Statistically correct moves and openings

I wanted to avoid white playing Be3 followed by several moves herding my Queen into a square of his choosing and slowing my long term development or possibly doubling my pawns on c4 making it easy for him to pick the extra pawn back up.

After 12. Be3 I'm not too happy with my options.  12...Qc4 doubles, if 11... Bb4 then 12 ... Qb4 is not available.  Only 12... Qh4 remains and that seemed unclear to me, like a trapped queen was in the air or at least a misplaced one.

These statistical evaluations are only important for people who play perfectly (nobody). I would like to see a statistic that shows how badly blunders effect the game in a given opening. This stat would be a much better indication of soundness for anyone <2500

ArcBlade - I think 12...Qh4 would be fine.  He would need quite a few more pieces to trap your queen.  In general, it's more important to develop the pieces sitting on the home row than to move a piece that is already sitting right smack dab in the centre of the board already.  The centre is an ideal place for most pieces.

Can someone do Grunfeld? ???

I play and have most success with the modern defense as black, but I am mostly a positional and defensive player, and like to avoid crazy tactical positions if I can. I'd also to comment on SylentSwords evaluation of 1.Nf3 ..... the point of 1.Nf3 and why it is good is that it does transpose to many solid positions for white while maintaining initial flexibility....but I don't play it as often as I see many players higher rated than me play it. Also I'm surprised at the number of people that still play the Scandinavian on this site.....I usually win relatively easy as White when they play that, but not always....

well that was wrong....it doesn't take my corrections for some reason, this is it:

Ok, I found out how to calculate. Not busy this summer. Do you want something to be calculated ? ( I only calculate lines with 100+ games in chessgames.com database )

yeah, just in that position, white has various continuations, but there may be 100 lines in that variation because I've seen a few masters that play it often as black.....

Thanks wayne_thomas.  I know pieces are strongest in the center it was just that in this situation I was concerned about him harassing my Queen for several moves getting development with tempo and eventually maybe trapping her, it just looked scary to me at the time.  I figured steal the pawn and get away, after all a pawn is a pawn.  But thank you for the analysis.  I will examine it.

I get that some openings are considered statistically better at the highest levels.  However, what does this tell us at lower levels?  Suppose GMs have better results against the Gruenfeld than the Sicilian (as they do), does that imply that those results will be the same at club/tournament level say U1800 or U1600?  It wouldn't seem to, do we have any data on what the results are for these lines when played by "suboptimal" players like myself?

Additionally, I agree with GM Simon Williams when he advises lower rated players to choose three openings that fit their personal style and learn those.  That is to say, if the Sicilian is statistically best but it is more aggressive than the player is comfortable with and includes lines that say suit the player with white more than the one handling black might he want to steer more into something he is more comfortable with, supposing he is a positional player?

Also, I wonder if directing chess toward one or two "correct" openings is right at all or if it will make for boring, rote, repetitive chess with every game following the same memorized pattern until move 25 and then following the same set of memorized endgame patterns for the last 20 moves with the only uniqueness being in the 10- 15 moves between the two.  Will we end up starting the game from move 25 of the Sicilian Dragon at some point?  

That brings up a different point, these stats seem based on the overall response but what about variations?

Yesn the pourcenrages are more accurate and there is also clear 2200+ vs. 2200+ restriction.

https://chesstempo.com/game-database.html

I'm now working on the summary of our calculations. I'll post it in 2-3 hours in a new thread.

+0.39 main line - +0.391 suboptimal ? you made me laugh. ( It's playable !!)

Chess is more complicated and varied than you suggest. If you learn 2 or 3 openings well--you will never or rarely be following a line memorized until move 25. In fact somewhere in the first 5 or 6 moves your opponents will respond with their own  moves which will be out of your preparation/memorization.

Besdies memorizing an opening 25 moves deep is not the way to learn that opening. You have to understand an opening--not memorize an opening.

I think we're missing each other a bit here.  I wasn't suggesting that learning the 3 openings, as GM Williams suggests, is what would lead to a sort of rote 25 move memorization.  And I absolutely agree that chess is much deeper and more beautiful than one line run 25 deep.  That was in fact my concern.  That is, whether or not this statistical labeling of lines like the French and the Gruenfeld as "suboptimal" and other lines as just wrong while suggesting that the only "right" idea is the Sicilian or maybe the Modern seems to be driving toward this idea of a statistically "correct" way to play?  I was lamenting the possibility that we are being told that to play say the French is wrong and you should only play the Sicilian.  Then might we be told that only the Dragon is statistically good?  At some point a very high proportion of games then becomes a dragon.  What then does that do to the game that, as you pointed out, is supposed to be wonderfully varied and beautiful and creative?  All we can then hope is that white finds some resource to refute the "best" black line.  And what about players who aren't comfortable in open tactical lines but prefer closed positional games (that is if only one line is "the best")?

As NM happytoad and I were hinting, straight win draw loss statistics are probably not the best way of finding good moves.  If you want to use an archive of GM games when building your repertoire, it's probably more helpful to at least factor in the ratings of the players, and try to figure our where the mistakes were made in the games.  If a whole bunch of GMs play a move that is later found to be unsound, the statistics are going to lead you to the wrong conclusion.

It's probably more important to understand the ideas behind your favourite openings, and what plans are associated with them than worry about stats in the first place.

I agree.  I mean as long as the statistical analysis is nothing more than an interesting curiosity I think that's cool.  But I think people should not use it as a suggestion of what moves to play.  As I mentioned before I think people need to choose openings that fit their style and I agree that once you do that you should go about understanding what the themes and plans of those openings are.

Great! I'll incorporate it later in my summary that I posted here:

https://www.chess.com/forum/view/chess-openings/4-statistical-subcategories-of-moves-and-openings

There are intermediate moves between Indian and Grünfeld. The classification of any move leading to Grünfeld depends also on the evaluation of the source position (just before Grünfeld). In particular, it's quite possible that a move m: X-->Y, leading to a position Y with good evaluation, is suboptimal (or even bad). In that case, it means that the source position X was even better than Y.

Btw we round up/down numbers to 2 decimal digits and +0.39 --> +0.39 is considered to be statistically good.