Is chess a draw? Yes, but finding it is the real skill!

With perfect play, chess is a draw

Most chess players are familiar with the famous hypothesis: if both sides play perfectly, the starting position is a draw. Nobody has proven this. The game tree is too extensive. However, Grandmaster practice and engine analysis strongly suggest it is true.

For years, I treated this as a fascinating piece of trivia. Today, at 2518 ICCF, I understand something I didn't know at 2200. The "draw by perfect play" hypothesis is not trivial. It is the theoretical basis for every difficult defensive and offensive task I face in chess. If chess was a theoretical victory for White, defending as Black would be a slow death march. But, because chess is almost always a draw, every position I defend has a path back to equality - sometimes narrow, sometimes nearly invisible.

Why is it important now?

In one of my current correspondence games, I need to find a draw before my opponent wins. The position remained balanced for the first 12 moves. 

Then I moved g7-g6 instead of b7-b6, and my opponent's response caused the engine evaluation of my position to drop by 2.9, which is a significant amount. In a matter of seconds, I converted a theoretical draw into a near-certain loss. And I have the horrendous task of finding a continuation that leads to equality. The engine gives my opponent a clear winning path - if I cannot find the single defensive move that holds. I'd rather not post the actual positions in an unfinished chess game, but I will publish the game once it's completed.

Correspondence chess allows you to use every possible aid (chess books, chess databases, engines, and so on) to help you make moves. I don't always use computers to play correspondence chess. I understand that at 2500+ ICCF, this may be unusual. Most of my opponents have engines running alongside every move, analyzing at a level that I cannot match. But I've always believed that relying too heavily on the computer impairs your own chess judgment. Sometimes you have to believe your instincts.

Your instincts can sometimes betray you. In one natural-looking move, I transformed a drawn position into a nearly certain loss. And because this is correspondence chess, my opponent has several months to find the moves that will lead to a decisive victory. He will eventually find it. If I don't come up with a draw before he wins me.

That's the distinction between knowing chess is a draw and finding it when it counts. I was familiar with the theory. Instead of trusting the machine, I went with my gut. And I'm now paying the price.

What's coming next?

In the coming weeks, I'll talk about correspondence chess, the tools we have to make our moves, the search for the perfect chess move, and how correspondence chess can help us develop chess opening theory. Over the next few posts, I'll explain what I've learned about reading chess engine evaluations. I will guide you down the narrow path of determining the one move that holds a draw when everything else fails. I'll explain what happens when tablebases tell you the truth. When seven-piece endgames transform speculation into certainty. You'll find real games, real decisions, and annotated positions from my own ICCF experience. Much of the theoretical issues stem from discussions on a national chess forum and a lecture I gave on the pursuit of the perfect move in chess.

A hard-fought draw in over-the-board chess can feel like securing half a point. A hard-fought draw in correspondence chess feels like a small proof. Not a proof that chess is a draw - that remains a conjecture. But a proof that this opponent, this game, could not be won.

BTW, we have played one move each since the catastrophic incident. The engine evaluation remains unchanged. At least the situation has not gotten worse.

Follow the story

This game isn't over. Whether I find the draw or spiral into a crushing loss, I'll document everything - the engine checks, the late-night second-guessing, the moment my opponent finally strikes (or fails to).

If you want to follow along:

• Bookmark this blog and check back in a few weeks
• Reply or comment with your own correspondence chess disasters - misery loves company
• Send messages to me using the simple "Please Contact Me Here" - option in the right column

Once the game finishes, I'll publish the fully annotated game: the position I couldn't show you, the engine's cold judgment, and whether I managed to prove that -2.9 wasn't the end.

The story continues.