London Chess Classic 2017 LIVE

Why putting scores 8.5 pts. to Caruana and 7.5 pts to Nepo? In fact it's only 6/9 each? You are confusing your readers. That's not the correct way to put scores in round robin. The tie-break score must not be added to the main score of the round-robin. I still don't like the scoring system...very confusing and very disappointing.

@ Claudioarrau : Taimanov reminds me also - going back many centuries... - François-André Danican Philidor, who, besides being probably the main chess thinker of his time, was also quite an important composer ! Decidedly, classical music and chess go well together !

@ Claudioarrau : Yes, it is quite true that classical music and chess meet quite often ! Myself, I'm interested more or less equally in chess and classical music, and one of my friends was precisely saying the exact same thing to me a few days ago !

Historically, Vasily Smylov comes also to the mind, who was a singer, and hesitated between a career in music and in chess, and Mark Taimanov, who didn't hesitated : he did both ! (How could he, I must say I rather wonder ! Two lifes in one indeed !) The collection "Great Pianists of the 20th Century" even feature a two-pianos album by him and his wife Lyubov Bruk !

Many thanks @Petrarlsen. It's interesting, isn't it, that chess and classical music seem to have a connection, e.g. the matches between Prokofiev and Oistrakh. And the pianist Nikolai Lugansky, a keen chess player, has performed for the players at events in Russia.

@ Claudioarrau : I quite like your approach ! (And, by the way, I quite like also your username ! Nice to see someone who must be quite interested in classical music !...)

After five decades of playing and observing chess, it seems to me that a player’s approach the game is in large part a reflection of their personalities, of which two are the most salient:

1. The player who values the sporting aspect of the game above all else, even it means “winning ugly”.

2. The player who values the artistic aspect of the game above the result, and is happy with a draw if the quality of play on both sides warrants it.

This is not to suggest either of these approaches is more virtuous. No one ever became world champion without the killer instinct. But fighting draws can be just as memorable as scrappy wins. And the best players usually combine the sporting and artistic instincts in equal measure.

But, at the end of the day, it is simply unrealistic to expect players to go against their nature, and play in a style which is foreign to them psychologically. That, in a nutshell, is my objection to attempts to induce players to play more recklessly to please many, but by no means all, the spectators. After all, the beauty of chess, unlike most other games, is that it is a game of logic, not chance. The object of the game is to mate the king. If you can’t do that, logic dictates the game is a draw. And there should be no no disgrace in that.

Bravo and congratulations to Fabio Caruana on his London Chess Classic win. On to the Candidates.......

About the "3 - 1 scoring system debate", the final results of this tournament confirm my hypothesis from one of my previous posts on this page :

This tournament's draw rate is 77.8 %.

The Chess Masters Final 2015 (featuring the 3 - 1 scoring system) had a 83.3 % draw rate (a higher draw rate than the present tournament's draw rate).

And the Chess Masters Final 2016 (also using the 3 - 1 scoring system) had a 76.7 % draw rate (a draw rate comparable to this tournament's draw rate - both are in the 75 % - 80 % range).

Some commentators said in substance, previously, that this tournament's draw rate was absolutely inacceptable, and that an absolute solution would be to use the 3 - 1 scoring system.

Nonetheless, this tournament's draw rate, compared with the draw rates of the Chess Masters Final 2015 and 2016 shows clearly that the 3 - 1 scoring couldn't at all be considered as an absolute guarantee against draw rates at a comparable level (and even higher) to this tournament's draw rate.

So, following these commentators, it would be possible to conclude that the draw rate of a tournament featuring the 3 - 1 scoring system can also reach what they consider to be a completely inacceptable level ; thus, this system certainly isn't an absolute solution to what they consider to be the "draw rate problem".

As for me, I must say that, even if this tournament started slowly, globally, I found it quite interesting to follow, so I don't complain !...

@ lajosarpad : Yes indeed !! And Caruana DID win !...

A system that encourages chess players to take higher risks, in whichever way (3 1 scoring system, prize money, Armageddon, ...) may have the averse effect, in the sense that players may anticipate the opponent to take the necessary risks. I think this may have to do with game theory, or with the "prisoner's dilemma". Or in other words, if taking no risks is the best strategy, then taking no risks is also the best strategy if you raise the stakes.

