I was taught, many years ago, that if any pawn of yours gets to the enemy’s back row, it automatically becomes a Queen (even if you still have your original Queen). Recently I’ve heard that the correct rule is that you get to choose what piece your pawn becomes.
Is this so? And if so, why in the world wouldn’t you choose to get a new Queen?
The only scenario in which you wouldn’t take a Queen is if a knight would be more beneficial. For example that knight might put the king in checkmate or fork another piece. Its a pretty rare occurance but its concievable that it might occur.
It’s possible that someone may need a knight more than a Queen at that point in the game. Knights are the only pieces that can jump over other pieces.
There are situations where you can force mate by not selecting a queen.
Check this site out .
Bolding added - and: not so. Check out my link. You might want a rook.
There is, as shown in my link, another situation - where you might want a rook, so as not to cause a stalemate.
As I would understand it, there are very few reasons why you wouldn’t. However:
Promotion to a knight sometimes gives a clear mate in a few moves, just because of the tactical sitaution. Because a queen cannot move as a knight can, there are some situations where the win would be less clear, at least, in those cases. (Though since a queen is generally a powerful and versatile piece, you’d probably have the edge anyway, still.)
IIRC there are some situations where it’s valuable to promote into a rook or a bishop, simply because those cannot be forced to move in certain directions the way a queen could. I might be misremembering that.
Hi, Opal!!
I realize others have already posted to this better, (not quite preview, just surfing to the thread in another window to make sure I hadn’t lost the board,) but I’ll send this along anyway, partly because I so rarely get to do an opal list. 
(looks at peter’s link.)
Ahh, that’s probably the example I was thinking of with rook (or conceivably bishop,) the stalemate situation. Because of the extra spaces a queen attacks, she might stalemate the other side, while a less powerful piece would give you the opportunity to press home a checkmate.
By the way, do any of the bright chess people here know where the ‘stalemate automatic draw’ rule came from?? It seems a little unusual to me… it seems like the stalemater is usually in a better strategic position than the stalemated, except for double stalemate. Why decide that it’s a tie??
I don’t know if there’s a definitive answer to this question, except that in order to win you must secure a specific objective: capture the king. If you fail to do that, you have not won - no matter what overwhelming power you have created.
I like the rule - but only because my brother is better than I am at chess, and I would almost always find a way to stalemate him.
After a bit of googling, I turned up this link, which claims that “stalemate was not considered a draw in England before the early 1800’s; before then, the stalemating player lost the game!”
If this is true (and I have no real reason to doubt it) then we should be able to find something written out there about why they created the stalemate = draw rule. My point being, since the change happened so recently, there may indeed be a discoverable answer to this question.
Wikipedia has an interesting link to this site.
:smack:
Forgot to add my conclusion: The practice of stalemate = draw derives from gambling. If you can’t outright “win” then you can’t take the other guy’s money.
Back when my father was teaching me chess by beating the crap out of me every time we played, he would from time to time deliberately not queen the pawn, but choose a bishop, rook or knight, to show me how to mate without a queen. At that point he could usually mate me in three moves with anything.
Of course in real life, it really only makes sense if you go for the knight.
Or the rook, see link in post #4.
A Knight? In 25 years of tournament play I have never seen someone promote to a knight. While most go for the Queen, I have seen the occasional Rook… but late in the game the knight is the worst piece to have. Its debatable if the promoting pawn is worth more than the knight. Unless promoting to a knight would cause instant checkmate, you should either promote to another piece or not promote the pawn. A pawn on the 7th rank is generally more of a threat than the benefit of having a knight.
Some things to think about:
Promoting a passed Pawn is something that normally only happens in the “end” game. “End” games normally are after most of the major exchanges of pieces, including the loss of many of the pawns…and in this case the opposing Pawn{s} that faced the promoting Pawn. You are probably 35+ moves into the game and if in a tournament, probably short on time. Generally, because of the lack of material, the opposing King is helping to stop the promoting Pawn. As most players would sacrifice to stop a promoting Pawn, often your opponent may be left with just a King. In this case you may want to go with the Rook to avoid a chance stalemate. Most players have experience with checkmating with a Rook; few players regularly are “up” a Queen on an opponent. In most games, one opponent will force an exchange of Queens well before the “end” game. Its been my experience that most players that have a Queen in the “end” play it as either an ad-hoc Rook or an ad-hoc Bishop. Because of lack of experience, you may want to promote to something you are more familiar with. This is especially true if you are trying to play under time control.
