< Earlier Kibitzing · PAGE 11 OF 11 ·
|Apr-09-18|| ||AylerKupp: <<Nf8> I suppose it depends on how you interpret the 2018 Grenke Classic Rules & Regulations. In my mind when they say:|
"Tiebreakers: 1. Number of wins, 2. Number of wins with Black, 3. Head-to-head score.
In case of a tie for first place: 2 games with 10 min + 2 sec a move; then if needed 2 games with 5 min + 2 sec a move; finally a game with 6 min for White and 5 min for Black (in case of a draw Black wins)"
To me that means that to determine the winner of the tournament you first look at the number of wins in the tournament by those players that tied for first place. If two or more players that tied for first place have the same number of wins, then they look at the number of wins that these players had with Black. If two or more players that tied for first place have the same number of wins with Black, then they look at the head-to-head score between the players who tied for first place. Only if these first 3 tiebreakers don't determine a winner do the remaining players engage in Rapid (semi-rapid?) and Blitz games.
That, of course, is just my interpretation, but if the intent of the organizers was to go directly into the Rapid and Blitz games to determine the winner of the tournament, why bother with listing the number of wins, number of wins with Black, and head-to-head score as tiebreakers?
Maybe this was the result of some cutting and pasting from an earlier Grenke Classic. I looked at the tiebreak rules for the 2017 Grenke Classic and they were the same as for the 2018 Grenke Classic. There was no Grenke Chess Classic held in 2016, but the wording of the tiebreaks in the 2015 Grenke Classic Rules and Regulations was identical to the wording of the tiebreaks in the 2018 and 2017 Grenke Chess Classic Rules and Regulations. I couldn't find the 2014 Grenke Chess Classic Rules and Regulations.
But, interestingly, the 2013 (the first time the tournament was held) Grenke Classic Rules & Regulations say "Tiebreakers (except for first place): number of wins, wins with Black, head-to-head score. Only for a tie for 1st place: a play-off with 2 games at 10 min + 2 sec/move; then if scores remain tied, 2 games at 5 min + 2 sec/move; if still equal an Armageddon game will be played at 6 min for White vs. 5 min for Black + 2 sec/move (in case of a draw the player with Black wins the tournament)"
That I think is much clearer, although I'm not sure why they would bother with tiebreakers for places other than first place. So I think that you're right and they intended in 2018 to go directly to the semi-Rapid, Blitz, and sudden death games for those players who tied for first place. And over time they decided to shorten the tiebreaker description to the point where, I think, it became ambiguous.
And, of course, it makes no difference now. We'll have to wait until next year to see if it makes a difference then.
|Apr-10-18|| ||Nf8: <AylerKupp> I happen to know with somewthing like 100% certainty that there was supposed to be playoff in case of a tie for 1st place, since they actually mentioned it in the live broadcast of the final day as information coming directly from the organizers; three major websites which are in touch with the organizers – chess.com, chess 24 & chessbase – mentioned that in their reports of the penultimate round as what's going to happen in case of a tie; and (as I already noted) there's the precedent of GRENKE Chess Classic (2015), where Carlsen & Naiditsch went to a playoff regardless of the technical tiebreaks mentioned in the regulations. |
It's really a bit strange that they had the clearer phrasing in the 2013 regulations and then changed it to something a bit more ambiguous. Usually the texts of previous years' regulations don't appear anymore, but I remember that when tournaments – like Norway chess – started using playoffs as tiebreaks for 1st place a few years ago they sometimes had this kind of somewhat ambiguous phrasing and later changed it to a clearer one (see http://norwaychess.no/en/regulations, sections 8 & 9). In general, super-tournaments are moving to playoffs instead of technical tiebreaks to decide 1st place (the only ones that still stick with the "archaic" method are the Candidates & Dortmund), but – as you can see in the current Norway chess regulations (section 9) – sometimes they still mention technical tiebreaks for places other than 1st. Not sure why, since usually it doesn't affect money prizes; perhaps to determine who will participate in a playoff in case there's a tie involving more than two players, perhaps simply for some kind of "good measure".
