Over at the Chess Cafe (I'd provide a link to it, but Hanon Russell doesn't appear to believe in permalinks, so any link I'd provide here would break in short order, hence there's no point in doing so) Mark Dvoretsky has written an excellent piece on the place of opening preparation in the development of a chess player.

He starts out by noting that Botvinnik only lists any sort of chess preparation and training as one of four factors in chess success, and the place of opening prep falls farther from there.

In large part, the article is a continuation of a dispute with another Russian coach over training methods. Along the way he quotes heavily from Jonanthan Rowson's Chess For Zebras (and if you don't have that book, run, don't walk, to your favorite chess book supplier and get it. It gets my vote for chess book of the decade, not just the year). But it's what he says for himself I'll be quoting here.

Following this kind of thinking, chessplayers forget that the opening stage must sooner or later come to an end. Even if the outcome of that opening is favorable, sooner or later, we have to search, move by move, for the very best continuations, and solve one problem after another – positional or tactical, technical or psychological. And the way in which a chessplayer deals with these problems has a far greater influence on the outcome of the game than does the position he gets out of the opening duel. When all’s said and done, the one who makes the last error loses.

In many events – chiefly children’s and teenagers’ tournaments – I have observed the same picture, again and again. The players run through the opening stage “according to the last word of theory,” and sometimes even introduce their own novelties. But after a half hour to an hour, a great change occurs. The level of play declines sharply; there are extended periods of thought over elementary moves; positional or tactical errors follow one after another.

A point I've tried to stress to my students. I frequently come out of the opening badly in my games, only to win later as my opponent's book knowledge runs out and the outcome of the game begins to depend on chess skill, rather than memory.

It's hard to make that point, especially to kids, who see that they can use the new opening line they've just memorized in their very next game, while this endgame position I've put before them won't come up for weeks, if not years, and they know it.

So – what’s more important for a ten-year-old chessplayer? To study the games with 13 Nd1 in the database or to train his eye for combinations and his calculation of variations? To me, the answer is as obvious as I hope it is to you. Concentrating a youngster’s attentions on the openings, which must unavoidably come at the cost of other more important developmental problems (time, after all, is not unlimited!), means that a trainer will do a poor job with his student: he will disorient him.

And again:

Of course opening knowledge is necessary. But in the first place, only part of a student’s time should be devoted to it (the stronger and more experienced the player, the less time needed). And in the second place, a player should never become a slave to his opening knowledge. What’s important is to learn how to make use of it in order to resolve the problems arising over the board yourself.

Maybe they'll listen to him, if they don't listen to me.