Squeezing Water from a Stone

An Unsurprising Outcome

"I Don't Believe in Fortresses" - Magnus Carlsen

A few days ago I tuned in to a broadcast of the FIDE World Blitz Championships just as they were analyzing Magnus Carlsen's round 16 game against Bu Xiangzhi. Around move 25 the commentators declared the game a likely draw and moved on to covering other more interesting and dynamic struggles. When I checked back in at the end of the day, while Magnus winning the event came as a bit of a shock, the result of this individual game was almost expected.

Arguably, Magnus' greatest contribution to the chess ethos is the idea that truly dead drawn positions are rarer than we thought. These victories are not the result of some idiosyncratic superpower that only Magnus possesses, in practical play there are always more opportunities to apply pressure than it seems at first glance.

On Engines and Endgames

The term "dead drawn" has become a pet peeve for me over the years, I find that it rarely means "I looked at all the plans in the position and they get nowhere" and much more often means "the engine is giving 0.0 after 5 different moves and there aren't many pieces on the board". Due to the monstrous defense capabilities of modern engines, and the perfect defensive capabilities of a table base, these resources can be shockingly uninformative when it comes to practical play. Normal and stratified engine evaluations are useful because they give us information on who is pressing for a win and how narrow the path is for both players. Weirdly, as a position becomes more comprehensible for an engine the less useful the outputs become for us.

All that to say, don't trust the engine's 0.0.

Understanding the Process

1. Generate Any Positive Imbalance and Limit Counterplay

How to arrive at a grindable position is a bit beyond the scope of this discussion but both of these components are integral to setting up an endgame grind. Obviously, it is difficult to cause problems in a completely symmetrical position (eg. if you have Rook and 3 vs Rook and 3 on the kingside it may be time to shake hands), and a position where your opponent has a clear plan, real winning chances or straightforward counterplay cannot qualify as a grind. One point that I would like to emphasize here is that as long as your opponent has limited counterplay, imbalances that seem completely inconsequential can explode into decisive advantages surprisingly abruptly.

Just to provide a list of some potentially decisive imbalances (for those who have not already read Silman's excellent books):
1. Material
2. Advantaged Minor Piece
3. King/Piece Activity
4. Pawn Structure/Weaknesses
5. Pawn Advancement/Space
6. Control Over Key Squares/Files

In Magnus' game, for example, he was able to win only with a slightly advantaged minor piece.

The above position will be explored more in the next section, at first glance, who (if anyone) do you think has the advantage? Is it enough to apply pressure? Can you come up with a plan for either side to do so?

2. Needle Your Opponent

This step is as much about psychological and practical exploitation as it is about the position itself. In general, as the "grinder" you have two types of moves you can play

1. Play a (ideally noncommittal) move that pretends to do something/makes a small improvement

2. Play a move that poses problems to your opponent

Why would we ever waste time with the first type of move when the second is an option? It has to do with what our opponent wants as the defender. Ideally, they would like to create a fortress where they can repeat a simple pattern of moves to hold a draw, and they are also hoping for a position where they can directly and simply respond to each of our threats. By alternating between these two types of moves and "pump faking" them, we can force our opponent to switch between these active and passive mindsets, ideally burning their clock, lulling them into a false sense of security and exhausting them in the process. After all, why do in one move what you can do in ten? Your opponent will struggle to sit still while you slowly improve your position.

Once your opponent figures out a defensive setup and repels all your threats, then it is time to change the setup by making a trade, pushing a pawn etc. and repeat the process with a new position. Often, either the defensive task will become so exhausting that it leads to a tactical breakdown, or a desperate, risky attempt to force the game into a draw. Most players are afraid to sit still and will quickly start seeing ghosts in the position.

This process is not simple at all, it requires creativity in terms of finding plans that disturb your opponent's equilibrium and you must constantly spurn your opponent's attempts to relieve the pressure and liquidate to a draw. That being said, it is always more fun to be the one pressing for a win and extending the struggle takes maximum advantage of that.

3. Pounce On Their Mistake

Self explanatory, but it is important to remember that as the one pressing you have almost limitless opportunities to pose problems. The defender only has to make one inaccuracy before a position that seems lifeless falls apart.

From Magnus' Game Against Bu:

Walking Through A Few Examples

Game 1: Creativity in a Dry Position

I was the victim of a very nice (though I couldn't fully appreciate it at the time) example of turning a seemingly small imbalance into a win a couple years back

There is always a way to ask questions of your opponent if you are brave enough.

Game 2: A Pure, Methodical Grind

To Summarize

The advantage you need in order to generate serious winning chances in a game is much smaller than many people think. Defending is very difficult psychologically and it is the nature of some positions that you can force your opponent to endure for a very long time, even longer if you know how to execute your plans at a meandering pace when appropriate. As long as you are creative, confident and patient, even the unlikeliest of wins can be within reach.