@Petrarlsen

Indeed and the last round is the icing on the cake. Carlsen has just beaten Aronian with Black, So has beaten Anand with Black and Adams is struggling against Caruana. Can Caruana recover the honor of the White pieces? :)

@ lajosarpad : Yes, I quite agree with your last post !

Meanwhile, after a slow start, this tournament is now very interesting, in my opinion !

"Magnus Carlsen somehow eeked out a win over Michael Adams"

eek: used as an expression of alarm, horror, or surprise.

Is there any video of Magnus doing his eeking?

@Petrarlsen

I think you are right when you state that average draw length and average length of decisive games conducted in separate analysis would yield more meaningful results than in my initial suggestion of getting the average length of games regardless of result. This separation makes sense, as draws and decisive games have different dynamics from this perspective. Also, I think a further improvement would be to gather the deviations in engine evaluation compared to actual results. If a game ends in a draw, but the engine says the position's evaluation is of 0.5, then there is a high chance white did not intend to play out the game. If a game is decisive, but the engines say there is a chance for a draw, then there is a high chance of premature resignation. Also, the data you have provided is interesting and your conclusions are accurate. The fact that a Bilbao tournament had a higher draw rate than this tournament )so far only means that the Bilbao system is not a guarantee for fighting chess. It does not disprove the hypothesis that the Bilbao system motivates players to fight. With examples we could have an argument towards both direction. A large analyses is needed to have some quasi accurate guesstimates about the effects of the Bilbao system.

@ Rooost :

As your question is very short, I am not completely sure to have understood it !

What I understood (if I am right...) is that you would want to download games in PGN, without the players' thinking times.

Perhaps there is a better solution, but one thing works : you click on "Unannotate" (below the game - the "Eraser" symbol), and then, you download the game by clicking on "Save" (on the right of "Unannotate" - the "Diskette" symbol). This will allow you to download a game in PGN without the thinking times.

can I load .pgn without timings?

One little interesting fact, about the 3 - 1 scoring system : for the moment, the draw rate of the present tournament is 78.6 %. This while the Chess Masters Final 2015 - featuring the 3 - 1 scoring system - had a 83.3 % draw rate.

It is obviously necessary to wait until the end of this tournament, but thus, it does seems to be possible that a tournament featuring the 3 - 1 scoring system can have a higher draw rate that a tournament using the 1 - 1/2 scoring system that has been much criticized by some persons for his - supposedly - much too high draw rate (the present edition of the London Chess Classic).

As, furthermore, the Chess Masters Final 2016 - for which the 3 - 1 scoring system was also used - had a draw rate (76.7 %) that was at the same level that the present tournament's current draw rate (both are in the 75 % - 80 % range), if the end of the present tournament confirmed this, we could thus conclude from this that the 3 - 1 scoring system cannot at all guarantee a draw rate within the limits of what is considered as "acceptable" by those who criticize present-day chess for its too high draw rate.

@ lajosarpad :

I think that, globally, we quite agree...

About the question about if the 3 - 1 scoring system could potentially encourage premature resignation of the games, I simply forgot to take this into account !!

But I would rather think that it would be better to evaluate separately these two questions : does the 3 - 1 scoring system has an effect on drawn games ? (longer games, lower draw rate, etc.) ; does this same system has an effect on won games ? (encouraging premature resignation). I think that these questions are separate problems ; obviously, they must both be taken into account for a general evaluation of the 3 - 1 scoring system, but I don't think it would be optimal to simply take the global length of all games, and to conclude directly from this on the general effect of this system. So I think that, as for the question of the length of the games, the best would be to take separately the average length of the drawn games, and the average length of the won games. What do you think of this ?

