6

In many of my games, I've come across opponents who can calculate really well and deeply. Generally I feel that I lose if I can't calculate as well as my opponents. So my question was really if calculation was the most important aspect of playing chess. Would the person who calculates better normally win?

5
  • Might be a bit too opinion-based.
    – Annatar
    Oct 1, 2020 at 6:01
  • 4
    In my opinion, the most important aspect is pattern recognition. Good calculation flows from that.
    – Annatar
    Oct 1, 2020 at 6:02
  • 1
    Does calculating well imply ability to identify promising paths to calculate and calculate deeply and accurately or is it just ability to calculate deeply and accurately any given line?
    – Akavall
    Oct 1, 2020 at 16:48
  • To add more info and for the sake of the argument it'd be interesting to note that quite strong chess engines can be build with a very simple evaluation function; material plus piece square tables (basically a very simple way to bonus pieces for being "well" placed on the board. No mobility, king safety, pawn structures, etc). For example PeSTO got very good results with this approach.
    – emdio
    Oct 2, 2020 at 10:27
  • 1
    @Annatar's comment is underrated
    – Darren H
    Oct 5, 2020 at 9:04

4 Answers 4

7

Would the person who calculates better normally win?

If everything else is equal then the answer is obviously "Yes".

However, everything else is rarely equal. Fixating on calculation is very common in chess players. Here is another question about calculation which clearly demonstrates this fixation. The fixation is so clear that the best answer easily pinpoints this in the first two paragraphs. BlindKungFooMaster then goes on to make the key point which is -

In my experience ratings first and foremost reflect positional understanding. Calculation skills make a much smaller difference, at least above a certain threshold.

This fixation with calculation is, when considered rationally, obviously wrong. The strongest chess is not pure computer chess, chess based on pure calculation, it is correspondence chess where a human being moderates pure calculation with their positional understanding.

This was certainly the case before AlphaZero. AlphaZero's recent success is down to an attempt to incorporate positional understanding into computer chess which has worked better than more simplistic ideas like adjusting scores according to ideas like doubled pawns, central knights versus dim knights on the rim, king safety in the middlegame, centralised king in the endgame, etc. It would be very interesting to see correspondence players being able to use AlphaZero-like computers in their competitions.

2
  • "The strongest chess is not pure computer chess" - citation needed. Sure, a human + a computer beats the computer, but does a human + a computer beat a bigger computer?
    – user253751
    Oct 1, 2020 at 13:04
  • Correspondence players DO use AlphaZero-like engines in their games (say, LC0) and they're all beaten handily by Stockfish (as of today) given similar energy expenditure, which one could say is a "pure calculating" engine. Humans are really bad at calculation in the grand scheme of things, but are good at other things, so we use pattern recognition and some refined heuristics while playing chess, but calculating more would definitely help.
    – 12345ieee
    Oct 1, 2020 at 22:22
3

Indeed. Accurate calculation is the source of all success in chess. Strategic ideas will only work if tactics support them. While other aspects of the game are also helpful and can make the difference, nothing can close the gap if there's a big difference in calculation.

2

evaluation is the most important aspect of chess. The ability to judge if the position is good or bad without needing for deep calculation is vital. There are limits to how fast humans can calculate and in the end, you need to know if the line you calculate is good or bad. Though admittedly, the terms are somewhat vague. Does a grandmaster beat 50 people in a simul because he calculates better than his opponents. In some way you could say yes, when he calculates he is able to pick out the most important lines. Of course he also wins because he doesn't need to calculate as much because of his positional understanding, his ability to evaluate positions without spending much time or effort on it. So depending on what you mean by calculation ability, you may come to different conclusions. As an aside, stockfish examines 100 million positions a second. Alpha zero examines 100,000 positions a second. How many positions do humans examine a second?0 to 10. No doubt if computers were limited to this computational limitation, stockfish and alpha zero would be crushed by professionals, probably even amateurs.

2

YES! Calculation is the most important skill in chess.

Why do you think computers are stronger at chess than humans? It is not their positional understanding or long-term planning; It is pure brute force calculation.

When I was rated 2080 I prepared for a tournament just by studying puzzles. I studied for hours a day and my head hurt from it. I took a couple days off before the tournament so I could be rested. Then I had a performance rating of 2500! I didn't get any opening advantages, in fact some I was slightly worse positionally; but I managed to get complicated positions and found nice tactics.

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.