I have recently picked up the Norwegian Defense (also known as Northsea Defense), 1. e4 g6 2. d4 Nf6.

All modern engines claim it is not good. But I have great results (14W-1D-5L in 15+10 on 2000 Lichess rating) with it and I am seriously considering adding it to my repertoire, so I would like to do some preparation work, look for critical positions where I would have to know an exact move or a plan to justify my opening.

First I have looked at games in the Lichess database. I was very surprised by Black scores in some of the variations. For example

[fen ""]
1. e4 g6 2. d4 Nf6 3. Nc3 d5 4. e5 Ne4 5. Nxe4 dxe4

scores 55% for Black in 2000+ rapid and classical on Lichess.

I have also looked at the Master database. I have noted that the character of the game is mainly set within 5 moves by White and gathered some common resulting positions. From there I wanted to work further, so I looked at master games more indepth.

Some are clearly surprised by the opening and choose a timid way to play and Black equalises before move 10. In the more critical tries White keeps the advantage for several more moves, but then lets it go or even slips into a disadvantage. Also, the games where both sides offer resistance are very rare. Very often on move 7 there are 2-4 games.

Following engine lines and picking up moves that were missed is rarely helpful, as the engine of course will mercilessly squeeze the position with inhuman precision and will disarm any initiative without effort, making any active attempts seem futile. Sometimes I do not see why White is better, so I follow the line into endgames that are slightly better, but probably considered equal at sub-master level.

Of course I analyse my own games, but the opponents crumble surprisingly fast once on their own, so the opening does not really get tested much in my own games.

So, how would one proceed to prepare in unorthodox openings apart from learning by his own games?

  • You're talking about millionths of a pawn. You can get by against real opponents playing common sense moves in an opening. Commented Sep 10, 2020 at 13:01
  • I assume you've looked at the transpositions into Alekhine's Defense that occur after 3 e5 Nd5 Commented Sep 10, 2020 at 22:24
  • 1
    @AlexanderWoo 3. e5 Nh5 is the way to go, an option that is not there in the Alekhine
    – B.Swan
    Commented Sep 11, 2020 at 10:41

2 Answers 2


Forget the engines. We are humans (hopefully) and are only interested in how human's play.

The King's Indian Defence (KID), a favorite opening of Tal, Fischer and Kasparov (and now played by Nakamura), has a reputation of being unsound by the engines (some positions are +1 and even +1.5) however for white playing precisely is very difficult.

For instance, the Mar del Plata Variation

[fen ""]
1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7. O-O Nc6 8. d5 Ne7

Stockfish evaluates it to be +1 at a depth of 40.

Simple advice for practical play is to ignore the engine evaluation and focus on the fundamentals in the opening (in the order):

  1. Exert pressure over the center (either classical or hyper-modern way)
  2. Develop minor pieces
  3. King safety
  4. Connect the rooks while developing the queen

Needless to say always look out for short tactics as you follow the above and also try to have all your pieces coordinated.

  • 2
    +1. Someone once said, don't trust engines when it comes to openings. I do think engines needs to change its evaluation function in the opening. Commented Sep 10, 2020 at 13:22
  • 1
    The luxury a KID Player has and a Norwegian Defense player does not is the huge amount of games and books on the KID which the ND lacks, and it is precisely the reason for the question. I am not overly worried by the engine eval, but how else to find critical variants when games are sparse? Even in the KID positions arise where you have to be precise as Black, but probably you would not find them by yourself from scratch.
    – B.Swan
    Commented Sep 10, 2020 at 13:36
  • It's not that engines need to change how they evaluate openings. it's that players need to change how they evaluate engines. Commented Sep 10, 2020 at 13:37
  • @B.Swan My 2 cents would be if you are able to get good results, keep playing and you will learn as you along (of course you will make mistakes but that is how learning is). For all you might know, you could end up rejuvenating the Norwegian Defence and end up as its leading proponent. :-)
    – Adhvaitha
    Commented Sep 10, 2020 at 13:42
  • @friscodelrosario Ok, I take your point. Commented Sep 12, 2020 at 6:44
  1. You should know how to evaluate an opening position- center, development, king safety. While an engine can be useful for missed tactics, I trust my own evaluation over any engine in the opening.

  2. Lopsided database scores can be a clue that the opening may be difficult for your opponent but they aren't perfect particularly at higher levels.

  3. Using a GM database and an engine find the most difficult line your opponent can play. If you can live with the end position then it's probably fine. If not, it still might be playable depending on how often that line happens in practice.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.