|
< Earlier Kibitzing · PAGE 27 OF 58 ·
Later Kibitzing> |
Dec-21-06
 | | Sneaky: <AdrianP> Yes, of course. So when you read figures like "30%" or "37%" that doesn't mean a whole lot, but I assume he set levels within the bounds of what would be considered normal, for example going by FIDE ratings and using the standard formulas for calculating win/loss/draw based on ratings. |
|
| Dec-21-06 | | acirce: Also, if it's mostly done to compare different systems, which is the most interesting in my opinion, I guess it doesn't matter (so much?) |
|
Dec-21-06
| | Sneaky: Right, acirce. I'm sure you'd witness large differences when the settings are near the extremes (e.g. if you stipulate that there is a player who can win 99% of his games) but for most sensible middle-of-the-road settings you probably will witness roughly the same results. Of course I'm just guessing, you never really know for certain until you actually do the work. When I get some time maybe I'll try to reproduce Fischl's work on the subject. |
|
| Dec-21-06 | | AdrianP: <Sneaky>
Player A: 2800 ELO
Player B: 2700 ELO
Players C-ZZZZZ: all 2600 ELO
Field 1 is Player A + Players C-ZZZZZ
Field 2 is Player B + Players C-ZZZZZ
Say System 1 favours the strongest player more than System 2 given Field 1. I tentatively think that that doesn't necessarily mean that System 1 favours the strongest player more than System 2 given Field 2. Or does it? I.e. whether or not one system favours the strongest players over another system may depend on the extent to which the player is stronger than the field. |
|
| Dec-21-06 | | YouRang: <An interesting experiment would be to define 127 players are mathematically equal, and place one "superplayer" in their midst, who is clearly stronger than the others. We could then figure out what the chances are that the superplayer becomes the challenger. You'd like to hope that the superplayer would have a better than 50% of succeeding but in my gut I am very skeptical that this would be the case.> In order to have a 50% chance of winning the KO tournament, the 'superplayer' would need to be strong enough that his/her probability of winning each KO match is about 90.6 percent. So how strong does he/she need to be to have a 90.6% winning percentage in each KO match? It depends on how many games are in each match, the probability of draws, and how tiebreaks work. The potential for draws and ties makes it a trickier calculation. |
|
Dec-21-06
| | Sneaky: <Say System 1 favours the strongest player more than System 2 given Field 1. I tentatively think that that doesn't necessarily mean that System 1 favours the strongest player more than System 2 given Field 2. > You're absolutely right. You might think that one would imply the other, and in many cases it would, but "it ain't necessarily so." In the studies I did with the 24-game system vs. the unlimited match format, I found that the skill of the challenger was paramount in determining which system was more favorable. A challenger only a tiny bit better than the champion would prefer, slightly, the traditional format--but if the challenger is a good bit stronger than the champion the he or she should prefer the unlimited format. This is very much like your comparison of two players at different rating levels. |
|
Dec-21-06
| | Sneaky: YouRang, when you say 90.6% are you thinking about the 2004 FIDE 128-knockout or the new proposal, where 128 players are divided into 16 groups of 8 players each? The new system is much harder to evaluate mathematically. What are the odds that the strongest player in a field of 8 will win an all-play-all round robin? And while we're on the subject of round robins, anytime you have an all-play-all, there is this ugly factor of coloration. The whimsy of coloration decisions can have a huge impact on the final results. You know what I mean: there are two giants in the field, and when they face off the guy with the White pieces has all the trumps. If he's behind and he needs a win, he has a chance to do it--and if he's already ahead and just needs a safe draw, well it's pretty easy to make a draw with the White pieces. |
|
| Dec-21-06 | | YouRang: Oh, sorry. Yeah, I was just thinking of 128 players playing 7 rounds of KO matches until there was one (winner) left. Anyway, it demonstrates why the old FIDE knockout system was a bad idea. |
|
Dec-21-06
| | Sneaky: FIDE wants to make chess like the World Series of Poker: a new champion every single year! |
|
| Dec-21-06 | | twinlark: More, if the rule allowing 2700+ players to challenge is implemented. It could become a rotating turnstile. |
|
| Dec-21-06 | | YouRang: <Sneaky: FIDE wants to make chess like the World Series of Poker: a new champion every single year!> Is this bad? I feel kinda strange about this, but I think FIDE may have come up with an idea that I like! Of course, it's more that offset by the horrible "2700+ challenge rule" idea, which robs the championship cycle of it's meaning. |
|
Dec-21-06
| | Sneaky: I sort of like the 2700 rule, because it's tied to the funding of the match. Does this mean that one player might become champion because they can raise the money, while another more worthy player can't come up with the funds? Sure it does. But what's so wrong about that? What chess needs is a large injection of money. In my opinion some of FIDE's biggest problems stem from the backwards policies which could only be conceived by some communist/socialist mindframe. Ideas like "minimum prize purses" remind me of Fidel Castro's stupid ideas about "rent control." It sounds great in principle, but in the end everybody lives in slums. |
|
| Dec-21-06 | | Archives: <FIDE wants to make chess like the World Series of Poker: a new champion every single year!> I'm gunna raise it up...
