Upgrade to Chess.com Premium!

what's the main difference between a 1300 and a 1800 player?


  • 3 days ago · Quote · #601

    Ed_Seedhouse

    Optimissed wrote:

    To try to explain it simply, you have the elo ratings, which, in theory, predict the win ratio between players according to the points difference. You are assuming that the model is perfect and that it applies perfectly all along the rating continuum.

    Um, I'm not sure who you are talking to here, but if it was me I certainly never said nor implied that Elo's model was "perfect".

    It is, however, a fact not a theory, that Elo's system predicts that the probability of winning losing or drawing is dependant only on the rating difference between the two players.

    Whether or not that prediction is correct or incorrect is a matter of empirical observation.  I doubt that Elo's system would have survived so long if that prediction was far off the mark, though.  Several sports have adopted, adapted, and used Elo.  Nate Silver's site uses plain jane Elo to predict NFL football and NHL hockey performances for example.

    No one in here has actually provided any evidence that Elo's assumptions are wrong, only ungrounded assertions and anecdotes.  Anecdotes ain't evidence.

    I am not claiming that Elo is "right", I am only explaining what in fact the assumptions of the statistical model Elo's system is based on actually are.

  • 2 days ago · Quote · #602

    DjonniDerevnja

    Ed_Seedhouse wrote:

    It is, however, a fact not a theory, that Elo's system predicts that the probability of winning losing or drawing is dependant only on the rating difference between the two players.

    I dont think Fide-ELO is a prediction. It is used to divide players into classes, but its not a prediction. ELO is summing up results from the past, and tells how strong the players have been.

    My Elo is based on how good I was as a beginner this winter, and increased when better results ticked in this autumn. It is not the same as my latest results (which more accurately describes my new strenght), but it is between my late results and my winterresults.

    Players that has been on the same level for years do have Elo-rating that is describing their current strenght quite well.

    Kids that is improving very fast are of course very underrated, because they are much stronger now , than they were back in August/September. I guess that todays  Elo might describe quite close how good the player was in August.

    I feel thatChess.com ratings are closer up to date, but they are also describing strenghts of the past, but a bit closer past.

  • 2 days ago · Quote · #603

    Ed_Seedhouse

    DjonniDerevnja wrote:
    Ed_Seedhouse wrote:

    It is, however, a fact not a theory, that Elo's system predicts that the probability of winning losing or drawing is dependant only on the rating difference between the two players.

    I dont think Fide-ELO is a prediction. It is used to divide players into classes, but its not a prediction. ELO is summing up results from the past, and tells how strong the players have been.

    What you think is beside the point.  Read Elo yourself before you make claims about his ideas. 

    Have you actually read Elo?  Well I have.  "The Rating of Chess Players, Past and Present" is still in print and may be purchased from Amazon. http://www.amazon.ca/Rating-Chess-Players-Past-Present/dp/0923891277

    It is simply an observable fact that he designed his system to make performance predictions based only on the difference in ratings.  You can deny this reality all you want, but you are living in a dream world if you do.

  • 2 days ago · Quote · #604

    DjonniDerevnja

    Ed_Seedhouse wrote:
    DjonniDerevnja wrote:
    Ed_Seedhouse wrote:

    It is, however, a fact not a theory, that Elo's system predicts that the probability of winning losing or drawing is dependant only on the rating difference between the two players.

    I dont think Fide-ELO is a prediction. It is used to divide players into classes, but its not a prediction. ELO is summing up results from the past, and tells how strong the players have been.

    What you think is beside the point.  Read Elo yourself before you make claims about his ideas. 

    Have you actually read Elo?  Well I have.  "The Rating of Chess Players, Past and Present" is still in print and may be purchased from Amazon. http://www.amazon.ca/Rating-Chess-Players-Past-Present/dp/0923891277

    It is simply an observable fact that he designed his system to make performance predictions based only on the difference in ratings.  You can deny this reality all you want, but you are living in a dream world if you do.

    I think I have Elo.  I didnt know it was a man. I thought it was ratingnumbers or a ratingsystem. Maybe its both? Heres my fidecard: http://ratings.fide.com/card.phtml?event=1524283

    My fiderating is1422.

