online chess against computer

Backgammon Online

Divide and you will find!
Funny name, real interests
[ Sign up | Log in | Guest ] (beta)

Have you ever played Backgammon online? It is a lot of fun. So come play Backgammon online with us!

PLAY BACKGAMMON ONLINE
Sign up now to play backgammon online for FREE!
backgammon online

keiserpaul 66 ( +1 | -1 )
The best opening move I found an interesting side to download some large games databases : www.uib.no/people/pfvaf/chesslib/
The side include statistics report about these games. The first DB consists of 555671 games and there we see that White scores 53.2 % . But there also are figures about the scores on each first move. These figures are :
d4 53.1 %
c4 54.0 %
Nf3 54.1 %
f4 48.3 %
g3 53.9 %
e4 52 %
b4 52.3 %
b3 50.5 %
Nc3 52.5 %
e3 42.1 %
g4 52.3 %
d3 45.7 %
a3 47.1 %
c3 46.2 %
h3 40.5 %
f3 45.2 %
h4 47.4 %
a4 48.4 %
Nh3 37.1 %
Na3 50.0 %

Remarkable are the good results of some so-called unorthodox openingsmoves as b4 (52.3 %) and especially g4 (also 52.3 %). Even a4 scores 48.4 % . So statistically you've more chance to win an a4 with White than another game with black (46.8 %)
brunetti 19 ( +1 | -1 )
Warning You posted here the statistics relative to ECO codes A*, just one fifth of the whole DB, and not conclusive especially about 1.e4, 1.d4, 1.c4, 1.Nf3.

Overall White winning chances, according to these DB's, are about 54%.

Alex
More: Chess
taoistlunatic 58 ( +1 | -1 )
I have seen very similar tables It seems that 1.c4 and 1.nf3 have the best winning percentage, but they also rank at the top of the list for drawing percentages, so to conclude, these openings are very strong and statistics show that it is very difficult for black to create a winning position from these openings, but because black can equalize easily and hold out for a draw if he/she wishes, These are not the best openings when you are playing against an opponent who wants a draw. I prefer 1.d4, which has the third hightest winning percentage, but does not produce as many equalizing, drawish positions as the others.

taoistlunatic 18 ( +1 | -1 )
by the way... If you like 'do-or-die' type of openings, Brunetti's favorite King's gambit (1.e4 e5 2.f4) is statistically one of the LEAST drawish openings, as this opening tends to push the limits of risk and reward!
acne 5 ( +1 | -1 )
i can't blame my loss games on e4 even it ranked 8th =D
keiserpaul 20 ( +1 | -1 )
Updated figures In view the "warning" from Brunetti, I have extented the exercise taken into account 2442134 games (all 5 DB's) .
White's score then is 54.4 %
The score of the first move (sorted descending) :
Nf3 55.4 %
d4 55.2 %
c4 54.7 %
g3 54.0 %
e4 53.9 %
Nc3 53.5 %
g4 52.3 %
b4 52.3 %
b3 50.5 %
Na3 50.0 %
a4 48.4 %
f4 48.2 %
h4 47.4 %
c3 47.4 %
a3 46.9 %
d3 45.6 %
f3 45.2 %
e3 42.6 %
h3 40.9 %
Nh3 37.6 %
peppe_l 10 ( +1 | -1 )
Stats are interesting sometimes But dont always tell the truth. Based on this both 1.h4 and 1.a4 are better than 1.e3 for example :)
keiserpaul 39 ( +1 | -1 )
Peppe_l e3 is a modest move while a4 and h4 are attacking moves. e3 can lead to a variety of different openings : a Bird's opening (having avoided From's gambit), an unusual English via 1. .. e5 2.c4 , an inferior line of the Queens Knight, or a reversed form of the French, the Queen's gambit, the Queens' Indian, or the Sicilian. All these are not preferable above 1.a4 or 1.h4. That's the thruth that you can learn from these statistics.
tulkos 10 ( +1 | -1 )
what does ar and h4 attack?! and e3 opens up lines for both the queen and the bishop,while ar and h4 don't.are you sure about these statistics???
brunetti 24 ( +1 | -1 )
The first move alone has no real value. Statistics about first move are quite meaningless. There's a high transpositional factor to be considered. For example, 1.Nf3 c5 2.e4 is a sicilian defense; 1.e4 d6 2.d4 g6 3.c4 is a King's Indian, and so on. A relevant statistics should be based on ECO codes.

