GM Joel Draws on his Experience
By GM Joel Benjamin   
June 11, 2008
Dear Joel,
In high-level chess, what would you estimate is the ratio of draw offers that are Accepted/Rejected?
Do certain players have a reputation for more frequently rejecting draw offers; and if so does that tend to deter their opponents from offering draws to those players?
Do certain players have a reputation for more frequently offering draws?

 I would say the percentage of accepted draws is pretty high.  Grandmasters generally do not offer a draw when they think they will be turned down.  There are cases where draw offers are made psychologically, but generally the player turning down the draw gains a little psychological momentum.

Also, though it may not seem to be the case to you and many other fans, grandmasters are trying to win their games.  I believe most draw offers occur when the game is clearly headed for that result.  Though draw offers are noted on scoresheets according to FIDE rules, they rarely survive into databases (much like clock times, a point often lamented by Daily Dirt columnist and ICC commentator Mig Greengard).  Thus there is really no way to obtain accurate statistics on declined draws.

Players with uncompromising fighting spirit are expected to turn down draw offers.  It’s likely that they receive them less often.  Examples would be Veselin Topalov and our own Hikaru Nakamura.

Players who draw most often probably offer them most often as well. [Though I think it is quite irrelevant which player offers a draw in a clearly drawn position]  If the rules are followed, the last player to move will be the one offering the draw.  So you could research the numbers, but it would be quite pain staking!

Dear Joel,

Have you ever known a case where a player included a draw offer in his written sealed move for an adjournment?
I’ve heard stories of players writing something other than a sealed move on their scoresheets, but these messages are usually much more unfriendly in nature!  I think players have offered draws verbally during the break, with opponents occasionally insisting on seeing the sealed move before agreeing.  In any case, adjournments are now relics of the past.

Dear Joel,

The following draw-related snip (page 69) from your book "American Grandmaster" helps me see why some people say that games from team events should not be officially Elo rated:
"... we lost to France in the last round, despite the fact they forfeited a game. Schiller asked me to offer a draw against Santo Roman to clinch fourth place for the team. I had an excellent position and agreed with reluctance. My opponent did not get to answer because the French captain jumped across the [table] to shake my hand. I didn't mind the draw too much because I had already clinched the board prize for first reserve."
(Note: Joel was White. After 31... Rc8-f8 material was equal, and Joel offered the draw. But Fritz_9 evaluates Joel as better by +1.8 with a heavy kingside attack.)
So, should games from team events be excluded from the official Elo ratings?
Would the Sofia draw-reduction rule be opposed by team captains at team events? Schiller would not have been able to request that you draw.
Would adopting Sofia at team events make the games more worthy of being officially Elo rated?

Thank you,
Gene Milener

Dear Gene,

Some team events aren’t rated.  U.S. Chess League isn’t, though those games aren’t played face-to-face.  As far as I know though, every major European league Elo (FIDE) rates the games.  Every major regional, club, or age-related team championship rates the games.   As for the Olympiad, there is no way the games would ever not be rated.  The Olympiad generates a ton of title norms, and for players from many countries, it provides the only opportunity to shoot for these norms.  A lot of players don’t have other opportunities for FIDE rated games anywhere else.  The Olympiad has to be FIDE rated, and it should be.  Why shouldn’t all those (relatively) slow time control games between grandmasters be reflected in the world rankings?  The percentage of games where your scenario occurs is pretty low.  We don’t want to throw the baby out with the bathwater.

Use in the Olympiad of the so-called Sofia rules would be opposed by team captains, players, heads of delegations, you name it.  Gene, I know you have expended a lot of thought and energy towards draw-reducing options, but sometimes a draw is okay.
We have to understand that the Sofia rules exist for tournaments that are sponsor and spectator driven, e.g. small round-robin GM only events.  The Olympiad is first and foremost a contest of national teams, and they should be allowed to compete without hindrance of artificial rules.  [By the way, I think move minimums are better than the Sofia rules, which turn top professionals into children when they have to ask the arbiter for permission to draw their game]

If the third board from Mauritius and his counterpart from Fiji want to draw, why should we stop them?  And how many of the dozens of arbiters at the Olympiads are qualified to make the judgments required by the Sofia rules?  In the Olympiad, some draws can be selfless and beneficial to the team.  A lot of captains have begged a tired or sick player to get out of bed and hold off a key player on the other team with a draw.

I think the Sofia rules would distort the natural course of events in the Olympiad.  If anything, they would make the games less worthy of being rated.