I didn't mean it's not "junk" just in terms of objectivity, it's also not junk in terms of fun and how it's an interesting opening. 

It is played a lot in correspondance according to MCO-14 "because of the fantastic complications that can arise"... perhaps not at the highest or most serious levels. but I still don't think White should win possibly unless you're a GM or strong master. 
  
I think Nunn's "ten-page refutation" idea/joke is silly because Kosten is showing the best lines and doing it in an excellent and honest way as Nunn states. You can write a book about a refuted opening no problem. The King's Gambit is worse than the Latvian Gambit and yet there are books about it. I don't think it's fair for Nunn to make that comment about "what GM Kosten probably really thinks". We see that Kosten gave the best lines and agreed that Black receives a little compensation but not much for the pawn which Nunn agreed was an accurate statement, so I don't see what more he wants.

Uhohspaghettio wrote on 07/01/11 at 17:12:52:


The Latvian is junk. John Nunn wrote a great little book, Secrets of Practical Chess (1998). In it, he discusses "books on offbeat openings" (pp. 70-76).  He specifically chose to analyse Kosten's book, The Latvian Gambit (1995) as an example of the problem with books on offbeat openings. 

Nunn stated that "for many players this opening offers a huge advantage: it provides a more or less complete defence to 1.e4. ... The question is, can the Latvian be considered playable?"  Nunn analysed several lines and came to the following conclusions:

By focusing on a simple line for white, he found that while Kosten suggested some positions were "fine" for Black, he found that Black "has violated virtually every principle of opening play, making weakening pawn moves, giving up a pawn and ...developing his Queen before other pieces.

Nunn continues, "Summing up, Black's position is probably just lost." Nunn was very harsh on Kosten, but he pointed out that Kosten himself was very honest about the Latvian. Nunn quotes Kosten as saying of the critical variation (6...c6): "Calmly leaving the e4-pawn to its fate;  what exactly does Black obtain in return? Objectively, very little: just a slight lead in development and a lot of fun."(Bold and italics added by SF) Nunn points out that that this is hardly "confidence-inspiring for Black, but of course having taken on the book he can hardly admit that the whole opening is just unsound. The publisher would doubtless not be amused by a manuscript consisting of a ten-page refutation!" 

Nunn points out repeatedly that Kosten is an excellent, honest author.  The problem isn't with Kosten, but the opening.

Trw has already dealt with your claims regarding the supposed merits of the Latvian in correspondence chess.

I think that studying games via opening books doesn't "severely limit" as he said my understanding of the middle game. It's the same as using a book with complete games except with a lot less material on the middle game. I do have books on middle game principles and strategies also. 

The Elephant and Englund gambits aren't "junk", they just give you a disadvantage in the opening. The Latvian Gambit is perfectly playable in serious competition and is quite popular in correspondance. Those openings have some meaning/ideas, they are not just ridiculous like h5. Following on from your logic, you also consider correspondance chess ridiculous because "junk" openings like the Latvian Gambit can be played there and win at a high level. 
 
I never said that you can judge openings from blitz, I was making another point. However you can take things from blitz even if you do drop a piece alter on. I don't know what you're referring to blitz for.    


Did you know that Tony Kosten, the administrator of chesspub, wrote a book on it the Latvian Gambit which you just called "junk"? http://www.jeremysilman.com/book_reviews_jw/jw_latvian_gambit_lives.html . 

Edited:

Uhohspaghettio wrote on 06/30/11 at 19:48:18:

Don't put words in my mouth, please. Where do I say I just study some random positions?
I do study openings, just not as a repertoire. I usually study a line when a certain game triggered my interest or sometimes a piece of analysis in here.
With repertoires I have seen too many players (myself included) fall into the same pitfalls:
-Overstudying lines before daring to play them
-Using a simplified repertoire and thus getting into problems when the middle game (or opening) is different. That is what often happens when people use a one-book repertoire as white (often advance French, advance CK, c3 sicilian etc).
-Using lines that just dont work for them, as a result of other lines in their repertoire or of economising the repertoire (see above) or of having a certain book.
-Basing a repertoire on a style which they dont have.

