I need some players who play at the ICCF (International Correspondence Chess Federation) to answer this question.

What would approximately be the ICCF Elo rating of the two following players?

  • Player A is a Grandmaster with an OTB rating of 2500 Elo FIDE. He will not use any engine, nor will he use any opening database, nor will he use any opening book, nor will he use any endgame tablebase, nor will he use any form of outside assistance. He will play using only his brain. He will use on average about 5 minutes per move, which also happens to be the average time per move of humans in OTB chess at classical time control. Player A is 100% human.

  • Player B is in reality a patzer, with an OTB rating of 1200 Elo FIDE. For the opening he will use an opening database and in each position he will just pick one of the popular moves which score well. For the rest of the game, he will use one (and only one) of the top 3 best engines: Komodo 9, Stockfish 6 or Houdini 4 (64-bit version). His engine will have 7-men endgame tablebases. At each position he will let his engine run for one hour (using 4 CPUs) and then he will blindly pick his engine’s first suggestion. He will not offer draws or accept draws early, he will continue his games until it is very clear that the position is a draw. He won’t resign early either, he will wait until he gets checkmated. Player B is 100% computer.

Note that I only need an approximation. But still, try to make your approximation as accurate as possible.

  • 1
    I really don't see how you can get a solid answer to this. It is patently obvious that Player B is a far stronger beast than Player A because the superior playing strength of the engine far outweighs other factors. But as to how each would perform relative to the pool of ICCF players, that depends primarily on the proportion of ICCF players that choose to compete with engine assistance, as opposed to those who prefer to play without it.
    – ETD
    Commented Jun 17, 2015 at 21:06
  • 1
    This question is not directly answerable. There is no formula because nobody will play seriously at ICCF without a computer. The point of ICCF is to play the best possible chess quality, think for a 5-min for a move is pointless. Why not just go for OTB chess? "He won’t resign early either, he will wait until he gets checkmated" is a bit silly, whether you resign early or not makes absolutely no difference in a correspondence game. If you're down by a piece, you'll lose, sooner or later.
    – SmallChess
    Commented Jun 18, 2015 at 1:40
  • 1
    Very obviously there are so many clueless people who play at ICCF just like player B. 99% of their strength comes from their computer. They just play at ICCF to inflate their ego. So one of these people (or someone who knows one of these people) will easily be able to answer this question.
    – Fate
    Commented Jun 18, 2015 at 7:16
  • "There is no formula because nobody will play seriously at ICCF without a computer." You're wrong, there are some people who play at ICCF without any engine. If they encounter someone with a strong engine, they will lose. So typically their ratings are low (but how low exactly I don't know).
    – Fate
    Commented Jun 18, 2015 at 7:16
  • If you want a simple proof that there are people who play at ICCF without any engine, just look at the ICCF ratings of some players. Some players have a rating of less than 1500. I even quickly looked at some of their games: they make some stupid blunders that obviously no engine would ever make (OTB they are probably less than 2000 Elo). Here I made a screenshot of my ChessBase's Correspondence Database: s15.postimg.org/6tsewa0dn/iccf_ratings.png
    – Fate
    Commented Jun 18, 2015 at 7:16

2 Answers 2


I play on ICCF. My guess is that the engineless GM might be able to maintain a 2000 rating by beating other engineless players or players who are bad at engine use.

I don't think the second player would do that well actually. Players in the 2300+ range are all using anti-engine strategies, following GM opening theory, have their novelties, so they can lure him into opening traps or positions his engine has trouble evaluating. I would guess he'd wouldn't go much beyond 2100. I've played against 2200 players who are pure engine once out of book, but they put more effort into their openings than your player B does.

There is another breed of player you haven't mentioned: the engine monkey master. These are players with 2000+ ratings over the board. They could potentially try to carry out strategic plans in the middlegame, but they get so caught up in the engine analysis they end up play positionally suspect moves. They do contribute a lot to the openings and endings though, so you see them in the 2300-2400 range.


Player A will have difficulties to achieve 2000-2100. Player B has something like 2400 guaranteed, but he will need to play many games to achieve that rating. Player A can start with his FIDE rating, but he will go down without any computer help.

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.