    I have not read Elo, so I am talking about what I believes chessratings describes.

    I think its strange to call it predictions, but at the same time I think the ratings is the best statistics we have to build predictions on. I will not say its predictions. It is numbers you achieve AFTER scoring aginst other players with such numbers.

  • 2 days ago · Quote · #605

    DjonniDerevnja

    ED_Seedhouse, thanks for the ELO-book link. Interesting!  :-)

  • 2 days ago · Quote · #606

    Ed_Seedhouse

    DjonniDerevnja wrote:

    It is simply an observable fact that he designed his system to make performance predictions based only on the difference in ratings.  You can deny this reality all you want, but you are living in a dream world if you do.

    I think I have Elo.  I didnt know it was a man. I

    So you are, in other words, almost entirely ignorant about the rating system and how it works and how it came about.  But this doesn't stop you from making pronouncements about it in spite of your ignorance.

    Fortunately there is a cure for ignorance, but you'll have to make the decition to implement it on your own.

  • 2 days ago · Quote · #607

    DjonniDerevnja

    Ed_Seedhouse wrote:
    DjonniDerevnja wrote:

    It is simply an observable fact that he designed his system to make performance predictions based only on the difference in ratings.  You can deny this reality all you want, but you are living in a dream world if you do.

    I think I have Elo.  I didnt know it was a man. I

    So you are, in other words, almost entirely ignorant about the rating system and how it works and how it came about.  But this doesn't stop you from making pronouncements about it in spite of your ignorance.

    Fortunately there is a cure for ignorance, but you'll have to make the decition to implement it on your own.

    I didnt know the background,

    It is not logical that rating is predictions.

    Everybody I know have got their numbers AFTER their performances. You dont have to read a book too see that.

    Do you disagree that rating is describing a strenght , or performance from the past? My FIDE rating is 1422, and it is set from results from the past, actually from results between when I restarted competing in january up to ca the end of november.

    If I should predict ratingscore, I guess it will be closer to Fide 1600 than 1422 in the next tournament. That is prediction. 1422 is new history/statistics/old facts.

    I admit I have not read anything from ELO, how could I, I didnt have a clue that Elo was a name. But what I write does make sense, or what?

  • 2 days ago · Quote · #608

    DjonniDerevnja

    Ed_Seedhouse, I found this:

    http://en.wikipedia.org/wiki/Elo_rating_system

    Interesting.  :-)

  • 2 days ago · Quote · #609

    AnarchyBrian

    Well personally I went from about 1200 to 1400 in a few months. The difference is now I know a few more complicated openings, I recognize more tactical positions and I recognize opponent mistakes better. That's probably the main difference between any two ratings.

    The difference is the 1800 player will probably beat the 1300, but who knows? Maybe the 1300 is better and will win?

  • 2 days ago · Quote · #610

    DjonniDerevnja

    AnarchyBrian wrote:

    Well personally I went from about 1200 to 1400 in a few months. The difference is now I know a few more complicated openings, I recognize more tactical positions and I recognize opponent mistakes better. That's probably the main difference between any two ratings.

    The difference is the 1800 player will probably beat the 1300, but who knows? Maybe the 1300 is better and will win?

    1300s that is better than 1800 are usually kids, maybe at the age of 9,10 or 11. I am a middleaged fastrising man, and I am at 1422 worse than 1800s , and probably closer to 1600 level, while Linnea, 9 years old at 1425 Fide is significantly better than me, more like 1700.

    I guess Magnus Carlsen, when he got 1300, can have been above 1800 strenght, because he elevated from ca 900 to 1900 in a year.

  • 2 days ago · Quote · #611

    AnarchyBrian

    DjonniDerevnja wrote:
    AnarchyBrian wrote:

    Well personally I went from about 1200 to 1400 in a few months. The difference is now I know a few more complicated openings, I recognize more tactical positions and I recognize opponent mistakes better. That's probably the main difference between any two ratings.

    The difference is the 1800 player will probably beat the 1300, but who knows? Maybe the 1300 is better and will win?

    1300s that is better than 1800 are usually kids, maybe at the age of 9,10 or 11. I am a middleaged fastrising man, and I am at 1422 worse than 1800s , and probably closer to 1600 level, while Linnea, 9 years old at 1425 Fide is significantly better than me, more like 1700.

