King and pawn vs. king endgame

Everyone is talking about Zugzwang.  The best form of it is reciprocal zugzwang, where it comes to the detriment of whoever is to move.  There are many chess positions where whoever is to move wins.  But reciprocal zugzwang comes in 2 forms:

A) If the side defending is to move, he loses, but if the side with the advantage, like an extra pawn, is to move, it is only a draw.

The easiest example is the following:

If Black is to move, White wins, but if White is to move, it is only a draw as Black has the opposition.  Move the White pawn back a square and White wins no matter who is to move - he can lose a tempo by moving the pawn if it is his move.

The other type of reciprocal zugzwang is the more severe case.  Whoever is to move loses.  The following is the simplest and most well known case:

The game is lost for whoever is on move.

Many, but not the norm. in most positions one particular player is winning or it's a draw whoever has the move. 

How can you have a zugzwang that's not reciprocal, by the way? Seems to me "reciprocal" is redundant.

Most often only one player is in zugzwang. Mutual zugzwang is when the player on move loses, or is now playing for a draw while the other has slight winning chances. 

@Ziryab

But if for W, say, having the move is a points disadvantage for W compared with W not having the move, then it must mean that W having the move is a points advantage for B compared with W not having the move and therefore W not having the move, meaning B having the move is a points disadvantage for B - that is to say B is also in zugzwang. Similarly with W and B interchanged.  

Of course only one player has the move in a game so zugzwangs don't actually occur in the game, only as hypothetical possibilities. They're features of a diagram rather than a position (the latter usually being taken to include side having the move).

It might be worth pointing out that zugzwang is a feature of more than just the diagram if FIDE competition rules are in effect. Those rules include the 50 move and 3-fold repetition rules.

So for example @ThrillerFan's first example in #61 is a zugzwang under FIDE basic rules (assuming nobody has resigned or agreed a draw), where the aforementioned rules don't apply, but is not a zugzwang under FIDE competition rules if the ply count is 97 or higher. Similarly for the second if the ply count is 95 or higher.

For a more practical example this diagram with ply count 0 is a zugzwang under FIDE competition rules but not under FIDE basic rules. 

Under FIDE competition rules White to play draws and Black to play loses, but under FIDE basic rules White wins either way.

The same diagram with ply count 2 or higher is not a zugzwang under either set of rules. (It's then a White win under basic rules whoever moves and a draw under competition rules whoever moves.) 

Similar considerations apply with previous positions that are considered the same for FIDE art. 9.2.2 (triple repetition) or with a combination of that and the ply count.

Not all zugzwangs are reciprocal.

The following is a prime example.

If it is White to move, he can simply move the Bishop.

If it is Black to move, he is in zugzwang.  If he could pass, this would be a draw.  But he cannot.  He must go ...Kf7 and he loses.

Only 1 side can be put in zugzwang here, namely Black.  This is NOT reciprocal zugzwang because it would not be to White's detriment if he is on move.  Just move the Bishop and White still wins.

Reciprocal Zugzwang SPECIFICALLY means it is to the detriment of whoever is on move in the position, meaning it must be to both White's detriment AND Black's detriment to be on the move for it to be Reciprocal Zugzwang.  If it is only to one side's detriment, it's just plain old zugzwang.

And note that detriment could be going from a win to a draw, a win to a loss, or a draw to a loss, compared to passing.

Zugzwang is a tactic, not a rule.  All zugzwang means is that the player on move must worsen his position, as in if he could pass, he would be OK, but because he must move, his position worsens.

In the example with the 2 knights and pawn, if either side's clock falls, the other wins as there is a legal sequence of moves that leads to make for either side.  It does not have to be forced.

USCF, this would be a win for either side as well, but change the Black pawn to say, a Rook, and only Black can win on time.  King and Two Knights requires at least 1 pawn of the opposite color present for it to be ruled a win, regardless of whether the pawn is behind or in front of the draw line, does not matter - the existence of a pawn of the opposite color is what matters with K+2N in USCF.  Again, FIDE it would be a win for White even without a Black Pawn if Black's flag falls.

Zugzwang does occur in games.

Not that long ago, over the board, I had White King on f6, White pawn on e7, Black King on e8 - a reciprocal zugzwang because if it is White to move, it is only a draw, but if it is Black to move, Black Loses.  The result is worse for both players if it is their move.

I was White and Black was to move, and I soon won.

That depends on how you define zugzwang.

I define it as a position with an unspecified side to move where if a side to move is specified that side is in a worse theoretical position than if the other side to move were specified.

You are defining it as a position with a specified side to move where the side to move would be better off if he could pass, but in that case if the other side were also to pass he wouldn't be better off. 

You say

If it is Black to move, he is in zugzwang.  If he could pass, this would be a draw.  But he cannot. He must go ...Kf7 and he loses.

but that is true only if he could pass and carry on passing indefinitely. In that case would you say Black is in zugzwang here (under basic rules, say)

If he's allowed to pass at will, it's a draw. If not he loses. (If he's only allowed to pass once he still loses, as is the case in your KBPvK position in #66.)

With my understanding of zugzwang your position would not be zugzwang, merely a White win whoever has the move. But It's possible my understanding could be adrift from the norm; I have seen the term used differently.

The definition makes a difference of course and, granted, with your definition not all zugzwangs are reciprocal.

There is the standard definition that chess players use, and there is a personal definition unrelated to how others use the term. Which should be preferred?

Which is which?

According to Wikipedia

Zugzwang (German for "compulsion to move", pronounced [ˈtsuːktsvaŋ]) is a situation found in chess and other turn-based games wherein one player is put at a disadvantage because of their obligation to make a move; a player is said to be "in zugzwang" when any legal move will worsen their position.

which doesn't specify whether passing or switching sides is envisaged.

It's also total nonsense because if all legal moves leave you with at best a particular outcome that's also the best outcome of the position before you make any of the moves, so it's impossible that all legal moves would worsen a position.

What I meant by zugzwangs not occurring in games is that only one side at a time has the move at any given point. Nobody is actually allowed to pass. Of course zugzwang diagrams can occur in games, but then the zugzwang is already too late to be of any importance because the side to move is already fixed and passing is not legal. The importance is that if such a diagram is likely to occur a player should try to avoid reaching it with the move.

The difference between our understanding of the term zugzwang appears to boil down to whether the person with the move can do better if he's allowed a single pass or he's allowed pass indefinitely.

A single pass is actually exactly the same as starting from the same diagram but with the opposite player to move, which is my understanding of the term. Indeed your sentence that I've highlighted in yellow appears to say exactly what I said of the the two knights v pawn diagram I posted.

If you're comparing the actual position with what would be the case were the player to move allowed to pass not just on that move but also in subsequent play, that makes a difference, but then very many diagrams could be considered zugzwang e.g. the rook and pawn diagram I posted in #69. (You haven't said whether you consider that a zugzwang.)

i understand

Ez