indication of strength so if you think otherwise then you are in the minority of which you are the only member.

4 comments:

Ilham said...

All this arguments and counter arguments are like saying GM Ludvig Hammer is the best Norway player because of the reality that he is the best Norway player in European Chess Championship 2008 with a 2792 elo performance. GM Magnus Carlsen did not took part in that tournament.

Hey btw in 1991 National Closed held in Kelantan,i finished 3rd!(Kamal Ariffin first). Jimmy, Peter, Kamal, Eric, Christy etc did not took part. I guess the reality is that mean in 1991, at 17 years old, i am Malaysia number 3. Damn!should have demanded from MCF that i am in Malaysia Olympiad team then.

...but as chess player, i know, i cannot say i am malaysia number 3 player based from one tournament alone. accumulation of result from several tournaments, reflected in rating, is the best yardstick.

Anonymous said...

True enough. Elo ratings are based on the logistic function, which is based on sound mathematical observations and derivation, developed and applied by professional mathematicians, Arpad Elo being one of them.

To cut a long story short, the rating difference helps to project your score against your opponent in the long run.

Of course, other factors have to be taken into account, such as opponent types (think Kasparov-Gulko). Having said so, at the end of the day, if you

placed a 1900 and 2100 player in a random tournament, my money's on the 2100 to finish higher. Over a large number of games, he'll score more. The number

players who the 1900 beats but the 2100 loses to will be much less than the other way round.

Variance has to be taken into account too. A player can play a tournament above or below his current rating (well, that's how your rating changes). The

probability that the strong player performs badly while the weaker play performs like a 'flash in the pan' is not as low as one might think. I said

'flash in the pan' because his win against higher rated opponents may be because they played below their strength.

So it's not so surprising if you see a low seed perform better than a high seed. Unless it happens over and over again. After which your rating does go

up.

In a nutshell, the Elo rating is not perfect; nothing in applied statistics is. But the system works well. It's reliable and considerably accurate.

Anybody who says that a rating means nothing obviously has no knowledge of statistics, and are no different from the wise guys who walk into a casino with their fancy betting systems, thinking they can bring down the house in a game of craps/roulette/.

Jimmy Liew said...

There will be those who twist facts and logic to fit their own agendas. We just have to recognize it for what it is.

Anonymous said...

Never get angry at stupid people

Post a Comment