|
< Earlier Kibitzing · PAGE 2 OF 14 ·
Later Kibitzing> |
| Feb-13-04 | | Benzol: <Benjamin Lau & ughaibu>
The point you two make about postal games is very relevant. I have posted a number of games of Cecil Purdy's and these games actually took years to complete. His preliminary games started in 1947 and finished in 1950, then the championship games started in 1950 but didn't finish until 1953 when he finally won the championship. Check out his profile. |
|
| Feb-13-04 | | S4NKT: Ughaibu > here's a site concerning the game you mentioned > http://www.correspondencechess.com/... Lau > No.. well.. I didn't mean immeadiately but if that's the case it would be interesting to know, if that was the case then every country would have a 'comparison rating' attributed to it, It might exist unofficialy at least. But yes after playing several games against players on an equal rating in a different country is what I meant. Concerning what you said about Fischer, he didn't really have too much resistance against the stronger players either, he only lost 2 games to Spassky in the '72 match for example, he only lost 81 games in total according to chessgames.com, If he was a kid today it would be another story, but the same applies to all masters from the past. Fischer himself said that amongst the old masters he thought that Morphy would fare the best if he was alive today and learned all the modern chess theory. Concerning what you said about Honza that's a good example, I guess players from strong chess countries could take a holiday to a weak chess country and play in a few tournaments there to get a few extra points added to their rating, if this theory is sound. |
|
| Feb-13-04 | | Benjamin Lau: <Concerning what you said about Fischer, he didn't really have too much resistance against the stronger players either, he only lost 2 games to Spassky in the '72 match for example, he only lost 81 games in total according to chessgames.com> Granted, but a lack of losses isn't the only factor you should realize. If Fischer had played in the USSR championships and tournaments instead, there would have been much more draws and his rating would not have become so bloated (didn't Fischer win the US championship like 6 times or more or something along those lines?) Btw, did you calculate 81 losses including simuls + non-classic time controls? If you did, then his real number of losses is actually probably less. I think one day they should have a group of similarly rated players between each combination of countries duke it out to determine the "exchange" rates for conversions. As it is, a high rating in say maybe Alaska is meaningless in comparison to a slightly lower one in perhaps New York. Thanks Benzol, yes I did check out Purdy's profile... I wonder what chessgames.com will do to correct the dates. Will they show a range (i.e. 1972-76), an average (1974), the first year (1972), or the last year in which a game was completed (1976)? |
|
| Feb-13-04 | | zorro: On the topic of ratings around differents countries, I think that a key factor is the number of players a country has. In a big country with many players the inflation of ratings will go at a higher speed than in small countries with few players. A chief arbiter told me some time ago that this factor is particurarly heavy in the US, to the extent that if a real comparison between USCF players and Fide players should be made the latters should be awarded a 80/100 pts. |
|
| Feb-13-04 | | ughaibu: Zorro: That isn't borne out by the Claude Bloodgood situation, rather the opposite. |
|
| Feb-13-04 | | zorro: Who is Claude Bloodgood and what is the situation he's in?
Anyway, I referred what an arbiter told me, it may as well prove wrong. |
|
| Feb-13-04 | | ughaibu: Claude Frizzel Bloodgood |
|
| Feb-13-04 | | zorro: <ughaibu> thank you, now I understand Claude Bloodgood situation. But I don't see why it should be contrary to what my arbiter said. And what is more the flaw he exploited is not mentioned. |
|
| Feb-13-04 | | ughaibu: Your arbiter was suggesting the larger the pool the faster the inflation, no? Claude Bloodgood's prison was a particularly small pool and this was the cause of the inflation. I think the problem comes from the ratings having floors, I cant see any reason to have floors can you? |
|
| Feb-13-04 | | zorro: <ughaibu> I don't know what to say. On one hand what my arbiter told me seems only logical, on the other hand you are saying that CB reached a high rating by playing the same few players over and over. Are you sure that that was the flaw he exploited? It seems strange 'cause following the Fide calculating system this cannot happen (but I reckon that USCF calculating system may be different). |
|
| Feb-13-04 | | ughaibu: The truth is I dont know how the ratings are calculated but from what I think I understand the problem is the floor. If there were no floor a larger pool would have a larger range so at the extremes players of the same level would have higher or lower ratings in the larger pool but there wouldn't be inflation. I can see how the floor causes inflation but I dont know how it interacts with pool size. I think there's a link on the first page of the Claude Bloodgood thread that goes into the process. |
|
| Feb-13-04 | | ughaibu: Maybe it's because with a smaller pool the difference in ratings between the players is on average larger and because of the floor this gives them greater steps up than occur in a larger pool with closer ratings(?) |
|
| Feb-13-04 | | zorro: <ughaibu> I'm not sure I understand what you mean by "floor". A sort of barrier between ratings? And they seem to work in a funny way: <and because of the floor this gives them greater steps up>. This cannot happen here. When your rating is much greater then your opponent's you simply don't get points, or very few, when you win. |
|
| Feb-13-04 | | ughaibu: The floor is a lower limit below which a player's rating doesnt go, lose as much as they like. |
|
| Feb-13-04 | | zorro: oh, I see. Now I begin to understand where the flaw is. But this doesn't directly affects my arbiter reasoning about inflation. To start with I don't think that a small pool necessarily sees a greater gap between ratings than a greater one. U can have a small pool with all the players tied around a rating x, and a greater one with players spreading from 1500 to 2600.