|Apr-10-18|| ||cunctatorg: He is the "top nerd chess player" but I like him because he is pretty inventive, honest and quite a fighter!!|
|Apr-10-18|| ||The Kings Domain: Fine game. Vitiugov seemed to have things going for him in the kingside but Caruana was just one step ahead with his dominance in the center.|
|Apr-10-18|| ||Ulhumbrus: 5...Qd7!! obstructing Black's own QB looks like a brilliant innovation and an alternative answer to the threat of Nxe4. Now on 6 Nxe4 dxe4 7 Qxd7+ Bxd7 develops the bishop.|
8...Ne6! occupies a blockade square and perhaps in this case the blockading knight works in the way in which Nimzovich would have wanted it to work.
If 10 Ng5?! makes the move f4 more difficult an alternative is 10 c3 and Nfd2
11 Nxe6?! moves the knight a third time to exchange itself for a knight moved four times so it gains a tempo, but it also strengthens Black's centre by drawing the f7 pawn on to e6 and it is a sign that the opening has gone wrong for White.
After 11...fxe6 any advantage may already have passed to Black. The pawn structure looks like that of a French defence or Caro Kann defence which has gone wrong for White. Black's d5 pawn turn outs to be strong instead of weak and so becomes a weapon instead of a target.
|Apr-10-18|| ||AugusteGusteau: Lesson learned|
|Apr-10-18|| ||cormier: |
click for larger view
Analysis by Houdini 4: d30 dpa done
1. <= (0.00): 14.c3> Bg7 15.f4 0-0 16.Bb5 a6 17.Be2 a5 18.Bb5 Rf7 19.Qg4 Raf8 20.Rae1 d4 21.Bc1 Bh6 22.Bc4 Nd8 23.cxd4 Rf5 24.Bd3 R5f7 25.Bc4 Rf5
2. = (-0.14): 14.f4 Bh6 15.Bb5 a6 16.Be2 a5 17.c3 0-0 18.Bb5 Qc8 19.Bc1 Ne7 20.Qg4 Bc6 21.Bxc6 Qxc6 22.Qh3 Bg7 23.Nd2 Nf5 24.g4 Ne7 25.Nf3 Rad8 26.Qh4 Qd7 27.Rd1 Nc6 28.Be3 h6 29.Qg3 Rf7 30.Kg2
3. = (-0.16): 14.Bb5 a6 15.Bxc6 Qxc6 16.Bg5 Bg7 17.a5 0-0 18.Be7 Rf4 19.axb6 Bxe5 20.Bxc5 Bxb2 21.Rb1 Bc3 22.Bd4 e5 23.Bxc3 d4 24.f3 dxc3 25.Nc1 Rff8 26.Qe2 e4 27.f4 Qc5+ 28.Qf2 Qxf2+ 29.Rxf2 Rad8 30.Na2 e3 31.Re2 Rxf4 32.Nxc3 Rd2 33.Rxd2 exd2 34.Rd1 Rb4 35.Rxd2 Rxb6 36.Na4 Rf6 37.Nc5 Bc6 38.Nd7 Rf4 39.g3 Rf5 40.g4 Rg5 41.Kf2 Bxd7 42.Rxd7 Rxg4 43.Ra7
|Apr-10-18|| ||Conrad93: <1) -0.03 (24 ply) 6. Be2 Nc6 7. O-O Be7 8. c4 Nc5 9.Nb3 dxc4 10.Nxc5 Bxc5 11.Bxc4 Qxd1 12.Rxd1 Bb6 13.Bg5 O-O 14.Bd5 Re8 15.Bxc6 bxc6 16.Rac1 Be6 17.b3 Bd5 18.Be3 Rab8 19.Nd4 Rbd8 20.Nf5 Rxe5 21.Nd4 Bxd4 22.Rxd4 Rh5>|
Not sure which Stockfish you are using, but Houdini thinks White is much better.