@Petrarlsen

I did not miss the need to segregate categories in the statistics, but for the sake of simplicity I did not mention this. I solely mentioned that we need to compare large samples of tournaments played in the classical system against large samples of tournaments played in the Bilbao system. I did not explicitly mention the need to avoid comparing apples against oranges, but that does not mean I want to do any invalid comparisons. I only pointed out the need to have large factual comparisons to even have a strong base for an opinion and that's still an opinion, far from a fact. So your point is valid, but is not opposing my opinion. Your point about short decided games is interesting and is pointing to a side of the coin. The other side of the coin is the set of games resigned prematurely without a fight to preserve energy for the next round. In a scale of 0-1 half a point is worth more than in a scale of 0 - 3 a single point, therefore under the Bilbao system one has lesser incentive to fight for a draw. Even a decisive game can be unplayed. Sofia rules and lower move number limits along with the draw percentage are also important aspects of such an analysis. I might be wrongly attributing too much significance to unplayed decisive games though, so you might be right here, but I am not sure about this, I have not enough data to build a well-founded hypothesis. I think we should learn the significance of short decisive games by analyzing them with an engine and see if the end position had drawish tendencies according to engines. Even this analysis has some flaws, as a player might miss the line leading to a drawish play rather than resign prematurely, or the engines might misundertstand a position due to the horizon-effect. Anyway, according to my current knowledge, intuitively I think that premature resignation could be a potential problem under the Bilbao system. I totally agree with you, regarding Koko48. He/she was cherry picking indeed, but we should not question his/her honesty. The problem was the fallacy used by the person, which is not proving he/she is wrong or right, the problem was solely with the reasoning method.

Koko48

You have shown us results which support your claim and ignored results which support the contrary. Again, I am not able to decide whether you are right or wrong with your opinion, I am just stating that if you are right, you are right with the wrong reasoning. At this moment I think it is premature to factually decide whether the Bilbao system increases fighting chess or not. I doubt the efficiency of the system in the prevention of unfought draws, you think it is effective, therefore we have different opinions. But this does not prevent me from respectfully disagree with you. If we can respect the plurality of opinions, then later, when we have more sample data we can come back to the subject and decide it.

@ koko48 : Don't get me wrong : I don't mean that you deliberately chose to eliminate the results of the Chess Masters Final ; I only think that the results of the London Chess Classic interested you VERY much, as you could use them, while, for you, the results of the Chess Masters Final weren't AT ALL interesting, because they absolutely didn't tally with you hypothesis, so you forget them quite quickly.

But in the Wikipedia article cited by lajosarpad, it is explicitly mentionned : "Cherry picking may be committed intentionally or unintentionally.", so, even if you did it completely unintentionally, you are nonetheless cherry picking !

@ koko48 : In one of your posts to me, on the page about the Gibaraltar Open that I cited in my last post, you said :

"Look at the games played under this format, particularly some of the older London Chess Classics (there were other tournaments around then that used football scoring...I believe Bilbao did one year, and I know I'm missing at least one)"

It shows indeed quite well that you are cherry picking (probably rather unintentionally, in my opinion) : at the moment of this old post, you knew quite well that the London Chess Classic wasn't the only tournament that used the 3 - 1 scoring system, but didn't memorize the tournaments whose results don't confirm your hypothesis... You only memorized the London Chess Classic, because you knew that you could use its results in a discussion about the 3 - 1 scoring system (even if, by the way, I don't agree with the significance of these results, because I consider - as I already explained before - that you don't take into account the MUCH wider range of levels in the previous editions of the LCC compared to the 2017 edition, which can perfectly explain all by itself the lower draw rates in these tournaments).

@ koko48 : And you couldn't even say that you chose to take the example of previous editions of the London Chess Classic because we are discussing the London Chess Classic of this year, because, when we discussed previously the 3 - 1 scoring system on a page about the Gibraltar Open (https://en.chessbase.com/post/gibraltar-rd05-peace-and-war), you also cited ONLY the London Chess Classic, and NOT the Chess Masters Final.

So you do this all the time indeed : you keep only the results that confirm your hypothesis, while discarding all the "unsuitable" results... And this does exactly correspond to the definition of "cherry picking"...

@ koko48 :

The Wikipedia article cited by lajosarpad define cherry picking as : "The act of pointing to individual cases or data that seem to confirm a particular position, while ignoring a significant portion of related cases or data that may contradict that position."

Cherry picking is then explained further by the same article : "The term is based on the perceived process of harvesting fruit, such as cherries. The picker would be expected to only select the ripest and healthiest fruits. An observer who only sees the selected fruit may thus wrongly conclude that most, or even all, of the tree's fruit is in a likewise good condition."

Lajosarpad is right, your reasoning is indeed an example of cherry picking : the tournaments that you cite feature systematically rather low draw rates, but it isn't because ALL tournaments with the 3 - 1 scoring system feature rather low draw rates. It is because you eliminated ALL the tournaments whose draw rates were too high, and thus "unsuitable" for your demonstration !

