Players should be given different openings. Some say that this may favor one player over the other, but at the top level the player must show they can draw in unfavorable positions instead of relying on their home preparation.
Let's face it, if it weren't for home preparation we wouldn't be seeing all these draws. Caruana would have lost two by now. Carlsen is going to ride him out and pulverize Caruana next week in faster chess.
In 2011, the Grandmaster Rustam Kasimdzhanov published an open letter to FIDE and the chess world on what he saw as the problem that draw were causing for Chess.
He began that letter by comparing Chess to Tennis, and he noted that there are no draws in Tennis, and, therefore if Chess were modified to eliminate the draw, it could be as popular as Tennis.
While much of his argument is valid, I think he got off on the wrong foot there. Tennis is often played by athletic young women in short skirts. Even when men are playing it, since it is an athletic competition, and not an intellectual one, the fact that the world's top tennis players are facing each other doesn't make what they're doing harder to understand.
Of course, these are differences between Chess and other more popular sports that we can't do anything about. But in addition to dispelling unrealistic hopes, clear thinking on this, I believe, also will help guide us in doing the things we can do.
Chess is not Tennis - nor is it Hockey or Wrestling. Chess has its own unique appeal to those who are interested in it. So the first thing to be conscious of is the unique nature of Chess.
Yes, Chess is a sport - it has competition, winners, and losers.
But Chess is also an art and a science. If we diminish either of those two aspects of the game in an effort to make Chess a better sport, that effort may not produce the desired result in making Chess more popular.
The popularity of Chess took a hit when the change from the Romantic era to the Modern era took place due to Steinitz.
Chess became more of a science because it was better understood thanks to him.
But not only did it suffer as a sport, as now there were more draws, but it also suffered as an art; games included fewer brilliancies, fewer daring gambits, fewer dashing piece sacrifices.
Obviously, this can't be fixed by asking players to pretend they didn't know about the things Steinitz told them. That would destroy Chess as a sport, since now results would depend on how conscientiously players followed this directive to make chess more appealing, versus how they shifted to worrying about results instead.
Really tight time controls, perhaps used in association with the Armageddon format, may help Chess as a sport, but besides being unfair to non-blitz players, they work against the appeal of Chess as a science, since the best chess can't be played under such conditions.
So in my opinion, the question is: how can we reduce the number of draws while doing as little violence to Chess as possible?
I've been thinking about this for a while, coming up with a number of different ideas. One that I have at present works like this:
Checkmate: winner gets 1.000, loser gets 0.000
Stalemate: winner gets 0.600, loser gets 0.400
Perpetual Check: winner gets 0.510, loser gets 0.490
Bare King Rule: winner gets 0.501, loser gets 0.499
By "Bare King Rule", I mean the following: a player who captures all of his opponent's pieces except the King may then, on his next move, replace one of his pieces, other than his King, by a Pawn. If he chooses to do so, he forfeits his ability to win by stalemate or perpetual check; instead, he can only win by checkmate, and when he does so, he only scores 0.501 to 0.499 instead of 1 to 0.
The idea is that when a player has insufficient material to even do perpetual check, but has achieved a slight superiority, then he can put points on the board, but only if he can play things so that a winnable K+P vs K endgame is created by the circumstances under which he captures the opponent's last piece.
Since now there are more outcomes, placed closer together in the continuum between the two checkmates, although there is still a small space for a draw, the chance that the outcome from one game to the next will vary - even if the variation is between White sometimes winning by the Bare King Rule and sometimes winning by Perpetual Check - is increased. And as long as the outcome is variable, instead of being almost always a draw, or almost always one particular kind of win by White, matches will tend not to be tied.