Michael writes:

I was going over the Lucena position with one of my students last night. This was the position:

I asked him what he would play and he moved the rook to a1. I said, "Well, the correct move is Rg1+ to push the black king away, but let's see why your move doesn't work." However, I quickly realized that after my student's next move, 2.Ra8, there was nothing I could do to stop 3.Rd8 and White eventually walks his king out and wins!

While Speelman's BCE mentions that White has two ways of winning from the starting position after 1.Rg1+ Kh7 and now either the usual building the bridge with 2.Rg4, or 2.Ra1, I have found no book that mentions my student's immediate 1.Ra1 which also appears to win.

Even Junior 10 can't hold the draw after 1.Ra1! Am I missing something here? And if 1.Ra1 does work, than why all the fuss about having to know Lucena?

SILMAN REPLIES:

Yes, of course Ra1-a8-d8 wins -- actually, 1.Rg1+ Kh2 2.Ra1 is even more accurate. Why not kick the King away first (It's good to stress this)? However, you hit on a question of form being taught over a functional alternative. In fact, you've stumbled across ONE OF THE SECRET TEACHING TRICKS OF TITLED PLAYERS!!! Let me explain:

First, check out diagram two

In this case of King and pawn vs. lone King, teachers always tell the defender that they will draw if they move their King straight back (when it must retreat). In this case 1...Kc6 is the move. However, you can also move your King to the side (...Kb6 or ...Kd6) and still draw easily. Later you WILL have to move straight back, but at the moment the straight back rule isn t a must.

The same can be said for most Lucena cases and other positions as well. So why is a big deal made of these things? There's a GOOD REASON!

When you teach, you want to give rules that stick in the player's mind, and you would like to avoid saying, "Here you have two ways to win. Just pick whatever you like or memorize both." Though true, it's another (unnecessary) thing to learn, and tends to confuse more than help.

Thus, in the case of the King and pawn endgame, it's just good form to always move the King straight back. You don't have to, but you might forget to do so when it matters most. By sticking to the proper form, you protect yourself from getting lazy and screwing things up.

In the case of the Lucena, your student's method works fine if the pawn is a Bishop-pawn or center pawn, but you MUST use the Lucena position to win if it's a Knight-pawn (see the next diagram).

Since he HAS to know the Lucena anyway, why not make that his method of choice (with Knight-pawn, Bishop-pawn, and center pawn)?

When one teaches, it's good to keep things simple. Lie if you must (it's for their own good), make copious use of plausible deniability, and herd them in the direction you want them to take. This not only helps the chess student, but also trains the teacher for an eventual career in politics.

This whole thing is probably a moot point since 99% of students won't notice the alternative method. If one does, and if he insists on making use of his solution (thereby challenging your authority), just say, "Yes, very good. That also wins. But I want you to learn the Lucena. If you don't, I'll kick your ass to pulp, triple my rates, and make you promise me your firstborn in recompense for the lack of respect you've shown."

Threats like this are very effective teaching tools. But that's another matter for another time.