|Apr-10-18|| ||Conrad93: 9. Nb3 is not necessary. 9. cxd5 Qxd5 10. Bc4 Qd7 (10...Qd8 is also possible, but the text move gives Black the option of ...Qf5 or Qg4) 11. a3! a5|
click for larger view
12. h3 (to prevent the active 12...Qg4) 12...0-0 13. Qe2
click for larger view
13...Qf5 14. Nb3
click for larger view
White has a slight edge, but Black has his chances.
|Apr-10-18|| ||BOSTER: <Ulhumbrus>:<5...Qd7>.
In many lines of the "Petoff's Defense" white play Qe2, blocking the bishop f1. Why black can't play the same?.|
|Apr-10-18|| ||ChessHigherCat: <Conrad93> I just used the built-in (premium-feature) SF analysis on this website. It stands to reason that Caruana wouldn't play a bogus opening in such an important game, in any case. He's much more likely to have prepared a line that would take his opponent by surprise because it's underrated (by Houdini, for example).|
|Apr-11-18|| ||AylerKupp: <Nf8> I have mentioned elsewhere that I prefer the technical or "archaic" methods rather than playoffs as a means of determining the winner of a tournament. With playoffs at time controls other than the time controls used in the tournament there's always an argument (which I agree with) that games played at Rapid, Blitz, Bullet, etc. time controls are different than games played at Classic time control games, just like games played at Correspondence time controls (with or without chess engine assistance) are different than games played at Classic time controls. While many of the skills needed are the same, many are different and FIDE recognizes that, hence there are different ratings for games played at Rapid and Blitz time controls than for games played at Classic time controls.|
But the technical or "archaic" methods are based on the results achieved in the tournament in question, so they should be more appropriate to determine the winner of that tournament in case of a tie for first place. We could argue indefinitely the "best" order of preference in which of these methods should be applied and, since the players would know at all times whether they have an advantage or disadvantage at all times, whether this would influence their approach in the final rounds of the tournament. Although I don't see anything wrong with that.
Playoffs in general can be a crap shoot, not just because there are greater chances of making mistakes as the time allowed per move are reduced, but because the smaller number of games played in the playoffs decreases the predictability of statistics derived from having played a much larger number of games. Still, knowing the statistics, players who think that have a statistical advantage in playoffs with shorter time controls might choose to go for them, particularly in a match situation.
But playoffs are clearly more attractive to spectators, so I think that they are here to stay.
And regardless of the method(s) used to break ties, I think that they should be clearly stated in the tournaments rules and regulations. They could get complicated if more than two players tie for first place, as witnessed by the tiebreak specifications in the 2018 Candidates Tournament. What would they have done at the 2018 Grenke Classic if 3 or more players had tied for first place and after the Rapid and Blitz tiebreakers had been applied more than 2 players still remained in contention for first place going into the sudden death or Armageddon tiebreaker? God only knows.
It would be ideal if these tiebreakers, regardless of their complexity, could be standardized by FIDE. Then each tournament could simply cite the FIDE regulations on tiebreakers as the method they will use, and the tiebreakers would not change from tournament to tournament. This would certainly reduce the number of questions on that topic asked at this site! And, given that matches are different than tournaments in many respects, the tiebreakers would have to be different since neither total wins nor Sonneborn-Berger would be appropriate. But, given FIDE, I doubt that this will happen anytime soon.
|Apr-11-18|| ||AylerKupp: <Nf8> BTW, I've been meaning to ask the meaning of your user name since there is often a good/funny story behind it. In case you wonder about mine, I am a wine aficionado and particularly fond of German wines. Wines from the Ayler Kupp vineyard in the Saar valley winemaking region were an early favorite of mine.|
Unfortunately the implementation of my avatar was poor since the both the contrast and resolution are poor. I think it was <morfishine> who mentioned that it looked like a washed up diploma and, whoever it was who said that, he is right.