    I guess Magnus Carlsen, when he got 1300, can have been above 1800 strenght, because he elevated from ca 900 to 1900 in a year.

    That's a good point. Some 1300 players have potential to reach 1800 or higher but may not have experienced as many mistakes and have not learned certain aspects of the game yet to allow them to improve.

  • 2 days ago · Quote · #612

    DjonniDerevnja

    AnarchyBrian wrote:
    DjonniDerevnja wrote:
    AnarchyBrian wrote:

    Well personally I went from about 1200 to 1400 in a few months. The difference is now I know a few more complicated openings, I recognize more tactical positions and I recognize opponent mistakes better. That's probably the main difference between any two ratings.

    The difference is the 1800 player will probably beat the 1300, but who knows? Maybe the 1300 is better and will win?

    1300s that is better than 1800 are usually kids, maybe at the age of 9,10 or 11. I am a middleaged fastrising man, and I am at 1422 worse than 1800s , and probably closer to 1600 level, while Linnea, 9 years old at 1425 Fide is significantly better than me, more like 1700.

    I guess Magnus Carlsen, when he got 1300, can have been above 1800 strenght, because he elevated from ca 900 to 1900 in a year.

    That's a good point. Some 1300 players have potential to reach 1800 or higher but may not have experienced as many mistakes and have not learned certain aspects of the game yet to allow them to improve.

    I think the very few 1300`s that is at 1800 strenght have low rating because it reflects their previous gameresults. So if this number tells us about half a year ago, they might have improved that much since then. Those 500 points are  a lot, and you shal improve at the speed of Magnus to do it in six months (which I guess is a normal timedelay between FIDE-ratingstrenght and official achieved FIDE-rating at that strenght).

  • 2 days ago · Quote · #613

    Ed_Seedhouse

    DjonniDerevnja wrote:

    So you are, in other words, almost entirely ignorant about the rating system and how it works and how it came about.  But this doesn't stop you from making pronouncements about it in spite of your ignorance.

    Fortunately there is a cure for ignorance, but you'll have to make the decition to implement it on your own.

    I didnt know the background,

    It is not logical that rating is predictions.

    Your rating is simply an average of your previous performances.  The Elo system assumes that your future performance will vary according to the standard or gaussian distribution around you "real" playing strength, of which your actual current rating is an extimate based on a sample.

    If this assumption is true then you can indeed predict future performances according to the current estimate of your "real" strength.  You can predict, for example, that about 70% of your performances will lie within +- 200 rating points of your "true" strength.  If you consistently perform better than your rating implies then your rating will increase to the point where the statistics of the normal distribution will predict your performance accurately.  That is, the estimate of your "real" strength will be adjusted to fit the observed facts.

    Elo ratings actually do in fact predict future performances quite well, so this lends evidence to support the belief that the underlying assumptions are in fact true, or at least a pretty good approximation to the truth.

    If your argument were true then, for example, election polls would also be way out when in fact when well done they are actually quite accurate.

  • 2 days ago · Quote · #614

    Elubas

    Yes, you must take into account how young the person is, because young people can improve fast. But if you have played 200 games, and your rating is at one particular area, and you predict your rating will increase 100 points suddenly next month, I'm more inclined to think you are just overestimating yourself (due to the temptation) rather than your guess being very rationally based (indeed, we don't really know what improves rating, do you need to know 20 more concepts, 50, something else? Which kinds of concepts, etc etc.). Indeed, the best kind of information we have for future ratings is past results/ratings. People will go nuts and say "oh but the future might be different from the past." Well, sure, but you have to go by something lol. One's rating can change but I don't expect it to increase 100 points in one day :)

  • 2 days ago · Quote · #615

    infrasonic_panoply

    This forum is mis-named, I learnt nothing about what the title says, apart from the obvious that a higher rating has more chance of winning.   Wow what an insight.

  • 42 hours ago · Quote · #616

    DjonniDerevnja

    Ed_Seedhouse wrote:
    DjonniDerevnja wrote:

    So you are, in other words, almost entirely ignorant about the rating system and how it works and how it came about.  But this doesn't stop you from making pronouncements about it in spite of your ignorance.