Im ALLLLIN baby! |
|
| Dec-21-06 | | twinlark: I don't like the possible strain the 2700 rule puts on the the WC. Why have an elimination process if any 2700 could challenge? |
|
| Dec-22-06 | | whatthefat: <twinlark: I don't like the possible strain the 2700 rule puts on the the WC. Why have an elimination process if any 2700 could challenge?> Agreed. It seems like a relic of the chaos, and I assume if FIDE has any brains (hah!) they'll eliminate it in the near future. Otherwise what's the point in the whole Candidates cycle? And when in the 2-year cycle would these additional matches happen anyway? |
|
| Dec-22-06 | | YouRang: <Sneaky: I sort of like the 2700 rule, because it's tied to the funding of the match.> Well, we disagree on this point then.
As far as funding goes, getting a million bucks every 6+ months is peanuts. What chess needs is corporate sponsorships. I think a good marketing agency could look at chess and see some serious potential. Major chess events are followed by millions of fans around the world, and better yet, there is a positive qualitative aspect to this audience since chess is associated with intelligence. I think many major corporations (Microsoft, Apple, Sony, etc.) would be interested in making an impression upon millions of intelligent people around the world. But first, chess has to be a fair and reputable product, and right now, it isn't. It has corrupt and inept management, and as pointed out by twinlark and whatthefat, the 2700+ rule robs the whole WC elimination cycle of much of its significance. Worse, the idea that the title can be up for grabs to the highest bidder destroys the sense of fairness. All this gives chess a bad smell that will keep corporate sponsors far away. I think Bessel Kok understood all this. Kirsan hasn't a clue. |
|
| Dec-22-06 | | acirce: <I think Bessel Kok understood all this. Kirsan hasn't a clue.> What's the difference? http://www.chesscenter.com/twic/bes... |
|
| Dec-22-06 | | YouRang: <acirce> Interesting article, thank you. I was unaware of this development. However, I'm not sure that it suggests that there is no difference between the two. It suggests (to me) that Kirsan and Kok are willing to cooperate, but as far as I know, Kok has not retracted his criticisms of FIDE management made during the FIDE presidential campaign. (Or has he?) |
|
| Dec-22-06 | | acirce: I think it means they have laid their differences to rest and realized that the best thing is to work together. This is nothing new that came all of a sudden from nowhere, of course. Kirsan called for unity from election day on and there have been rumours about various kind of plans. FIDE's Presidential Board meeting in Elista Sept 22-23 mentioned what looks like what's now being announced. http://www.fide.com/news.asp?id=1113
http://www.chessninja.com/dailydirt... from Sept 22 refers to a Dutch article <As described, FIDE would put Kok in charge of all professional chess operations, including the current world championship cycle. You know, the one with the disappeared candidate matches that lead to the San Luis-style world championship tournament in Mexico City next Fall. (Also the Olympiad and the women's and junior's titles.) FIDE president Kirsan Ilyumzhinov and Kok met last Friday in Ukraine to discuss it. Amsterdam would likely be the base of this new operation.In the article, Kok said "I let him know my conditions and he shook my hand and promised he would come back to me with a complete proposal. It's mad, but everything is bizarre in FIDE. We had a deal about this in 2002 and nothing came of it."> Curiously, the Olympiad is not mentioned in the Memorandum. The 2002 deal he is talking about is of course the Prague Agreement - it's interesting that this was also signed in Prague - calling among many other things for <The professional chess world requires a professional management body to be constituted on the basis of a business plan, which will be submitted by Bessel Kok and to be discussed and agreed with the World Chess Foundation and the FIDE Presidential Board within 90 business days after May 6, 2002. on approval of the business plan, FIDE shall issue a license (excluding Einstein Group pre-unification events) to this body to manage professional chess as a profitable business.> http://www.chessbase.com/newsdetail... Kok was out-maneuvered then, but I think FIDE has more to gain this time. |