I have done a little research, and I have found two tournaments that illustrate this very well :


- Chess Masters Final 2015 : draw rate = 83.3 % (https://en.chessbase.com/post/bilbao-so-wins-after-tie-break).


- Chess Masters Final 2016 : draw rate = 76.7 % (https://en.chessbase.com/post/bilbao-rd-10-three-draws-carlsen-is-first-nakamura-second).


These two tournaments' draw rates are quite high, MUCH higher than the draw rates of the tournaments that you cited as examples !

Coming back to the Wikipedia definition : "(...) while ignoring a significant portion of related cases or data that may contradict that position.", it is exactly what you did : you didn't kept the tournaments which didn't give "suitable" results for your demonstration !

So lajosarpad was quite right : what you did IS indeed "cherry picking"...

@LajosArpad "Cherry picking some examples are no way proofs to your point."

How is it 'cherry picking' when I gave you four consecutive London Chess Classics - 2009 through 2012 - that used Bilbao Scoring and had drastically higher win rates than this one? And drastically higher win rates than most elite tournaments in general?

And all four had virtually no GM draws?

@ lajosarpad : Another very simple but very important thing that must be taken into account : the use or not of "Sofia rules" ! And, if "Sofia rules" are used, with which moves limit ? 30 moves ? 40 moves ? Quite obviously, this will have important consequences on the length of the drawn games !

One more point : I think that the draw rate must also be taken into account : for example, if, in two tournaments, the average length of drawn games is 25 moves, but in one tournament there is a 95 % draw rate, and in the other a 55 % draw rate, the situation cannot be considered as being the same in both tournaments !

@ lajosarpad : "(...) the best I can imagine you can come up with to support your argument is a large statistical data of all the tournaments played under the Bilbao system, with the average length of the games in comparison of all the tournaments not played under that system with the same data as control pattern."

In my opinion, you miss something important there : the range of levels of the participants must absolutely be taken into account for this study. (How, I don't know exactly, but I'm sure that, as a mathematician, you will have ideas on that subject !) In my opinion, logically, the more similar the levels, the shorter the (drawn) games will tend to be, in average. Two reasons tend to point in this direction, in my opinion, when two players have approximately the same level :

1) Neither of the two players will have an "easy task", to win the game, so they can easily tend not to "push" too hard, as this could generate important risks of losing the game (on the contrary, if you have, for example, a 200 points advantage on your opponent, you will know that if you "push" rather hard, it will - normally !... - not be too dangerous for you, and the chances to see your opponent wilt under the pressure of your play is quite high, so there are high chances that the highest-rated player will continue to play for quite a long time, hoping to win the game - even in a very drawish position, and even quite possibly in a slightly inferior position).

2) Between two players having the same level, a draw is a "normal" result ; you cannot be, normally, extremely disappointed by a draw in these circumstances, and neither player, in most cases, has a very big incentive to win (to say it differently : a draw can never be objectively considered as a catastrophe or a near-catastrophe between two players who have approximately the same level). While if, for example, Carlsen draws against a random 2600+ GM - especially if this has possible consequences on his global results in a given tournament -, I am quite sure he will not be happy at all with such a result ! With a big Elo-difference, even a draw is quite a bad result for the highest-rated player, and I think that this must nearly necessarily tend to generate longer games, as the highest-rated player cannot be satisfied with a draw ; for him, a draw is a "half-loss", and the global tendency will be that, most of the time, he will continue to "push" until there isn't anymore any hope of winning the game.

So, a comparison between the lengths of the games, IF the levels of the participants aren't taken into account, will, in my opinion, always tend to favor the tournaments featuring the widest range of levels for their participants. And thus, it wouldn't permit a reliable comparison between tournaments featuring the 3 - 1 scoring system and tournaments featuring the usual 1 - 1/2 scoring system, because there would be a strong doubt as to wether the difference in length of the games is due to the scoring system or to the different ranges of levels of the given tournaments.

And, by the way, I think that what should be taken into account shouldn't be the length of the GAMES, but the length of the DRAWN GAMES, because a "short win" isn't - obviously ! - less interesting than a "long win" !!

