We've all worked tactics problems. I can do those pretty well. Yet I have never been able to pull off one of these amazing combinations in my games. Not once.

These tactical situations don't arise from nowhere - they are created.

I had a bit of an epiphany recently. I had viewed positional play as an alternative to Tal-like tactical play. You did one or the other, Karpov or Tal. Now I don't think this is true. Fischer said, "Tactics arise from a superior position." Now I think that positional play leads up to the combination by leaving the opponent no good squares. When he moves on a bad square, a tactic likely appears.

I would like you all to give me advice on how I can learn to find the moves before the combination. I seriously think this is an enormous gap in my knowledge. This is why my games are boring. I shuffle my pieces without a clear plan, hoping for some combination to appear. How can it, when I do not actively seek to make it happen? When I win, it's because I defend well and the attacker destroys himself on me. To make matters worse, when I realized I defend pretty well, I started concentrating on defenses that were solid. A little is good, more is better, right?? This has probably only enabled me to strengthen my bad habits.

Anyway, please, save me from me.

  • The more you practise the more you will be able to pull of amazing combinations. Look for weakness in the opponents style of play and exploit it.
    – pbu
    Jul 24, 2015 at 14:52
  • This sounds exactly like how I play! Nice question!(+1) Dec 23, 2017 at 13:00
  • Minor nitpick: what Fischer actually said (well, wrote) was: “tactics flow from a positionally superior game”. (This is in his annotation of 18.Nxh7 in Game 1 of My 60 Memorable Games.)
    – Stephen
    Jul 1, 2020 at 19:46

6 Answers 6


Your situation sounds a lot like my own - in the past, I tended to use a very "defensive" style of play, preferring closed openings, subtle position play, etc. If the opponent made a mistake and I could win a pawn with some clever tactics, it'd be enough to win the game, most of the time.

This approach is probably perfectly fine if you are aiming for a respectable rating (as we've often heard, simply not blundering will see you through to the 1800s), but for some reason, I began to lose interest in chess (probably from too many draws and eeked out wins).

After a break of several years, I started playing "strictly for fun", with friends in person, and online. This has given me the opportunity to experiment more with play style, and I found out that I actually prefer an aggressive style. It was surprising for me just how often this style of play would lead to the opportunity for exciting tactics. My experience on making the transition:

Play unrated games: It's difficult to experiment (wildly) with your play style while trying to play your strongest game. You should still review your games and try to learn from them, but you shouldn't worry if you lose.

Stop playing drawish openings: Some openings are simply more suitable for aggressive play than others. Prefer open games versus closed ones. As white, you have a tempo advantage, and you should try to make the most of it. As black, you can try to unbalance the game, as opposed to simply letting white run the show (e.g. Sicilian Defense). I've always found the Danish Gambit to be a wonderful opening. Get someone to play this opening with from the "Accepted" position, to appreciate what an aggressive style can lead to - you're two full pawns down, but have a huge advantage in development (aside - unfortunately, the Schlechter variation is strong for black, and almost everyone knows it nowadays).

Don't use your aggressive style on computers: Perhaps it's just me, but computers nowadays are way too good at tactics. They see things coming a mile away, and simply won't walk into your clever tactics, unless they are 6+ moves ahead. It's not fun to keep getting floored by the computer. Humans make mistakes more often, but more importantly, do not see every possible combination to at least 4 moves (unlike computers).

Piece position is important: This goes without saying, but having the pieces there when you need them makes all the difference. Attacks are often won or lost on a single tempo difference, so don't get tricked into giving up tempo, and try to get some from your opponent!

Build pressure: If in doubt, build pressure, complicate the situation. Look for a weak spot, and start aiming pieces at it. Again, you'd be surprised at how often a tense situation can lead to a brilliant combo / quick win.

These are all the basic, non-tactics related tips I have. To reiterate, I think you should just play many games in a more aggressive style. You will probably lose a lot at first, but you will also get a lot more practice in attacking, both strategically and tactically. Eventually the results will come, and if you decide to switch back to a more defensive style, the lessons learned will stay with you.

Edit: To expand on the comment about playing against computers, I find that their simulation of "weaker" human players is quite badly done. One well known program, for example, seems to change the player's heuristics in favour of different pieces ("likes knights", etc), and throws in a few major blunders at lower levels, while playing 99% accurately most of the time, thinking many moves ahead, and having an excellent opening repertoire. I don't really find this to be a realistic representation of a real player. Setting the AI up yourself, to have a shorter think time than you (and to not think on your clock time) may work slightly better, although typically leads to weak endgame play. In the end, it's more fun to just play against people, and with the internet, there's plenty of opportunity for that.

