What well known gambits are unsound?

I'm pretty sure 5 g3 is far better than 5. Nbd2, which seems totally negative.

I'm not totally convinced by e3, Pacifique, but 7.ed is wrong. Probably 7.Bg2. Then 7 ...d3 is a possibility but I don't think it poses white too many problems? Still, I don't have a chess board to hand.

Is not the above  a plausible continuation and does not White have considerable pressure on the black position?

What does unsound mean? Is it the same as refuted?

Looking from white's perspective, my view is

refuted < unsound < worse < equal < average < better ... etc ... but they are all just words. Some gambiteers take "refuted" or "unsound" as a personal attack on their beloved gambit.

From black's perspective, things are a little different, because of white's opening advantage. So if (let's say) it turns out that the Traxler with Nxf7 is completely equal, then this is a black success, theoretically speaking.

When it comes to gambits, you can really evaluate them along many dimensions.

1. Are they easy to play OTB?

2. Are they popular, and how likely is your opponent to have studied it?

3. How easy is it to learn?

4. How likely is it that I can steer the game in this direction?

5. What is the time control?

6. Is there a forcing draw line?

7. How difficult are the main ideas in the opening to discover OTB?

And lastly... what is the theoretical evaluation? This question is almost an afterthought for many gambiteers, and rightly so. Unless you are a pretty well-known player who gets a reputation as someone who exclusively plays the XYZ gambit (and therefore easy to prepare for, especially if some smart person decides to publish a blog "How to beat player ABC" or something).

Of the numbered list above, I think that the most important thing is #4. This is about utility of time spent more than anything, but if you're going to learn some very specific tactics and ideas about a very specific opening or move order, doesn't it help a lot if you can actually use them? Obviously the gambits occurring in the early moves (Englund, Albin, Budapest, Elephant) rank well here... but they also tend to be the ones which are more likely to be previously studied by your opponent. In a nutshell, everybody has to do their own calculus as to how to spend their time preparing these sorts of things.

Let's look at two [successful] gambits, from two high-level games. Both were featured in Chess Life, and one features our own David Pruess.

Mackenzie - Pruess: http://www.chessgames.com/perl/chessgame?gid=1648780

Dana, by his own admission, had prepared this gambit extensively and was about as prepared as one could be. Now it's worth noting that almost no matter what the opening, this is a difficult position for an opponent, even an IM like David, to find himself in, but especially so for an imbalanced position. How would this gambit fare based on the factors above?

1. It's difficult for black to play, with no easy plan. Yes, he's up a queen for two pieces, but there's no clear way to get an advantage.

2. Nearly impossible that the opponent has studied this one.

3. Difficult to learn

4. Medium to low likelyhood. Obviously the Sicilian is very popular, and the 2...d5 has been considered black's best at various points in time, but the real key is whether black will play ...Bg4 going for the queen sac. Add it all up and it might be a bit of wasted effort. Also, I bet that the publishing of the game in Chess Life really kills Dana's future chances of an encore performance, which is really unfortunate - and it might be against a player who also counter-prepares.

5. This would work pretty well in long or short time controls.

6. Only on the white side.. :-)

7. Pretty hard. You can watch Dana's blogs, articles and videos on this, and he makes it sound simple, but OTB it is not that easy.

Verdict: This is a great find by Dana, and he was able to really make his preparation count.

Esserman - Van Wely: http://www.chessgames.com/perl/chessgame?gid=1630005

This one was really interesting, because the article in Chess Life was written by Alejandro Ramirez, one of my favorite CL authors. He's funny, direct, and honest, and in this case he wrote about how Marc chided him for declining the gambit in one of their games - but Ramirez said that he really thinks that it's a mistake to accept the gambit, which I found interesting.

The worst part about the Smith-Morra is that it is SO popular, even with people of very low ratings, just because white's game basically plays itself, and because it is so easy to enter. But Marc was able to produce quite a convincing game, which almost makes you really question some of Van Wely's ideas in the game - but I also read that the setup with ...f6 was held in theoretical high regard at the time. I don't know if that verdict has changed.

--------------------

On a personal note, I'll write about the Englund Gambit. I entered a tournament here on chess.com back in 2008, so I could learn more about how this opening works. I'll give some of my findings about it here.

http://www.chess.com/tournament/englund-gambit---open

First, it is pretty bad. It became clear that having white is such a large advantage (comparable to having a killer serve in tennis) that in the double round robin you're always looking to hold serve as white (win) and then try to scrap a draw as black. Obviously a win would be incredible, but I'll go on as to why that's sort of unreasonable.

