I have not got a statistical mind, so I don't pretend to analyse the formula.
But "the rating" is quite an amazing piece of science. If I play someone more 50 places higher I will lose on roughly 90% of occasions, if I play someone 50 places lower I will win on 90% occaisions.
The rating also plays with the mind.
e.g. If I'm losing against a 900 ranked player I will always hang on and wait for the error.
if I'm beating a 1400 player I will do as many quick exchanges as possible in the hunt for the cynical victory.
I apprieciate the rating is just a number, but it is a very clever number.
Your in wonder
Invigorate
I wonder...what is the "if you are x pts higher/lower you are expected to win y percentage of the games".
And I do agree that if I'm playing someone rated lower than me and they have the advantage...I'll hang in longer waiting for the error. Then again, the lower rating is probably reflects their tendency to do just that.
I'm sure my own rating would be higher judged solely on my "best moves" ignoring my blunders.
Nonny
Originally posted by BLReidYou might be playing lower rated people most of the time and have an inflated rating.
Haven't seen the chart, but the rule of thumb that I learned was that a 200 point difference usually means a 2/3 win expectancy. (My own experience has not backed this up though, as I am nowhere near 1/3 against the 1750-1850 crowd).
Originally posted by flexmoreI don't know actually. I'm just going by what someone said.
this is an interesting observation ... many people seem to believe it ... is it true?
or just some kind of illusion.
it runs counter to the intention of the elo system.
I'm talking about the RHP system. Is it the same as the ELO system?
Let's try a hypothetical.
Suppose we take a 1600 named A and have him play, over and over, people with exactly 200 rating points less.
The Win Expectancy for A will be according to the formula 76%. If this is accurate, then, assuming his Loss Expectancy is 24%, he come out even in the long run. He will on average remain a 1600 which is what we want.
If such a person played people 200 points higher, then the WE would be 24%. So that's consistent. And again, A will on average remain at 1600.
So, if there are problems with the system, it's in the Win Expectancy and implied Loss Expectancy figures. Either the fact that the Draw Expectancy is always zero throws off the calculations or the Win Expectancy calculation is simply not accurate based on the statistics. I can't explore this any further as I don't have the numbers I'd need to check to see if reality fits the model. These numbers would be the ratings of both players at the beginning of the game and the result of the game.
Originally posted by invigorateWhat's a "place"?
I have not got a statistical mind, so I don't pretend to analyse the formula.
But "the rating" is quite an amazing piece of science. If I play someone more 50 places higher I will lose on roughly 90% of occasions, if I play someone 50 places lower I will win on 90% occaisions.
The rating also plays with the mind.
e.g. If I'm losing against ...[text shortened]... ciate the rating is just a number, but it is a very clever number.
Your in wonder
Invigorate