Normally when I play against a computer I try to avoid any positions where I feel the knowledge-base of the computer would give it an advantage. For example I find computers very strong in typical protracted opening sequences and in endgames.

I seem to fare best against strong computer opponents when I play more positional, "by the gut" chess than when I attempt to apply a more strategic and tactical game plan.

I am wondering if this is something other people have experienced; if there is a practical basis for this belief and what type of chess situations have proved fruitful for them when playing against a digital opponent.

  • 1
    Perhaps try avoiding crowded opening ? The more pieces the more theory and calculations ( in the advantage of the computer ) I think.
    – mick
    Commented Oct 12, 2012 at 15:44
  • 4
    Pull off the power supply when you're losing...
    – SmallChess
    Commented Jan 2, 2017 at 3:26

5 Answers 5


Interesting question, but it's a little bit hard to answer generally. Any engine has its own quirks, but I'll try to give a few general hints and things to consider.

First, any modern engine on sufficiently fast hardware is going to wipe you off the board. Even top GM's lose to engines with odds. 1 However, most engines have the hardest time in closed positions where long-term strategy is more important than tactics on any given move. Games that are very closed (think classical KID lines) are won or lost based on chess understanding, and engines still don't truly understand chess.

One very interesting thing, however, is that some engines actually are vulnerable in extremely tactical games. Most notably, engines are not always as good as humans when one side has a large attack. Engines are notoriously bad at overestimating defensive resources and going into a line that is losing for the defender. However, there is absolutely no margin for error in this case since a single misstep means the loss of the game. This method (i.e. attacking the computer's king), is a very easy way to win against slightly weaker engines - for example the ones built into Windows and Macs (also the chess games on international flights, but those are really awful...).

To finish, probably the only way to have good chances against a strong machine are to memorize the moves ahead of time. This might sound fantastic, but it's already been done. The point is that the engine will spit out the exact same moves every game (under identical conditions), so it's possible to find a hole in the engine's repertoire and steer the game into this gap.

1. Rybka has played many odds matches - most notably against IM Meyer, GM Ehlvest, and GM Milov (who actually won the match!).


Yes, that generally happens with me as well. I have been playing a lot of chess at a younger age with my gut instinct and now, when I move to a more tactical game by seeing professional games and chess sites, and try to implement the same during my game, I am completely lost.

Gut-moves generally work well as these are the moves that one has generally made while playing for a considerable amount of time. Going for a drastic change from the "Gut-game" to a tactical game plan would be a hindrance as the board position would be something totally new to what one is acclimatized. A gradual move to a more tactical game plan is suggested.

I also feel that the earlier in one's chess life, they learn the tactical game plan, the better it is.


Positional play, even sacrifices like "poisoned" pawns. The computer won't see an immediate threat and hence grab the material, but in the long run you can use that weakness (double pawn, bad bishop etc.) to your advantage.


I believe that the OP's observation is correct. A computer is worst at evaluating "intangible" advantages, and best at calculating "lines." So to the extent possible, a human player should aim for the former and avoid the latter.


I've had good results against the computer by avoiding it's strength, which is tactics. Try to keep the game closed and positional for that reason. Also, book opening lines are programmed into the computer, so avoiding them is a good idea. As far as end games are concerned, I've found them weakest there, where the lines extend beyond their horizon. Of course, I'm not speaking about the very highest settings where that wouldn't apply. If fact I've had my best results by heading for the ending with at least an equal position and outplaying them there. And since the computers are so materially oriented, I've been able to make successful sacrifices against them when the position warranted since they'll typically grab the material.

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.