|
| Dec-22-06 | | YouRang: Okay, well this is encouraging, I think. But the 2700+ rule still reeks, IMO. :-) Do you know if Kok has taken a position on it? I would be surprised if he approves of it. |
|
| Dec-22-06 | | acirce: Well, the 2700+ rule is scandalous and should obviously be removed after Mexico to reinstate fairness in the cycle. I also doubt Kok has approved it, but I don't remember him clearly denouncing it when it was introduced. Can anyone fill me in? I'm not sure FIDE intended to keep it anyway, though. The suspicion was that the rule was mainly introduced to be able to arrange the (de facto) unification match without having to recognize Kramnik as a World Champion. |
|
| Dec-22-06 | | twinlark: <acirce>
During his election campaign, Kok drew up a proposal for a fully revamped World Championship cycle which had no mention of the 2700+ rule. The proposal for the men's championship would have been a very neat and elegant format. The women's cycle was completely different. I asked Susan Polgar about it on her player page, but she didn't want to comment until after the election results as she planned to speak with the winner on the subject. Kok was looking to construct a sensible world championship cycle and one can safely assume the 2700+ rule would have been inconsistent with his proposal, and therefore did not even enter the picture. Having said all that, I don't recall he specifically repudiated it, but it's difficult to imagine in the cirucmstances that he would have retained it. |
|
Dec-22-06
| | Sneaky: <As far as funding goes, getting a million bucks every 6+ months is peanuts. What chess needs is corporate sponsorships.> That's a reasonable point. However, I still don't see the 2700-rule as a provision which sells the title to the highest bidder--all it does is force the champion to defend his title. I don't know the details of the 2700-rule, but if it has proper constraints (limitations on how many times per cycle it can be invoked, limits on how many times the same player can do it, etc.) I don't see any great harm in it. Besides, may actually get to watch the Topalov-Kramnik rematch in 2007 because of this provision. I don't know about you, but that's one rematch I'd like to watch. It's certainly not the worst thing that FIDE has done lately. |
|
| Dec-22-06 | | square dance: <Besides, may actually get to watch the Topalov-Kramnik rematch in 2007 because of this provision. I don't know about you, but that's one rematch I'd like to watch.> havent had your fill of slavs yet? ;-) i dont mind that a challenge can be made to the champion, but i dont quite like how fide has set it up. 2700 is not a very meaningful rating in terms of wc material. something closer to 2750 would be more realistic imo. i also dont like the idea that the champion is required to defend his title upon being presented with a challenge especially if he is to defend his title every other year along with competing in supertournaments, etc. i think there are some positive though that have already been noted such as the money being brought in and the excitement in the chess world over a wc match. |
|
Dec-22-06
| | Sneaky: If I could come up with the $1M I think Kramnik should be forced to defend against ME, dammit. Who cares what the rating of the challenger is? If the challenger is strong, we'll see a fight--if he's weak, then we'll have an exhibition. The biggest problem with the 2700-rule is, in my opinion, just like you say: the champion is forced to overdefend his title. In 100 years we've gone from Lasker defending once every decade or so, to now when FIDE forces him to defend every other month. Here's a thought: maybe FIDE is just trying to push Kramnik out by bombarding him with title defenses until he loses, then they'll change everything around, again, once they get him out? |
|
 |
|
|
< Earlier Kibitzing · PAGE 27 OF 58 ·
Later Kibitzing> |
Kibitzing