"And throw away a point? Let's be realistic. I don't see that happening. "

If you have a position which is probably lost, then it might be a reasonable tournament strategy to resign it instead of playing it out even if you have drawing chances, in the case you will end up very tired. The strategy would be to sacrifice the point instead of fighting for it, not necessarily being rewarded by it and ending up tired for the next round, possibly losing further points. Note that Karjakin sacrificed a better position, where he might have won. In case he succeeded winning the game, he would have earned half a point under the traditional scoring system, which is worth more than one point under the Bilbao scoring system. So yes, players might be willing to sacrifice potential points due to tournament strategy. Karjakin's example shows that it is not impossible to sacrifice points. Note, that here an example is a proof, since proving that a general statement is false can be done via showing an example, but examples are not proving that general statements are valid. Examples can be used when proving a general statement, but that's mathematical induction, which consists a factual proof that any example will comply to the assumed pattern.

If you have a position which is probably lost, then it might be a reasonable tournament strategy to resign it instead of playing it out even if you have drawing chances, in the case you will end up very tired. The strategy would be to sacrifice the point instead of fighting for it, not necessarily being rewarded by it and ending up tired for the next round, possibly losing further points. Note that Karjakin sacrificed a better position, where he might have won. In case he succeeded winning the game, he would have earned half a point under the traditional scoring system, which is worth more than one point under the Bilbao scoring system. So yes, players might be willing to sacrifice potential points due to tournament strategy. Karjakin's example shows that it is not impossible to sacrifice points. Note, that here an example is a proof, since proving that a general statement is false can be done via showing an example, but examples are not proving that general statements are valid. Examples can be used when proving a general statement, but that's mathematical induction, which consists a factual proof that any example will comply to the assumed pattern.

Guys, let's not insult each-other. It is normal to disagree and to do some debates. I, for one enjoy debates when they are intellectual exchanges.

koko48 you have provided some examples which, according to your point of view support your argument. That's absolutely fine, so long as you treat those correctly and do not attribute them more than what they are: examples. Your general claim that the Bilbao system leads to more fighting chess cannot be proven by examples, even though, examples can support your claim. A proof should point out factually that there is a high probability that all the tournaments played under the Bilbao system will lead to more fighting chess. Cherry picking some examples are no way proofs to your point. I do not say you are right or wrong, I am not making such claims, since I have not seen a factual proof that your point is valid, but I did not see a factual proof of the contrary either. It might be true. Honestly, the best I can imagine you can come up with to support your argument is a large statistical data of all the tournaments played under the Bilbao system, with the average length of the games in comparison of all the tournaments not played under that system with the same data as control pattern. I did not do such statistics, so I do not know the potential result, but, as work hypothesis let's assume the statistics will support your claim. In that case you would have a strong hypothesis supporting your claim, founded on statistical data. But for one, it would be a strong argument, but not a proof and secondly, it will contain all the data, not just the examples which, according to your point of view supports your argument. Cherry picking is a logical fallacy: https://en.wikipedia.org/wiki/Cherry_picking

Also, when you start to argue about the people you do not agree with and assume their goals in the debate to be something is another logical fallacy, called argumentum ad hominem (https://en.wikipedia.org/wiki/Ad_hominem), as the participants of the debates might be right or wrong independently of their goals in the debate:

"because they don't seem to sink in among the more willfully ignorant and hardheaded commentators here"

In a cultured debate it is reasonable to assume that the other participants of the debate are pointing out their honest opinion and if you disagree with that, you can come up with your arguments.

"I've also heard people claim players could collude to 'split the points' in a Bilbao scoring double round robin by mutually throwing games."

It is very much possible, since a player would have a better score with a win and a loss than with two draws under the Bilbao system. I do not know whether this actually happened, but it is a possibility.

@ koko48 : No, I didn't insult you. I really wonder why you say obviously false things like this (and you do it all the time) ; everyone can check and see that what you say is false ; it really is completely absurd ! And I think that to compare the contents of my posts to you to the rather incredible strings of insults that you thrown at me is rather enlightening on that matter ! (For the links to these pages, cf. my previous post.) And YOU insult me another time now by saying I am an "immature troll" ; you prove again and again and again that you cannot discuss with someone who don't agree with you without insulting him !

For the rest, following the same links, everyone can read our exchanges of arguments... And I maintain that, in my opinion, your arguments don't hold water at all ; I'm sorry, but, as for me, I consider that you don't give one single convincing argument...