I could go on a little, but this usually applies to people who work on a repertoire and are below 2000 elo. I used to have a "one book" repertoire (Italian, Closed sicilian etc), but I noticed that at some point this wasnt working for me anymore. So I decided to change my repertoire (1.e4 main lines). That did help, but only a little. So I decided to just stop with it and play whatever fancies me at that moment. Since then results have only gone up. Whether that is due to not having a repertoire or because I play mostly main lines (Cox had convinced me ) is open to debate.
Quote:

I dont know, he did win Hoogovens didnt he?

Edited:

trw wrote on 06/30/11 at 14:32:27:

I dunno, it would seem that studying the middle game via this route severely limits the broader middle game understanding as it is more difficult to "swith". Personally I think you learn more by going into terra incognita earlier. You get more diverse positions and due to being forced to work it out behind the board I get more benefits from analysing the game. For me usually things only stick when I thought about it during a game 
Plus otherwise all those opening books go to waste...

And yes I do have some sort of repertoire
-I only play 1.d4 or 1.e4 main lines
-Vs 1.e4 I dont play the Pirc, Alekhine, Scandi and other weirder openings
-Vs 1.d4 I dont play the Benoni, Tarrasch, (semi)-slav and other weirder openings (modern and the likes)

How far we have come since the days when 3-minute chess actually wasn't rated.  I continue to wonder why such things get brought up so often in the context of opening repertoires, the merit of particular opening variations, etc.

Willempie wrote on 06/30/11 at 09:23:52:


I would argue the exact opposite. How does one get an understanding for the resulting middlegames if you don't have a repertoire? If you jump opening to opening to opening as a u2000 player you'll never understand the middlegames cause you'll never understand the opening it came from. Rather the better way is to slowly gain knowledge of the resulting middlegame positions/concepts by constantly seeing similar ones from the same opening.

also TN has written about repertoires in the general chess section and has answered a multitude of questions pertaining to repertoires.  so I would check past posts in the general chess section.  Many of the posts are answers to my questions in particular.  You have gotten some good ideas by these masters.   
Points to get in my opinion:
1. Play open games or at least semi open games as all closed games open up.  Save closed games for last.   
2. play main lines aka play 1...e5 1...c5 1...e6 or 1...c6 to 1.e4 and not 1...b6.   
3. Time if you have it play the super sharp main lines.  Since attack is probably first you want to learn.   
4.  I think TN made a good point to play no more than 2 openings vs 1.e4 1.d4 ect.  and play no more than 2 first moves!   

so a sane repertoire that will teach you chess and not openings (took me a few years to get this through) is 
1. vs 1.e4 play 1...e5!!!  For get sicilian or French till later.  Vs Ruy Lopez play either Breyer variation(now big with Magnus Carlson) or Open Ruy Lopez.  Maybe even Archangel with Bc5 and not Bb7 due to d3
2. vs 1.d4 play either Queens Gambit accepted or Tarrsach.  Tarrash limits what you have to learn as it can be played vs everything pretty much.  If you decide on queens gambit accepted you should play 1... e5 vs 1...c4 and a quick d5 this is a good system vs English one you can play for life.
2. play 1.e4 as white
sicilian: either open or c3 sicilian, many here would recommend some easy lines in open rather then other anti sicilians.  But if you work ect then 2.c3 is a good variation.
Caro- under 1800 play panov botvinick.  other wise at 2000 play Short advance system.  Why?  Because even Shirov is playing it, it develops your pieces and gets you catled and then play c4.
French: either exchange and early c4 or Nc3.  This has been a big debate if you see other sections on repertoires.  Decide on one and stick with it.  (I know chessmaster PC program Josh Waitzkin analyzes the exchange line well)  Or you can play the Korshnoi gambit as Neil McDonald recommends in his starting out 1.e4 (great sections on Caro kann, French, sicilian (basics Dismantiling the sicilian is a good addition), and follow his recommendations vs everything else....   
so you "need" 1. starting out 1.e4 by McDonald, Meeting 1.d4 by Aagaard, and Nigel Davies book on 1...e5 and Open Ruy lopez by Glenn Flear!  If you like the QGA get James Rizzatano book vs 1.d4.  Now you can blow your money on Tactics and endgame books!!!   
Zatara