The Knight is a powerful piece in the “opening” game and still strong during the “middle” game. But as the game develops, the other major pieces become far more powerful because of their mobility. It takes Knights four turns to cross the board while Bishops, Rooks, and Queens can do it in one. As most of the action in the “end” game happens on the rim of the board, you fall under the “Knight on the rim is grim” prophecy.
A King and a Queen can checkmate.
A King and a Rook can checkmate.
A King and two Bishops can checkmate.
A King, a Bishop, and a Knight can checkmate with difficulty.
A King and two Knights cannot force a mate.
Neither a Knight nor Bishop with the help of a King can checkmate.
So promoting to a Knight requires another piece for mate, and if you have a Queen or a Rook, why are you promoting? If your opponent has other pieces… promotion should be to a Queen automatically. If stalemate is possible then you should promote to, at the very least, the highest value piece…Rook…then Bishop…then Knight. There is no stalemate position that will be possible by both Rook and Bishop.
So in summary, if promoting to Knight isn’t instant checkmate…why would you ever think of a Knight?
Footnote: In around the house games just for pure fun, I never promote to a second Queen, I promote to whatever is the next piece in line that is missing. It would be a rare day that I am promoting a pawn with a Queen, and all my Rooks and Bishops on the board. 
What the OP’s describing (automatically going for a new Queen) is called overpromotion. Depending on the position at the moment, going for a new Queen may not be, as a few posters have mentioned, the best choice.
BTW, “double stalemate?” How can one have that? Western Chess doesn’t permit one to pass his move.
You wouldn’t. Nobody ever said that promoting to a knight (or a bishop or rook, for that matter) is common. But it is at least conceivable that a knight might be an instant checkmate, where a queen wouldn’t be. So one might at least conceivably promote to a knight.
As for stalemate, you’d only want to underpromote if it was an instant stalemate. Any player with any skill at all can avoid stalemating with a queen he’s free to move, and the mate with king and queen is even easier than that with rook and king. But again, there are conceivable situations where promotion to a queen would be instant stalemate, so it’s conceivable that one might want to underpromote to rook or even bishop (though it’d likely be better to just wait a move before promoting, rather than promoting as low as a bishop).
And by “double stalemate”, I suspect that chrisk was referring to “draw by insufficient power”, as, for instance, if both sides are reduced to just a king. It’s not a stalemate, but I can see how a person might think that “stalemate” and “draw” are interchangeable.
Depends on what you mean by instant… I’d promote to knight as long as I could get INEVITABLE checkmate. An inevitable mate in nine moves isn’t instant, but I’d take it if I couldn’t see inevitable mate with a queen.
And, I’m sure I’ve heard the term ‘double stalemate’ somewhere, maybe in some obscure chess column. Of course, it doesn’t make any practical difference, since it would just be like any other stalemate if neither side can legally move. I think it’s possible, at least theoretically.
In that link, they show a diagram where promoting to queen results in a draw, but promoting to rook lets you win. Can you explain? Given that queens can do everything rooks can do, how is this the case?
In those diagrams when the queen is promoted there are now no legal moves for the opposing king but it is not in check. Given that the king is the only piece that player has it is a stalemate. Stalemate has been considered a draw since about 18 something according to other posts.
In Chess, one must make a move when it is his turn to make a move. There’s no passing. If that player has no legal move and is not currently in check, then the game is a stalemate (also known as a draw: in tournament play a loss = 0, a win = 1, and a draw or stalemate gives each player 1/2 point). If one promotes a pawn to the wrong choice for the position, then it could happen that the other player does not have a legal move.
Remember also that Queens don’t just “do everything that Rooks do.” They have the combined powers of a Rook and a Bishop. That means that the Queen covers more squares than a Rook does, thus denying the opposing King more squares than a Rook would.
BTW, the reason I specified Western Chess in regards to the “no passing” rule is that Korean Chess does permit a player, in certain situations, to pass his move.