@ Petrarlsen You initially insulted me when I originally argued for Bilbao Scoring. You said I was "grumbling" and "complaining".

Then you insulted me by responding to everything I said with the same ridiculous, intellectually dishonest arguments. You claimed I gave no proof, after I gave you proof. You claimed you looked at every tournament and every game played under Bilbao Scoring and saw no evidence that it made a difference - which was clearly a lie.

It's obvious I'm dealing with an immature troll who needs to be contrary and get the last word. So I'm going to give it to you. I'm sure you'll respond again and I will not answer, because I'm not interested in filling up this page going around in circles.

My arguments and the proof I gave below, are for the thinking readers and commentators here. Because we have a serious problem in the chess world that needs to be addressed. Saying "everything is fine the way it is" with elite tournaments is not going to increase sponsorship nor lead to increased popularity of the game

@ koko48 :

It isn't because you will state three, four, or five times the same false arguments that they will become true...

One more time, I already answered extensively to absolutely ALL your main arguments.

If you consider that your positions are convincing, this is your opinion, but I completely disagree...

Everyone interested can make his own opinion with the two links that I already gave on this page : https://en.chessbase.com/post/gibraltar-rd05-peace-and-war and : https://en.chessbase.com/post/london-classic-round-4-caruana-breaks-the-deadlock.

A detail, but quite significant in my opinion : each time we exchange on this theme, you end up insulting me (as everyone can see on the two pages that I cite in the present post) ; nearly every time, when a commentator resorts as openly as that to insults, it means that he completely ran out of arguments...

@Petrarlsen Yes, I'm repeating the arguments because they don't seem to sink in among the more willfully ignorant and hardheaded commentators here. Those who don't have an intelligent argument but argue for the sake of arguing and just live to be contrary, take my advice: Being a troll is no way to live life.

You claimed I had no proof of Bilbao Scoring system's effectiveness at reducing 'draw rates', when the proof is right here. I gave it to you before, and you refused to look at it.

Then you claimed the reduced draw rates had to do with a few non-elite players in the field, which is nonsense. The GM draw rate was eliminated, and overall draw rate reduced, among the games between two elite players as well.

So do you have any new arguments, or are you going to keep rehashing the same ones that have already been refuted? I hope this isn't the way you play chess, Petrarlsen

And then you wonder why Karjakin accepts draws in better positions with black, and say "I can't see the objective justification for that!" You are an example of the blind traditionalism among certain people in the chess world, I was referring to

As koko48 always states the same arguments everywhere without stating anything significantly new, and as I have already answered extensively to all his arguments elsewhere, I will not explain my own positions one more time.

But, for those who could be interested, I will point out two pages where we debated on all this at great length : https://en.chessbase.com/post/gibraltar-rd05-peace-and-war (on a page about the last Gibraltar Open) and https://en.chessbase.com/post/london-classic-round-4-caruana-breaks-the-deadlock (on a page about the 4th round of the present London Chess Classic).

Final example: 2012 London Chess classic: 19 of the 36 games (over 50%!) decisive:

http://www.chessgames.com/perl/chess.pl?tid=79427

Does anybody need more proof that Bilbao Scoring raises the fighting spirit of players, and reduces draw rates?

Why was Bilbao scoring scrapped, despite these drastically improved results? I believe it was due to the blind traditionalism among the decision makers in the chess world, combined with some vocal players who opposed Bilbao scoring. Probably because they didn't like having to work so hard and play for a win every game. They like their extra 'rest days', and prearranged draws in the last rounds.

@lajosarpad "the problem is with unplayed games which comes from the players' lack of will to risk."

That's right. And that unwillingness to risk is directly related to the scoring system. That's why Karjakin was satisfied with a draw with black in better position. Under traditional scoring, draws still keep you in contention.

"it might induce unplausible risk taking which could damage the quality of the games"

I've heard these arguments, I find them unconvincing. People worried about the quality of the games with faster time controls as well, for the most part those fears have been unfounded.

The greatest damage to quality of games is GM draws and non-games. Which is due to computer analysis, combined with a scoring system that rewards draws and discourages risk taking

"A player might want to avoid playing a drawish endgame for hours, resign it quickly to have time to prepare for the next game."

And throw away a point? Let's be realistic. I don't see that happening.