Alex
keiserpaul 28 ( +1 | -1 )
Tulkos The original idea of a4 ( by Mr. Preston Ware of Boston, at the Fifth American Chess Congres at New York in 1880) is to rush the pawn to a6. Look also at Paul Valle's contributions to the Forum with the same idea for h4-h5-h6. So these pawns are attacking black's pawn structure.
keiserpaul 24 ( +1 | -1 )
Brunetti " The first move has no real value ". Splendid, that exactly is what I wanted to make clear with the statistics. 1.a4 or 1.g4 or 1.h4 are not bad moves by themselves. The position you get on the board after some further moves is what really matters.
taoistlunatic 60 ( +1 | -1 )
Alex makes a good point. 1.D4 seems to be the best opening move when all the transpositions are considered, for example the queen's gambit is white's overall best opening statistically. The nimzo-indian is black's strongest defense against 1.d4, but it is not nearly as strong as the sicilian kan/taimanov variation is against 1.e4

Statistically, (in grandmaster play) 1.e4 c5 2.Nf3 e6 has an extremely high winning percentage for black, I can't remember what it was, but is convinced me even more to prefer 1.d4 and I am now playing this variation of the sicilian against 1.e4

You can say its all a bunch of statistical garbage, but I am interested in winning, and I wwant to play the openings which give me the greatest chance of winning, :D -Carl
taoistlunatic 33 ( +1 | -1 )
by the way, Alex beat me in our first and only game with a strange defense 1.d4 ...Na3?!

I opened the game very strong and had a considerable space and development advantage, I knew I was doing well as Alex began taking a long time on his moves, but when the middlegame got started, I played poorly and Alex's skill prevailed...:(
zdrak 146 ( +1 | -1 )
It's important to realize WHY some openings have the better winning percentage for one of the sides, rather than trust blind statistics.

Sometimes it's just a matter of fashion - a strong player or two take up an opening and "improve" it's statistics.

Example: in the 70's, the statistics of the variation 1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 b5!? were clearly in favor of black. But if you'd examine the games more carefully, you'd see that most of them were played by the inventor of the variation, Polugaevsky (a strong GM), who was trying out various ideas in this line against weaker opposition. Does this mean this line is good for black ? Not necessarily.

Same with Bent Larsen, who won a lot with 1.b3, giving it a good percentage (he won because he was a better player than his opponents, not because 1.b3 is such a great move!!)

Sometimes the opening would go into a slump - a strong idea is found for white, and the theorists of the "black" side have a difficulty to keep up for a while. For example, currently the black side of the KID is in a slump, as the Bayonet attack (9.b4) is a strong idea for white, to which black didn't find strong countermeasures (yet!!). But on a club level, the small plus that white gains after 9.b4 is negligible and the opening can be played with no problems.

To summorize: it's much more important to find an opening that suites your style of play (open/closed positions, attack/defense, etc) than follow the percentages.

My 0.02$ (or, in the spirit of the discussion, my 2% or a dollar ;-)
peppe_l 377 ( +1 | -1 )
Hmm "e3 is a modest move while a4 and h4 are attacking moves. e3 can lead to a variety of different openings : a Bird's opening (having avoided From's gambit), an unusual English via 1. .. e5 2.c4 , an inferior line of the Queens Knight, or a reversed form of the French, the Queen's gambit, the Queens' Indian, or the Sicilian. All these are not preferable above 1.a4 or 1.h4. That's the thruth that you can learn from these statistics."

Ok, if you believe this kind of stats always tell the absolute truth, thats fine. My opinion is there are differences like the popularity of certain moves among certain level of players and so on...Anyway, I guess the real reason why top players never play 1.a4 or 1.h4 is because they havent analyzed these moves properly. Im sure when they do, openings like 1.a4 and 1.h4 will be as popular as 1.e4, 1.d4, 1.c4, especially since first moves doesnt have any importance whatsoever.

Now Im sorry if my sarcasm offends someone, Im sure all unorthodox openings mentioned here are playable on amateur level, and might lead to interesting positions. But...(ok I wont say more, for some reason critisizing unorthodox openings in any way leads to situations I prefer to avoid).

"It seems that 1.c4 and 1.nf3 have the best winning percentage, but they also rank at the top of the list for drawing percentages, so to conclude, these openings are very strong and statistics show that it is very difficult for black to create a winning position from these openings, but because black can equalize easily and hold out for a draw if he/she wishes, These are not the best openings when you are playing against an opponent who wants a draw. I prefer 1.d4, which has the third hightest winning percentage, but does not produce as many equalizing, drawish positions as the others."

Also, I bet 1.Nf3 and 1.c4 are favoured by positional players, who are more prone to draw games not only because of their openings, but also because of their playing style. So it is possible the difference in drawing percentage isnt as large as statistics tend to show.

"1.D4 seems to be the best opening move when all the transpositions are considered, for example the queen's gambit is white's overall best opening statistically. The nimzo-indian is black's strongest defense against 1.d4, but it is not nearly as strong as the sicilian kan/taimanov variation is against 1.e4

Statistically, (in grandmaster play) 1.e4 c5 2.Nf3 e6 has an extremely high winning percentage for black, I can't remember what it was, but is convinced me even more to prefer 1.d4 and I am now playing this variation of the sicilian against 1.e4

You can say its all a bunch of statistical garbage, but I am interested in winning, and I wwant to play the openings which give me the greatest chance of winning, :D -Carl"

How much this slight difference in games between GMs affect winning percentages on amateur level? For example many Sicilian lines are theoretically good, but in real-life play they work well only in the hands of strong players who know them exceptionally well. But, if these stats have any affect on your opening choices, then you are lot stronger player than I am and you have no reason to waste your time for reading my opinions...On my level knowing the basic ideas of the opening (and ending to positions one likes to play!) is most important, and the statistical difference between Nimzo and Sicilian is completely irrelevant. Even theoretically dubious moves can give good results if one knows them well, as for example Keiserpaul and Paulvalle from here have proven in real-life situations.

PS. Is the level of Sicilian Kan/Taimanov players and Nimzo-India players exactly the same? How about opponents? If the average rating of Sicilian Kan/Taimanov and Nimzo players and their opponents isnt exactly the same, statistics do not give reliable results about the value of these openings.

taoistlunatic 69 ( +1 | -1 )
all good counter-points But I really don't have any preferable style, I can play in any situation, and I neither prefer open or closed positions, I just want to win.

Further, if you read my forum about 'opening repertoire' you can see how hard it has been for me to chose an opening I like, because I like most of them, but I had trouble deciding which ones to play and develop skill at. So the statistics were a great way for me to make a decision about my opening repertoire. Are they relevant at amateur level? Probably not! But because my decisions were based on the stats of GMs, at least I feel confident in my choices, and if I ever do get that far, I will have a formidable opening repertoire which I have studied and practiced for years....
brunetti 67 ( +1 | -1 )
It's not possible that you feel the same in any situation, any opening, any middlegame type, any ending. You're not Alekhine (are you? :), and you must play better in some conditions than in others :)

So, instead of GM's opening statistics, which are good only if we're choosing the global repertoire of a GM's team, one should classify all his games by opening and by middlegame type; then analyze his performances for the different classes he made, and then he will have a meaningful base to choose with a clever repertoire.

If I don't play like GM's, I don't need to know that they score 55% with 1.c4, while maybe I am not good in strategical english positions, and I feel at home with, say, Grob opening.

Alex
taoistlunatic 25 ( +1 | -1 )
Why is it not possible? I like many openings and can play well in many openings, if that makes me like Alekhine in some way, well that's great!

Your way sounds good also Alex, but mine will work just as effectively...