King's Gambit Refuted

Once again this recommendation is from *last year*. It is an old move (again your line is even older than the Nc6 line) but the theory stands up to computer scrutiny thus far. By all means keep playing the old mainline; it's certainly not bad, and keep trusting engines and ignoring theory if you'd like. As hardware continues to accellerate, someday, probably in your lifetime engines will be strong enough to find the best moves in the opening.

Here is a blitz game I played in the variation last night (if anyone cares.) My moves were not the best.

Trusting the computer over the book in the opening is a huge mistake. There are just too many possiblities and long-term plans, etc early on that computers have a very hard time accurately assessing openings. Decades of experience, trial and error, on the hand, is more reliable for the one stage of the game every game goes through.

thanks, interresting

That's interesting, since computers have a knack for refuting ideas decades old. What takes GM's years to understand takes my coomputer a few nanoseconds to grasp.

The notion that humans understand chess on another level is interesting and very romantic, but it's not realistic.

Yereslov, are you kidding me?  Computers can't understand strategy the same way that humans do - they have a core of cold, hard calculation.  

Computers can calculate better than humans, but there is a horizon problem. The deeper a line, the more power it takes to calculate by something like an exponential factor. This is why engines play the opening very badly on their own, so badly they are never programed to play on their own but have opening books written for them by GMs. In order to understand the opening you have to understand goals that may not be apparent for many moves, many more moves than an engine ever thinks when it calculates. All this is fairly common knowlege. (A sort of line like Nc6 in the Salvio Gambit is a particularly good example. It's not apparent that Black has enough for the sacrificed rook for many moves so the computer doesn't even consider it a candidate move unless you make it look deeply at it, at which point (as I said) Houdini eventually likes the move very much.)

Tmk, you make an interesting comment that everyone knows - computers rely on game databases and openings books to help them get a kickstart.  

Ones rating usually as a rule of thumb indicates exactly how much myopia one sufferings from in relation to ones understanding of chess theory. This is shown in the fact that you would slate an opening has a whole based on one all be it unsound line. if you understood what you are on about you came to the appreciation that Bc4 against g5 is largely insufficent only if Black does NOT play g5 to g4 playing either h6 to fortify the pawn chain and or d6 to take away e5 to intend h6 and or Bf8 to g7 all depending on the move order used by White. If Black does play g5 to g4 then 0-0 (Muzio) or Nc3 (McDonnell) is a gamble in the face of a stronger an opponent (although some may argue the attacking merritts of the Muzio, I am doubtful). This is perhaps why in the face g5 for Black players of the White peices prefer h4 as the best try...As i have tried to tell others in various threads, the king's gambit is only a generic place name for a well over a dozen plus variations (gambits) and the King's Gabit will only ever be bust, toasted, refuted and any other name you can think of when each and every one of these dozen place varations can be shown to be a loss by force.

I would hate if someone discounted my input because the low rating I have on this site; I find myself agreeing with with Yereshov on that point. Ideas can stand on their own merit. (I agree with Hadron too, the real test of Bc4 after Nf3 is the Hanstien variation without g5-g4, not this. Plus there are a million other branches to deal with you're going to try to refute the whole opening. It's just that this thread was started on a variation I already knew the real refutation to.)

Of course people will discount input from various people because of their rating. You would be stupid not to IMO. I for one virtually never have anything worthwhile to contribute to specific lines or ideas because I am not nearly good enough to understand any position correctly. When I see weaker players than myself trying to do so I don't even always read what they have to say let alone give it serious thought, in fact I invariably find it amusing that they think that they have something meaningful to add.

Not only do I disagree with that point but I am confused by another that you make. You realise that there are potentially people who disregard your analysis because of your rating whilst at the same time imply that you are in fact stronger than it suggests "...low rating I have on this site;". There is nothing in your stats that I can see that would suggest that you couldn't be even a GM. You want people to take you seriously and you think that because of your rating they may not (most people won't check your profile) yet you do nothing about it? It wouldn't take much effort.

Technically an idea can stand up on its own merit as you say but lets be realistic. Any accurate complex idea suggested by a weak player is going to be the result of computer lines or passed down advice 95% of the time. In either case the the player suggesting such moves is not going to be able to understand or explain any of the ideas or moves that he/she presents making them useless. This thread is the perfect example of just that - some moron posting computer lines and attempting to pass it off as analysis. Futile in an attempt to do anything but amuse the right audience.

Fair enough. You picked up It was implicit in my comment that I did want people to know my rating here isn't a reflection of my playing strength. I still think it's possible for even poorer players than us to have good ideas, though, particularly in this age when things can be "checked" with computers if you learn how to do it right. I'm also suspect when higher rated members post glibly without explaining themselves and expect people to believe them based on their rating. Regardless by my line and actually Yereslov's line as well were both "passed down" (for more than 100 years) and looked at with computers. I do understand mine and play it. From the game posted at the start it seems like Yereslov understands his too. We're just mainly disagreeing about how useful engine suggestions are in the opening.

You forgot to add "with very rare success."

"Any accurate complex idea suggested by a weak player is going to be the result of computer lines or passed down advice 95% of the time."

Almost every ounce of chess knowledge is passed down. No one has ever become a GM just from playing chess. Experience doesn't make good chess players.

A troll.  Tracking turned off.

 Almost every ounce of chess knowledge is passed down. No one has ever become a GM just from playing chess. Experience doesn't make good chess players. 

.. standing on the shoulders of giants .. (Stephen Hawkin)

It's easy to claim someone is a troll, but it's much more diffuclt to actually prove it.

Who's the troll really: the guy who started a thread on a way to defend against a pawn sacrifice, or the guy who is trying to start an 80 page flame war?

I will go with the latter. 

If you have nothing useful to contribute, move along.

Nh6 keeps the position far more stable. 

There is no reason to complicate it.

If you want a "safe" option there are quite a few more before this positions. There are systems with an early d5 against either KG setup which give Black enough for a good game and are popular at the GM level. There is also not playing g5-g4 and holding onto a safe advantage against this kind of play by White. That kind of play is the real reason Bc4 after Nf3 is not played more. Also you can play the old line with Nh3 and f3. The old line certainly not bad for Black, but it's not winning either; Nc6 is.

The analysis of 5... Nh3 ends with positions where possibly Black has some pull; it's up for debate. The analysis of 5... Nc6 ends with an unstoppable attack proving Whites entire idea unsound.

In general an artificial move like Nh3 is not something you should consider playing unless it's absolutely necessary. In the position in question actually both 5... Nf6 and Nc6 will do. Nc6 is the best move. There is no need to fear the attack on f7 and go into contortions with the odd defensive Knight move.

If you feel the positions after Nc6 are too complicated by all means play the old line with Nh6 and f3. If you're looking for a refutation of the variation, as you seem to be, the most accurate way to punish White is Nc6. Did you look at my game? or the analysis in the articles I posted?

The KG is not refuted and anyone who thinks that for lack of a better word is an imbecile. It is not played at top levels because it is not as safe as other openings. We saw today in the FIDE World Cup a KG in a must win situation. No one is going to believe a refutation from some scrub, sorry it just will not happen. Showing one line of the KGA where white is not good is redundant. If it is not good, white will not play it. None of the moves made in your petty refutation were forced. You are acting like if one variation of an opening is bad, the entire opening is bad... and that is ignorant. And back in the thread someone posted a Muzio gambit which was not even close to being a Muzio gambit. I love the people the that say the Muzio is refuted... if it is refuted, post the refutation.