|Apr-12-18|| ||Conrad93: I can't imagine German wine being that good, but then again I've never had it.|
|Apr-13-18|| ||Nf8: <AylerKupp> No interesting story behind my username, I'm afraid, except perhaps a vague memory of Larsen's comment that "you can't get mated with a knight on f8"... Anyway, playoff tiebreaks are indeed in all likelihood here to stay (and soon be implemented in the Candidates as well), not just because they provide attractive drama for spectators but also because players - at least from the younger generation - seem to prefer them (which means two strong incentives for organizers to implement them). A representative opinion is that of Caruana in a recent interview after the Candidates, when asked about the tiebreak system there (https://www.youtube.com/watch?v=v8H...) - he says playoff would be the best by far because it's the most fair; of the technical tiebreaks he says head-to-head is also ok (but often not enough, as we know), whereas number of wins or Sonneborn-Berger (he says Buchholz by mistake) feel completely arbitrary. I also remember Karjakin, in a press conference during the 2014 Candidates, calling the system of technical tiebreaks instead of a playoff "stupid." Especially with regard to Sonneborn-Berger, I think it's obvious why players prefer to feel that their fate is in their own hands rather than be dependent on the results of other players. |
Btw, in previous posts I focused only on the factual question of what tiebreak they would actually be using, not on what tiebreak would be more desirable, but since you mentioned it I happen to fully share Caruana's view about the desirability of direct playoffs - indeed, not just because they provide drama, but because they're more fair, despite the problem of using different time controls. That's since both number of wins & SB seem to me almost entirely arbitrary: in a round-robin, if two players have the same number of points, a bigger number of wins also means a bigger number of losses, and a higher SB means better results vs. higher ranked players in the tournament but also worse results vs. lower ranked players. In both cases I don't see any compelling reasons (at least if we talk in terms of "fairness" or "justice") to reward the former rather than punish the latter. SB does make more sense in a Swiss format.
|Apr-13-18|| ||morfishine: Not all German wines are of the Sweet-White variety. Germany has made great strides over the past 25 years producing a number of Pinot Noir Red wines, though it remains a challenge due to the climate|
|Apr-14-18|| ||qqdos: <Nf8> & <AK> Surely the point of tie-break rules is to encourage participants to play for a win in every game, including and particularly against the better/best opponents; not to settle for "lazy" or strategic draws; not to adopt a passive attitude with Black; and to discourage a non-losing "thou shall not pass" attitude. If a player decides to choose one or a mixture of the above approaches, why is it unfair if he/she loses on tie-break to a player with e.g. more wins? A secondary point is that a rapid play-off game(s) can produce an unfair or anomalous result of a tournament at classic time-limits.|
|Apr-14-18|| ||AylerKupp: <<Conrad93> I can't imagine German wine being that good, but then again I've never had it.>|
It's not a question about German wine being that good, but whether you like them or not. I happen to like them, others don't. But I think that everyone should try them at least once, and there are many different styles.
At the risk of being persecuted by the off-topic police, let me say that German wines, at least in this country (USA), are perceived to be sweet. But there are several degrees of sweetness and, as <morfishine> said above, Germany also makes some (dry) red wine, mostly from the Pinot Noir grape.
