We would be considering, of course, the over the board rating of the respective machines and/or players. I'm ultimately wanting to see evidence that suggests there is an increase of draws amongst machines with modern technology paired against one another, over humans paired with similar ratings. For example, if we compared games with both players at 2700 strength, would they have a higher draw rate than the best machines, when paired off opposing one another?

I know that this tends to happen when considering humans alone. Does the trend extend, ratings wise anyway, beyond human playing strength?

  • Difficult to compare these two, since the machine draws are always "fighting draws" while with humans other factors such as the tournament situation, piece color, tiredness.... can play a role. In the hypothetical situation that you find two human players at 2700 strength and can make them play at maximum strength (using all of their precious home preparation), I would expect the drawing rate to be lower than for computers because humans (even 2700s) tend to blunder more. Commented Mar 10, 2017 at 10:10
  • I'm closing as this link asks a similar question: chess.stackexchange.com/questions/14813/… Apologies to the board. Commented Mar 10, 2017 at 13:49

1 Answer 1


The draws among chess-playing machines tend to be far more than players of 2700-plus Elo Level. Of 100 games, and I mean in 100 matches, machines will have 60-70 draws compared to humans where they may have 50 draws.

Now the score can fluctuate a lot from the players side owing to different conditions .

  1. Both players are not in a Spirit/mood for fighting as like Kasparov/Karpov 1985 WCC.

  2. Unlike a computer, a strong 2800 player may not be in his best form. He may have different reasons like health/family problems, or other factors which may cause a setback .

  3. A Player may even make a basic tactical mistake and lose the game like Anand did against Carlsen in 2014 WCC match.

These are some reasons that a machine does not encounter at all, and plays freely. You can even check the database among players, and if possible there is a lot fo history associated with famous chess games. You could find many interesting stories which would adjudge many of the reasons above and even attract various other info.

  • 1
    Where did you get your numbers? Commented Mar 10, 2017 at 11:01
  • These are not exact but approximate . If you want precise then please check in database and filter out the Games of 2700 Players over the last 1 to 5 years . I wrote numbers can fluctuate but the above factors are sincere and cannot escape . Commented Mar 10, 2017 at 11:23
  • 1
    So you just invented the numbers? Commented Mar 10, 2017 at 11:27
  • Are the numbers very incorrect ? It is based on some Chess Articles I read in past. Commented Mar 10, 2017 at 11:34
  • 1
    Well, that's exactly the question. I think Paul expects this to be true, as you do, but he's asking what the truth is. I think the TCEC engine championships have about a 52% draw rate, which I think is less than the top humans, but there are also some larger differences between the strengths of different engines so some more analysis is needed. Commented Mar 10, 2017 at 11:37

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