  • 3
    +1, especially for "build pressure". Humans can't stand pressure for a long game and eventually they'll crack and allow a winning tactic (or hang something). Building up enough pressure is enough to even beat computers too! I've had good results in blitz against strong computers when they underestimate my attack and eventually crack.
    – Andrew
    Jul 31, 2012 at 4:10
  • @Andrew absolutely. I've had slightly less luck with computers (probably because I'm not as strong a player as you), but I've also noticed that I can sometimes beat them on blitz. It's just a bit frightening that even during blitz, they have 2 to 3 full moves computed, meaning you have to play very accurately to get ahead.
    – Daniel B
    Jul 31, 2012 at 6:24
  • 3
    If you've carefully amassed an attack against a computer that is thinking 7 or 8 ply ahead, there's that moment when you're thinking "hmmm, there must be a fatal attack here, but where is it exactly?" and the computer starts helplessly sacking its pieces for a few random checks, because it's already determined that you have a forced mate. Which is hardly what a human is likely to do :P Aug 1, 2012 at 3:59

I'd say study some Alekhine. He is perhaps the best tactician before Tal and the best tactician ever IMHO (though, not flawless), since Tal was always somehow easygoing and missed Alekhine's steel willpower.

Amateurs would just praise Alekhine (he is known as the genius of combination, while Tal is known as the wizard of combination) and would judge him by the results of his games. Study some classical combination games played by Alekhine and commented by himself. You will often notice how he would invoke a reason for a possible combination (a weakened field, a material superiority on some board sector etc), how he'd start analyzing variants and find a defense, thus concluding that the combination is still premature. Then he'd make another preparation move. And at some point the position on the board just "explodes".


I shuffle my pieces without a clear plan

Here lies the heart of your question, IMHO. I'm not a GM, but I can tell you how I find my combinations during play.

  1. "Study the battlefield". Each opening has weak points. For example, in the French Defense, Advance Variation, Black may have problems with f7 and g7 pawns. Other example, in the Italian Game and many gambits after 1. e4 the weak point for Black is f7. You should always keep an eye on those spots, waiting for the right moment to strike. Sometimes you can look for combinations related to those squares (in the example above, a Queen exchange followed by a Black King capture, then Nxf7 forking King and Rook), or try to exploit other weaknesses when your opponent's pieces overdefend those squares. Of course, if you know where to watch you'll find those combinations more easily.
  2. focus on every possible tactical problem your opponent may encounter. Then you'll spot them in the exact moment they appear. Does your opponent may be mated on the 8th rank? Does your opponent's Queen protect 2 vital squares at the same time? Does your opponent have a weakness and a hanging piece at the same time? And so on. Detect these situations, in 10-20% of the cases they may become the key for a combination.
  3. improve your position threatening something. If your goal is an aggressive play that might win a full point during the middlegame through a combination, you have to choose from many possible moves the one that make your position better AND threats an attack. If you are able to conduct your pieces well, then one of those threats will become the combination you search.
  4. when you're in great positional advantage, look for every possible way to worsen your opponent's position. If sacking a piece may crush a defense, don't exitate. Be careful, though: we're not Tal. :) If you find a safer way to win the game, DON'T force an attack.

I play always with these points in mind. I usually win my games during the middlegame, gaining decisive material advantage or even forcing a mate. Of course you cannot forget the golden-rule: improve your position at every move. Don't go chasing combinations that do not exist. But I think you're not that kind of player. :)

A final note: since you're a good defender, you may play solid games waiting for your opponent to destroy himself with his own hands. Petrosjan became World Champion that way. :)

  • 1
    Tactics over the board are generally the result of opportunity, rather than planning. You learn tactical patterns, then watch for the opportunity to play them, while playing good moves in the mean time. You can improve your chances of the opportunity arising by positioning your pieces to take advantage of them, but you can't force them to appear. A combination is the result of an opponent's mistake; it's what happens when preparation meets opportunity.
    – Arlen
    Oct 22, 2012 at 12:52

If you want to get better at combinations, play tactical moves, such as pinning pieces with a bishop, moving rooks to open files, launching double attacks with a queen and a second piece, etc.

In one of my first combinations (playing Black), I had a rook on a king file with two bishops between it and the enemy king. I moved the first bishop e7xNa3. White had to recapture. Then I moved my light-squared bishop, e6-g4, attacking the queen on d1, and discovering a check with my rook on e8.

If you look for them, there will be plenty of opportunities to set things up. Then let your opponent look for a way to avoid them.


One idea which helps a surprising amount when considering tactical possibilities is to look at what you are thinking about, and reverse the order of the moves.

If you think you can checkmate the king by sacking the bishop first then the rook, try briefly considering the opposite, and then consider other permutations. Doing this can make things clearer, especially if you are trying to calculate many moves deep.


Possibly the most pertinent question I have seen on here so far. For me the answer is simple, get a book on attacking. The classic "Art of Attack in Chess" by V. Vukovic is highly regarded, and a for a modern classic the "Attacking Manuals" by Jacob Aagaard is also good. Both of these books focus on how to get an attacking position where combinations are much more often seen.

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.