Unfounded self-confidence (or any unfounded belief) is very harmful.

Citation needed. Bluff (i.e. unfounded confidence) seems to be a very efficient strategy in many games. Apparently, even in chess:

UNIDENTIFIED MALE #2: Rook to D1.

CAMPBELL: And this particular move was really bad, and so it caused us to give up the game right away.

FOO: This really bad move confused Kasparov. Murray says he heard Kasparov's team stayed up that night trying to analyze the logic behind that move - what it meant. The only thing was - there was no logic.

Self-confidence will only help you in games where other factors already favor you.

A charitable reading of the article would be that the other factors are already okay.

Maybe not good enough to guarantee a 100% victory, but let's say they give a 20% chance of victory (but only if the writer believes in their victory, otherwise they cannot find enough motivation to train properly), the gains of victory are huge, and the costs of training are relatively low. So it could be instrumentally useful to believe that victory is almost sure. (Of course, with all the caveats and proper compartmentalization. For example, author shouldn't make bets about their victory. Only to believe in the victory while training.)

Self-confidence also has beneficial social effects, since it signals that you are capable of winning. As a result, people tend to be more self-confident than can be rationally justified, and they're also built to counteract those effects. For example, they are risk-averse. If you are only as confident as is rationally justified, but you are still risk averse etc., you will avoid taking actions that would benefit you.

I already know this. But you weren't supposed to mention it. I'd appreciate it if you'd delete this comment.