WSS wrote on 01/28/14 at 10:44:10:


I like Yasser Seirawan's take on that game: http://www.youtube.com/watch?v=g1gWIfVajlM&list=PLVWaFpMwtaGiVZ77NhhvGGGzvF7oFSW..., especially beginning around the 15-minute mark.  Seirawan is a genuine Pirc guru, having played it his entire life, and I respect his opinion a lot in general.  That game is a pretty good illustration of how not to play for a win as Black, in my opinion.  It's clear that Kramnik didn't know the nuances--not from a theoretical perspective, but a practical one.  Seirawan may not be as strong as Kramnik anymore (and likely never was, though Seirawas was an extremely strong player in his day), but with his practical experience he immediately knew where Black began to go wrong and suggested improvements.  

Of course, Kramnik faces an entirely different problem in the last round of a tournament that he needs to win than we do!  So we can forgive his choice.  

In my own experience, I found that Alekhine's Defense was a good offbeat/secondary weapon, because I could get it on the board on move 1.  On the other hand, I played it out of curiosity and want of variety rather than searching for winning chances.  But a secondary defense should be at least 95% sound, and you should be able to get it on the board very early.  I wouldn't think that 1.e4 e5 makes a very good secondary defense, unless you already have a ton of experience with it in the past. 

All good advice indeed.  Practically speaking, I would do the same regarding sticking with what I know well - particularly if contending for a tournament title.  

I guess what may have inspired my question (aside from the pain meds I am currently taking!) was Loek van Wely's terrific win over Aronian in the last round of Tata Steel.  The Dutch defense was a fascinating choice (which he doesn't seem to play it too often according to my db) which led to a tension filled game (in addition to being quite thematic!) and an exciting finish.  A somewhat different scenario than I painted in my OP as Loek was not near the top of the standings and had less to lose.  The flip side of this is Kramnik's loss to Ivanchuk with 1.e4 d6 and the Pirc in the Candidates trying to create a fight at all cost - a decision I'm sure he wishes he could take back.

All of this led me to thinking about which defenses which probably would not be your main every day defense (thus adding a little surprise factor) that give the most chances of leading to a tension filled, decisive game with a corresponding higher risk of losing.  I didn't select the Sicilian or Slav (both of which I play) because they are so "well trodden" and lines like the c3 Sicilian and Exchange Slav can be a "buzz kill" for the Black player looking for a fight.

Anyway, interesting discussion and thanks for the responses!

I think that people should play their normal openings, unless you are playing someone with a very high level of overall chess technique: IM or at least 2300ish, let's say.  You're much more likely to win when you know what's going on, what to do in the positions you get, have a lot of experience with the structures, and don't fall behind on the clock working everything out.

On the other hand it's easy for me to say that, as my normal openings give excellent winning chances anyway: the Taimanov/Paulsen/Kan complex vs. 1.e4, and against 1.d4 the Noteboom/Semi-Slav or Classical Dutch have been the defenses I played the most.  

Obviously it doesn't hurt to be able to play multiple defenses (I've also played the French, the Caro, Alekhine's, the Nimzo/Blumenfeld, the Grunfeld, the Tarrasch, Classical Slav, various QGDs over the many years I've been playing), but in general I've never had problems finding winning chances with whatever was my "normal" opening at the time.  You can often vary within your main opening to get more winning chances instead of scrapping it for something else entirely.  If you play the French you can switch from Classical to the McCutcheon, and if you're worried about the Exchange, there are always opposite-side castled positions or playing 4.Bd3 c5 with an IQP, which should be enough to outplay an opponent.  Overly sharp tactical lines are often NOT what you want to play anyway in my opinion, because they are prone to perpetuals, mass exchanges in some variations, having been worked out very deeply, etc.   

If you're facing someone much stronger than yourself your best bet is to stay "in book" and play like a Grandmaster as long as possible, wouldn't you think?  Which means playing your normal stuff, the stuff you know best.  I'd think that the only time you'd consistently get more winning chances by doing the opposite is when you're facing someone much weaker than yourself and trying to avoid them following a long theoretical line.

1.e4 - Sicilian Najdorf
1.d4 (1.c4, 1.Nf3) - King's Indian