But German white wines can be dry also, and these are usually identified with the word "Trocken" on the label. I don't know the ratio of dry German white wine to sweet German white wine and couldn't find it on the WWW but German dry white wine has only recently been imported in quantity. I think I have read that Germans typically keep their dry white wines to themselves and export the sweet wines since dry wines tend to go better with food. But I don't really know.
|Apr-14-18|| ||AylerKupp: <<Nf8> No interesting story behind my username, I'm afraid, except perhaps a vague memory of Larsen's comment that "you can't get mated with a knight on f8">|
A useful thing to know. I think that from now on whenever I play Black I'll give high priority to the sequence ...Nf6, ...0-0, ...d6/d5, ...Re1, ...Nfd7, and ...Nf8. I may not win many games with what is basically a time-consuming and passive setup, but presumably I won't lose any. :-)
As far as technical or "archaic" methods of breaking ties in a tournament, yes I agree that they are all arbitrary. But no more arbitrary than awarding 1 point for a win, ½ for a draw, and 0 for a loss. What I think is important is to determine who played the "best" in the tournament (whatever that means) among all the players that tied for a given position. I am not too impressed with players' feeling that they would like to have their fate in their own hands rather then be dependent on the results of other players. After all, they had the opportunity to "have their fate in their own hands" during the tournament; all they had to do was win all their games! In a round robin doing that will ensure sole first place. And I don't see why a Rapid/Blitz playoff between a player with a high Rapid/Blitz rating and a player with a much lower Rapid/Blitz rating should be considered "fair", except of course in Rapid/Blitz tournaments.
I would think that a technical system like SB could be improved. But it does reward a player for doing well against players that also did well in the tournament in question, and I think that is what's important, not how well they did against players that did well in other tournaments (i.e. based on pre-tournament ratings). I also don't think that it makes sense to reward a greater number of wins or a greater number of wins with the Black pieces, if these came against the players that did poorly in the tournament, regardless of their pre-tournament rating.
Using TPRs might be an alternate approach since it also rewards players that did better than expected in the tournament. But the player with the highest rating would be at a disadvantage since, of necessity, the average score of his opponents would be less than the average score of the opponents of any of the other players. Is this "fair"? I don't know. But then the player with the highest rating is expected to win the tournament outright and, by virtue of his tie, he didn't. Perhaps a case of "better luck next time".
|Apr-14-18|| ||AylerKupp: <<qqdos> Surely the point of tie-break rules is to encourage participants to play for a win in every game, including and particularly against the better/best opponents>|
While I think that encouraging participants to play for a win in every game is important from the perspective of encouraging more exciting chess and increasing the attractiveness of the game, I think that the primary (only?) purpose of the tie-break rules is exactly that, to break ties. Anything else, while possibly desirable, needs to remain secondary. A system that breaks ties in a "fair" way <and> also encourages exciting chess would of course be ideal.
Perhaps the 3-1-0 scoring system (also known as the Bilbao System) as has been used in some tournaments instead of the usual 1-½-0 scoring system, might be useful in encouraging going for wins in every game. Is this "the" solution? I don't know, but for some opinions see https://www.chess.com/blog/SpaceOdd... and https://en.chessbase.com/post/some-....
As to whether a player's choice of a particular strategy is "fair" or "unfair", I don't know either. But as far as your secondary point, whether a Rapid play-off game(s) can produce an unfair or anomalous result of a tournament at Classic time controls, I think that it's more a case of an irrelevant result. What does a set of victories at Rapid, Blitz, or Armageddon time controls have to do with the results of a tournament at Classic time controls?
|Apr-18-18|| ||qqdos: <AylerKupp> You have clearly given this matter (tie-breaking) a great deal of thought. No perfect system exists but I wonder if awarding a minus point or points for a loss might be the answer. For me a win against a "better" player deserves more recognition than a win against a "worse" one. Conversely a loss to a worse opponent might need a greater sanction. How do you define worse and better? It would need fine-tuning and possibly for that reason could be impractical. What about wins and losses with Black?
BTW I fully agree your final question, I don't favour play-off games unless at the same time limits as the main tournament.|
|Apr-18-18|| ||Muttley101: Surely the best looking chess player should win a tiebreak.|
Chess needs more poster people.
|Apr-18-18|| ||AylerKupp: <<qqdos> You have clearly given this matter (tie-breaking) a great deal of thought.> (part 1 of 2)|
Yes, I have. And I hope that at least some of it was good. :-) I agree with you that a win against a "better" player deserves more recognition than a win against a "worse" player, where "better" or "worse" are of course relative. All these players that we discuss are top-notch players!