Going in to the tournament, I had found some page (can't find it now) which had recommended 1.d4 e5 2.dxe5 Nc6 3.Nf3 Nge7, and gave just a couple of simple variations to back it up. I said heck let's go with it, and off we went.

So I might be totally off about this gambit, but if you discount the ridiculous trap where black mates on c1, what's the upside? A good gambit should have some good upside, and this one just didn't. Again, maybe the upside doesn't mesh well with my chess style, but it just didn't seem to be there. You can see the difficulty of trying to get a positive result as black (draw or win) just by looking through the rounds. Interestingly, round 2 and round 3 had exactly the same people and pairings, because nobody got eliminated in round 2.

Even if black can get his pawn back, it's not even clear he has equality. He's just not down a pawn. Or black can gambit a second pawn with f6 or d6, but note that white doesn't HAVE to accept that second pawn. And if he doesn't, black just get an equal game. If he does (and he should), at least black gets some freedom for his pieces. I mean, a pawn isn't that much in the grand scheme of things, in the lower rated area, and especially in the non-engine area. People make mistakes.

And so of course the tournament ended, I even blogged about my win as black in the final round. It was a miracle that we didn't have several more rematches in the final round, because my opponent was a pretty good player.

-- Oswald

p.s. What is the theoretical verdict in the Albin Spassky variation (4.e4)

[

I try and be objective]  re 6. Bg5 

However I still would like an opinion on this line:

It should be realized that if the early move for White of a3 works well that all of this analysis is beside the point?

My own opinion is that I wouldn't castle or play Q b3 for that matter but I think I'd start advancing my queenside pawns very early. I'm not convinced by that position: I think Black may be better.

My experience in such positions with an early a3 is limited but basically white gets a3 in and then has a bit of defending to do. Then typically white gets time for about three attacking moves. Generally black's only recourse is a counter-attack against white's king. If white survives it, and he should do so, then he wins.

Oh and nobody seems to have mentioned the Cochrane Gambit in the Petrov Defence. I think it's sound (only just) and I've played it otb several times and won with it.

When I used to play e4 I specialised in the Vienna Opening. Not the Vienna Gambit but the Vienna Opening, which is far wilder. It's generally black that gets to sacrifice though.

Another amazing gambit for black is Tal's Gambit in the Veresov. The Veresov-Richter attack is out of fashion, unfortunately, since lots of people thought that Tal's Gambit (in the variation with 4 f3) is unsound. Again, it isn't unsound at all. I used to love playing it as black.

I used to play a completely unsound piece gambit in the Modern Benoni because of its surprise value and the practical results I achieved with it against players up to about 1950 FIDE. (Is that about 2050 USCF?)

I wish people would give diagrams and actual lines rather than speaking in generalities.

It is quite ok to use unsound gambits if your aim is to win and you think that will help you win. There can be a later problem with that but for the immediate goal of winning it can be a good idea.

Fred, how would you continue with Black in that particular position?

I would like to see the Cochrane Gambit...

1)latvian

2)smith morra

3)halloween

4)If played poorly scandinavian

What is this piece sac gambit in the modern benoni? Never heard of it

People are starting to talk in generalities not specifics. I have a line for which the only comment was something like "I'm not convinced by that postion, I think Black may be better" I do not think they have examined the position

very well--so will try again with some extra moves...

I`m still awaiting Ponz (and others) opinion on 1. d4 d5 2.c4 e5 3.dxe5 d4 4.Nf3 Nc6 5.g3 Nge7 6.Bg2 Ng6 7.Bf4 h6 (with idea to take on f4 and play g5)

After 4.e4 Nc6 5.f4 f6 5.exf6 Nxf6  Black has a dangerous inititative for a pawn. Here are two illustrative games.

Pacifique, I don't mean to be unnecessarily picky but why do you choose an inferior line for white on move 8? Evidently 8 B X e6 is correct. I know that white may lose his fianchettoed bishop but it seems that he gets quite good development. I may be wrong and this idea of e6 in the first place was off the top of my head but it needs to be tested properly and fairly. However, thanks for pointing out that 8 Q x d8 apparently loses.