    Fortunately there is a cure for ignorance, but you'll have to make the decition to implement it on your own.

    I didnt know the background,

    It is not logical that rating is predictions.

    Your rating is simply an average of your previous performances.  The Elo system assumes that your future performance will vary according to the standard or gaussian distribution around you "real" playing strength, of which your actual current rating is an extimate based on a sample.

    If this assumption is true then you can indeed predict future performances according to the current estimate of your "real" strength.  You can predict, for example, that about 70% of your performances will lie within +- 200 rating points of your "true" strength.  If you consistently perform better than your rating implies then your rating will increase to the point where the statistics of the normal distribution will predict your performance accurately.  That is, the estimate of your "real" strength will be adjusted to fit the observed facts.

    Elo ratings actually do in fact predict future performances quite well, so this lends evidence to support the belief that the underlying assumptions are in fact true, or at least a pretty good approximation to the truth.

    If your argument were true then, for example, election polls would also be way out when in fact when well done they are actually quite accurate.

    I follow you better now, and it did help to read the wikipediaarticle,

    What I meant was that rating isnt prediction, but aftergamesstatistics,

    I understand that ratings are the statistics that is the best tool for making predictions, and I think its almost spot on for experienced players, and a bit too old to describe players that is improving too fast.

    My impression is a bit off, because I play in an environment where players have faster ratinggrowth than normal (my club got three Norwegian champions and two silver in the kid championship, and those kids are flying for example from  1000 to  1500 Nsf-elo very fast).

    That kidboom is kidding with the statistics; Isak Sjøberg was at 1063 NSf-elo in the clubchampionship this winter, in november he got 1409 (and 1632fide). I am fresh in chess and play in the kid-class (c) All the ratings in the clubchampionships gets far behind the normal Norwegian level, because I cant beat a 1063 Kid with 1063 strenght when he actually is far above 1200. I went to a tournament in november (Kiwi-Konnerud), with a nsf-elo below 1000, and no players below 1600 could beat me.

  • 22 hours ago · Quote · #617

    eastyz

    I think the difference between a 1800 player and a 2300 player is that the 2300 player spends is energy playing and improving his chess rather than on blog tennis.

  • 9 hours ago · Quote · #618

    DjonniDerevnja

    eastyz wrote:

    I think the difference between a 1800 player and a 2300 player is that the 2300 player spends is energy playing and improving his chess rather than on blog tennis.

    And the difference between Fide 1422 and Fide 2300 is probably that the 2300 in stead of relaxing with internetdebates  works out physically, and gets som fresh air. A 2862 does play a lot football and basket. Actually no players in the history of chess have ever reached 2860 fide without playing basket and football (soccer). In Norway there is a top athlet school, and that school have classes for both chess, football and basket, and a lot of other sports. The Icehockeyplayer  Mats Zucarello Aasen  also went to that school. GM Simen Agdestein is teacher there.  

    The difference between 1800 and 2300 also might be the difference of a first year and a last year chesstudent at this school (NTG)

  • 9 hours ago · Quote · #619

    Benzodiazepine

    DjonniDerevnja, excellent post!

    I think the old saying still goes: healthy mind in a healthy body.

    If your body is healthy then your mind will be healthy, too.
    Fat people are, usually, not the brightest.
    Not eating breakfast and lunch doesn't help them the slightest bit.

    Exercise is what helps.

     

    I noticed that my cognitive performance increased dramatically once I started cycling again and now that I go to GYM I can learn and comprehend much quicker and better.

  • 4 hours ago · Quote · #620

    DjonniDerevnja

    Benzodiazepine wrote:

    DjonniDerevnja, excellent post!

    I think the old saying still goes: healthy mind in a healthy body.

    If your body is healthy then your mind will be healthy, too.
    Fat people are, usually, not the brightest.
    Not eating breakfast and lunch doesn't help them the slightest bit.

    Exercise is what helps.

     

    I noticed that my cognitive performance increased dramatically once I started cycling again and now that I go to GYM I can learn and comprehend much quicker and better.

    You are right, I will both start at the gym early january, and go skiing. :-)


Back to Top

Post your reply: