trandism wrote on 10/27/11 at 14:07:25:


Thanks for the link! I wonder how long it'll take for somebody to reach 10 Mg (Megagames) ...

Edited:

fling wrote on 10/27/11 at 20:43:56:

Agree.. 8...Nbd7 exclusively against the higher rated players.  If the second player wants to win 8...0-0 is a must because 9.Qb3 can be easily met by 9...Qe7 while 9.Qb3 after 8...Nbd7 threatens the b7-pawn and black has no better than a draw variation. So to speak black has to go along with 8...0-0 and Bf5-g6 variation if the win is a must.

Gilchrist is a legend wrote on 10/22/11 at 22:13:53:

Not quite!.. Why 8...0-0 when you can play 8...Nbd7 first. Then 9.Qe2 (9.Nh4 Bg6 10.Qb3 a5) 9...Bg6 10.e4 and now 10...0-0 (thus transposing to 9...Nbd7-line and slightly improved move order, in fact). Now 11.Bd3 Bh5 12.e5 Nd5 13.Nxd5 (13.Ne4 can be met by Timman's 13...c5!? or the traditional but somehow passive 13...Be7 14.Ng3 Bg6 15.Bxg6 hxg6 - and I don't like 15...fxg6 because of the weakness of e6-pawn) 13...cxd5 14.Qe3 Re8!? (intending Nd7-f8 if needed) 15.Ng5 (15.Bd2 Qa5 13.Bxb4 Qxb4) (or 15.Ne1 Rc8 16.f4 Bxe1 17.Rxe1 Bg6 =) 15...Bg6 16.f4 Bxd3 17.Qxd3 f5 can be an usual continuation for both sides. Imho, black is ok here.. 

If you really want to find regional games - especially under 2000 - then you have to forget about megabase and co. Either for statistic or other reasons you have to start from here --> http://www.chessgameslinks.lars-balzer.info/

Great idea! I could supply a review of the theory to help you make your tournament better and more entertaining as well as useful for making any conclusions.

I didn't have in mind any specific statistics, but i spoke from my own experience as a coach. For 10 years now i see juniors trying my proposal and play the piece sac (and 99% of those games are not in the megabase) and the results are more than encouraging. As White i propose them to play the e3 line and avoid that stuff.

In general i think that doing statistic researches that concern players < 2000 are dangerous because many of those games are not in the databases. For example in the big open tournaments only the games of the first boards enter those databases. Doing that research would require consulting databases like greekbase for example that is full of games that do not enter the big databases. This database concerns Greek club players, so i assume that similar databases exist in other countries also. It would be interesting to see how those statistic will change if you use those tools. But again, do the amateurs know a bot of theory? In my experience if Black is well prepared (not big things, just know that you should castle K-side and know when to play for the ...b5 attack when White commits a mistake) at club level he wins again and again...

Zwischenzugzwang wrote on 10/22/11 at 21:56:18:


I think this statement is what I really thought of when I read Ametanoitos post. In fact, I haven't seen it played ever in games between non-titled players. I think both sides might be too afraid of playing into a line full of theory like this endgame.

Ametanoitos wrote on 10/22/11 at 14:11:53:


Interesting statement. I've just made some little statistics for the following position:



Data Set 1: Elo between 1000 and X

X = 2000:    4 / 50 %
X = 2100:  10 / 45 %
X = 2200:  22 / 48 %
X = 2300:  55 / 55 %
X = 2400: 104 / 59 %


Data Set 2: Elo between 0 and X

X = 2000:  30 / 53 %
X = 2100:  39 / 50 %
X = 2200:  54 / 49 %
X = 2300:  99 / 54 %
X = 2400: 148 / 56 %

Both data sets are based on my database of appr. 5 mio. games. I only considered those games where both players are rated between the lower (1000 resp. 0) and the upper margin ("X"). I would like to draw the border between "club player" and "(semi-)professinal" somewhere around 2200 Elo.

Data Set 1 seems to indicate that Ametanoitos is right insofar as Black has slightly better chances than White. Data Set 2 is more problematic, as the lack of an Elo rating does not necessarily imply low playing strength. To avoid the worst blunders, I excluded all games where the lack of an Elo rating has obviously nothing to do with playing strength, e.g. Reshevsky-Smyslov, Wch 1948.

Maybe the most interesting figure in DS 2 is in the first line: 30 games, White scoring 53 %. It's interesting insofar as here we have 30 games (compared to 4 in DS 1), showing a slight superiority for White. If those 30 games were really played by players "weaker" than 2000 Elo, then Ametanoitos statement is probably not true.

Anyway, these figures indicate that the Slav piece sacrifice variation is very rarely played at lower levels.

Best regards,

Zwischenzugzwang