As far as definitions of "better", two come to mind:
(1) "Better" in the overall sense. Here ratings provide a means to differentiate the "better" from the "not so good". The ratings take into account the results of many games so that, on the average, the player with the higher rating is the "better" player.
(2) "Better" in the sense that they played better than other players in <this> tournament. After all, players' performances vary over time; sometimes a player is in good form and does better than expected in a given tournament and sometimes a player is in bad form and does worse than expected in another tournament. That's why Dr. Elo modeled players' performance using a normal probability density curve.
To me what's important for breaking ties in any given tournament is how well the players involved in the tie played in <that> tournament, given their results in that tournament (and at that time control) and nothing else. That's why I favor Sonnenborn-Berger as a tie breaking measure since a player gets more SB points if they do well against other players that do well in the same tournament and are penalized if they do poorly against players who did poorly in the same tournaments.
I also favor most wins as Black (or probably best <score> as Black, taking draws into account) since it's harder to win or draw with Black than with White. Of course, being in a first place tie and having the edge in the best score as Black tiebreaker of necessity means that the player didn't do as well as they should have as White, so that cancels out.
For the same reason I don't attach too much importance as a tiebreak measure the most wins tiebreaker (encouraging going for wins is a different and desirable issue). If two or more players involved in a tie have the same number of wins and one of them had all his wins with the White pieces, one of them had all his wins with the Black pieces, and the others had a balance between the two colors, then the player that had all his wins with the Black pieces had the more impressive accomplishment.
I also don't attach much importance in the head-to-head competition results because that's often just one game unless it's a double round robin, and in either case that's a small number of games and a poor statistical sample.
Another possibility I mentioned was the use of TPRs since that rewards the players that do better than expected (based on their rating and their results against their opponents) and penalizes them if they do worse than expected against their opponents. The clouding factor is that it bases the results partly on the players' pre-tournament ratings, something that I would prefer to avoid since previous tournament results should not affect the results in a subsequent tournament, and the player with the highest rating would have a handicap since the average rating of his opponents would naturally be less than every other player's average rating of their opponents. Then again, the highest rated player should be expected to win the tournament outright and, if he doesn't, I don't think it's too bad of a thing.
|Apr-18-18|| ||AylerKupp: <<qqdos> You have clearly given this matter (tie-breaking) a great deal of thought.> (part 2 of 2)|
Only if all tiebreak measures are inconclusive do I favor tie breaks at Rapid and Blitz time controls. In my opinion they have nothing to do with the results of the tournament played at Classic time controls.
So, going out on a limb, my preference for ranking the tie-breakers would be as follows:
2. Best score with the Black pieces.
And, of course, that's just my opinion and a minority opinion in all likelihood.
Finally, if all else fails, I don't have a problem in declaring all the players that tied for first place as co-champions. Other than for the Candidates Tournament it is not strictly necessary to have only one player be declared the winner so, since all the with the same score are already splitting the prize money for those places, I see nothing wrong with sharing the title as well.
But, since not only is it more satisfying to have one winner but the fans (and the players as well) prefer using playoffs as tiebreakers, it's clear that playoffs are going to remain the main method of breaking ties.
|Apr-18-18|| ||AylerKupp: <<Muttley 101> Surely the best looking chess player should win a tiebreak.>|
Yes, I agree, and I should add it to my list above. After all, following my "logic", they are the players that looked the best in <that> tournament.
Of course, that then requires determining the criteria for "best looking". And if you thought that the criteria for determining "best player" is difficult, that's nothing compared to determining the criteria for "best looking". :-)
< Earlier Kibitzing · PAGE 11 OF 11 ·