I've also heard people claim players could collude to 'split the points' in a Bilbao scoring double round robin by mutually throwing games. Also highly unlikely, and there's no evidence this has ever happened. Meanwhile GM draws are real collusions that happen almost every tournament. People are worried about a highly unlikely hypothetical method of collusion, meanwhile they accept the collusion that happens regularly.

"but we do not really have a factual proof that it is indeed raising the fighting spirit of players."

Yes we do. The games themselves. I gave this example on another page, the 2010 London Chess Classic. 13 of the 28 games were decisive and none of the 28 games were GM draws.

Also five of the eight players had a chance to win the tournament or tie for first going into the last round - which led to every player playing real, fighting games in the last round (this almost never happens under traditional scoring. Usually the leaders agree to short draws in the last round to secure their places at the top)

http://www.chessgames.com/perl/chess.pl?tid=72655

2009 London Chess Classic- 11 of the 28 games were decisive: http://www.chessgames.com/perl/chess.pl?page=2&tid=70344

2011 London Chess Classic - 17 of the 36 games were decisive: http://www.chessgames.com/perl/chess.pl?tid=76213

@koko48 even if @Petrarlsen would have complained about Karjakin accepting a draw - which he did not, by the way - that would mean nothing if we look at accepting draws in general. A draw is a normal result of the game, the problem is with unplayed games which comes from the players' lack of will to risk. If one introduces the Bilbao scoring system, it could raise the number of decisive games, but we do not really have a factual proof that it is indeed raising the fighting spirit of players. However, if it increases the ratio of decisive games, it might induce unplausible risk taking which could damage the quality of the games. Yes, under the Bilbao system you score 3 points for a win and a loss and 2 points for two draws. However, if you take some risks, you have a higher probability to lose a game and you will need to take additional risks in the next game to remedy the loss in points, which could cost you another loss, which might convince you to play solidly instead. Players might have the incentive to play solidly, waiting the opponents to risk. Anyway, besides the fact that I do not see factually proven that the Bilbao system would raise the decisive games' ratio and even if it does, it could lower the preciseness of games, on the other hand a well-fought draw will be punished. A player might want to avoid playing a drawish endgame for hours, resign it quickly to have time to prepare for the next game. So, if the Bilbao system raises the ratio of decisive games, those decisive games could in a certain amount mean unplayed games, where the weaker side resigns to be able to prepare for the next game and win it, since after many hours playing an exhaustive endgame, possibly, but not necessarily reaching a draw could easily mean losing the next game due to tiredness, after all.

re draws
The rules of the game are sufficient for a high winning percentage as Fischer and Kasparov have shown.
The key is motivation in reducing the percentage of draws.
Hopefully popularity and sponsorship would benefit -there must be a balance in appealing to the elite and the populace.
Tournament and match results should be decided as follows:
1] win points 2] tiebreak rules 3] coin flip - rapid games cannot be allowed to decide the best classical player.
Points and rating should be awarded by a new system:
1] win with black 1.2 2] win with white 1 3]draw .3 4] loss 0
This would also eliminate the need for Sofia rules and increase the percentage of wins - and losses with black..
Also to address knockout matches - there should be a minimum of 6 games per knockout, with the option of a qualifying tournament to accommodate the necessary reduction in the number of players in the main tournament.

@ koko48 : One more time, my post about Karjakin has no possible link with the 3 - 1 scoring system.

To elaborate a little more : if you have, for example (in terms of engine evaluations), a "0.5" position with White, or a "-0.5" position with Black, there is absolutely no possible objective reason to accept more easily a draw with Black than with White.

This was exactly what Albert Silver and I meant. And, nonetheless, as many players consider that a draw is a correct result with Black, for a completely similar position, they will content themselves with a draw more easily with Black than with White. This isn't logical, but hasn't any link whatsoever with the "3 - 1 scoring system" theme !

@Petrarlsen Karjakin's reasoning was not erroneous under the 0-1/2-1 scoring system, because a draw keep you in contention under that format. Therefore a draw with black is a satisfactory result, even in a better position.

But you know as well as I that Karjakin would not have accepted the draw under the 3-1-0 scoring system, and none of your intellectual contortions will change that obvious fact

@ koko48 : I wasn't at all "complaining about Karjakin accepting a draw in a better position against Aronian" ; I only fully approved the author of the article on the fact that it is an erroneous reasoning to accept more easily draws with Black than with White for similar positions.