But more, inflation simply depends on the number and frequency of people playing. Rather, I'd say that the case of CB shows that in US, because of floors, you can have a sort of inflation effect even in small pools. |
|
| Feb-13-04 | | ughaibu: These days with internet site specific ratings manipulation there's all sorts of stuff the workings of which I dont understand. One is "bottom feeding", I dont know why but it seems that by playing much lower rated players it's still possible to increase one's own rating. This may have been Claude Bloodgood's situation if he was the only strong player in the prison. My view is that regardless of the mechanics the conclusion must be that ratings have serious shortcomings. |
|
| Feb-13-04 | | marcus13: I love his game against Karpov, such a strategical game for so yuong player. |
|
| Mar-14-04 | | Benjamin Lau: <Honza>
You're a serious correspondence player aren't you? Would you mind putting Andersson's high (2800+) correspondence rating into perspective? It's supposedly the highest currently. Is it the highest ever? Do you think Ulfy fully deserves it? Do you think that Andersson would do well against the corr world champs? So far, Andersson hasn't lost a single corr game in any of the tournaments he has attended (although he did come close once T Hamarat vs Ulf Andersson, 1996), but I don't know if has entered any world class corr tournaments. Have you analyzed his corr games before? Just curious, thanks in advance for any reply. |
|
| Mar-14-04 | | Benjamin Lau: Incidentally, this is my favorite Andersson game: Ulf Andersson vs Leko, 1996. Very nice work, Leko was young but still very strong in the endgame in 1996. |
|
| Apr-03-04 | | babakova: Uffe! Sveriges bäste schackspelare genom tiderna |
|
| Apr-04-04 | | acirce: Long live the (other) great Ulf ;) |
|
| Apr-04-04 | | nikolaas: <translated from babakova> Swedish best chess player ever.
(If it's translated wrong don't hesitate to tell it, I was just trying) |
|
Apr-09-04
 | | Honza Cervenka: <Benjamin Lau> Ulf is undoubtedly very strong player and he is certainly one of best players who are now playing correspondence chess seriously, but his present rating (2737) is calculated from 35 games only. (By the way, it is not the highest rating now as Joop J. van Oosterom has rating 2748 calculated from 172 games.) It is possible that Ulf's rating over 2800 from years 1999-2000 was a little bit exaggerated thanks to his excellent performance in few contests with small total number of played games. Of course, it does not mean that he cannot be the best. For Top 200 players list see http://www.iccf.com/downloads/topli... |
|
| Apr-09-04 | | Benjamin Lau: Thanks for your concise and objective reply Honza. Does rating mean anything at all in correspondence chess or do you find that it tends to be a bit misleading? Umansky for instance is only ranked 11th on that list, but he won the Jubilee Corr (I think that's the name) tournament in which former CC champions participated. |
|
| Apr-09-04 | | ughaibu: Those ratings do seem rather odd, Oim in 133rd place and I didn't notice Penrose at all. |
|
 |
|
|
< Earlier Kibitzing · PAGE 2 OF 14 ·
Later Kibitzing> |
Kibitzing
