After playing a few club games, I realized how frequently queen endgames tend to arise with my style of play. However, this is a major weak spot for me, as I have very little theoretical knowledge to rely on. My opponents seem to know even less, so we typically shake hands on it.

However, this lack of theory will eventually catch up to me, and I need to be educated. So, with that in mind, I ask the following questions:

 

 

 

 

 

 

 

What are the defending ideas for a position like that? Should the weaker side trade into a position like that following the general rule of trading pawns when one is down, or should he maintain pawns on both sides of the board where it may be possible to stop the stronger side from creating a passed pawn?

The position you have posted looks like an explosive win for white.  See Silman's Complete Endgame Course for a better explanation than mine.  (Silman puts this info in the 2200-2399 rating section.) 

Yes it is possible for white to win.  White needs to make a housekeeping move such as b3 to guard against back rank mate.  Then with the king on b2, the queen needs to be stationed so it shelters the king from checks and protects the advancing pawn.

Queen endings are special because blockade is not possible.  If the black Q ever stands in front of the pawn, the white Q will just pull up along side, protected diagonally by her own pawn, threatening the black Q and forcing her to move away.

The defender should trade as many pawns as possible.

Tony and Ghost explained great, even for me haha, but I think IF your opponent is dumb enoguh to trade queens DO IT, and you will  have a big chance of promoting the g pawn.

All of what ghostofmaroczy said is right (or, at least I agree) of course tony too -- I also do an overview of it, but if it all sounds too involved then Tony gave the best advice, just push the pawn and shield your king from checks and you should win

The main feature of having that extra passed pawn is that a queen can't blockade it, the queen and pawn vs a lone queen can push down the board by themselves -- the black king would like to help stop it, but he becomes very exposed to checks.

A dream position for white in this is to play g4-g5 Qf6 b3 and Kb2.  Putting the king and queen on the long diagonal (where the king is shielded from checks) is a common pattern in this endgame.  After pushing the pawn pretty far down the board, the queen can leave the diagonal for the final push -- this is when you have to calculate and make sure the enemy checks will run out.  Another bit of knowledge, generally giving  checks on a diagonal are more effective in this endgame than rank and file.

As for the defense, basically black is screwed :) but if he could magically trade off all the queenside pawns he would have some drawing chances -- especially if his king ended up in front of white's remaining passer.  If he ends up behind, he likes to hide in the corner and try for a perpetual -- the further away from the passer the better because if white can block a check with his queen and give check himself, the queens come off and the game is lost -- therefore the king likes to be in the opposite corner from the action so the chances of such a fork are less possible.

Thank you all for the terrific information. The information about putting the king on the long diagonal and the queen on this diagonal as well turned out very useful. I had an inkling that a position like this would yield a win, but how about on like the following?

Please ignore the specific pawn structure, I'm interested in tips for when each side has a majority on one side of the board. What should the plan be? I believe a huge percentage of these are theoretically drawn, but what would the ideas be for each side?

A recently played game, all commentary is welcome. Specifically looking for info on whether trading rooks was a good idea or bad, and where I could have played the position better from there.

---

I think the considerations go in order of importance 1. King safety 2. Queen activity 3. Weak pawns  Pawn majorities are nice, but like you said many of these games are drawn.  For there to be a win a player would have to have a large advantage in one of the three listed things and only after try to create a passer with a majority.

King safety is an obvious one, especially because your opponent may at the very least be able to get a perpetual check for a draw.  The activity of the queen was a new idea for me when I learned it (I thought she was always very active) -- she likes to control or occupy the middle of the board.  If your king is safe get your queen active!

Last look for pawns that could be targets.  Other than looking at the middle of the board concerning the queen, if you can tie your opponent's queen to the defense of a weak pawn (or group of pawns) that's the best, and a very sizable advantage if it's tied on a square away from the center.

In the position you give, based on king safety alone I'd say white has a very small advantage -- the fact that black's f pawn is missing from f7 is a big factor concerning his king's safety, but like you said these are most often draws.  King safety being #1 white would like to be able to loosen black's king more, breaking his pawn shield up with moves like h4-h5.  Also simply grabbing the center with Qe4 is a very good move too.

White would love to occupy black's vulnerable 2nd rank with the queen.  White's position is more solid but his majority can't create a passer (his pawns can't leave his king) so white is basically very slightly better but can't win unless black makes a mistake concerning his king safety or queen activity at which point white may be able to win some Q-side pawns.  Black should just stay compact and defended, his queen can't leave his 2nd rank open (again with the king safety), so can't really support his majority -- without a blunder I'd guess like you said this is drawn.

Hmm, as for the game I think you have an advantage either way you capture, not only do you have the passed pawns, but you were correct thinking the knight is superior to his bishop.  Confined to a fight on the Q-side the knight is better + his pawns on dark squares don't favor a dark square bishop + all your pawns are on light squares so if there's not an attack using those holes you're minor piece is superior hands down (and it is).

So your minor piece is better + you have passed pawns -- this mean every time you trade off a heavy piece you're getting closer to a technically won endgame.  So trading rooks was a good idea.

My impression on the R/R vs Q/Q endgame is getting the queens off was better.  Not only is your king safer from all those dark square weaknesses, but in the rook endgame your rook was so much more active vs it's counterpart + the passers, I agree this was a winning advantage.  The line ending 35. exd4 black simply plays Rc4 and marches his king to d5.  For that reason 33.Bd4 in that same line is a mistake for white, the bishop + rook were about as well coordinated as white could want covering the c file and blockading on b2, instead white needs to run his king over to the Q-side to help.

35. Qc5 does look better you ask what he'd play after this?  Well his position is worse, I'd go for Qe4 with g4/h4-h5 type of ideas in a desperate bid to not lose hehe.  White can't defend passively on the Q-side, so has to drum something up against black's king.

Thinking of d5 as a permanent home for you knight was a mistake :) I like the look of 36...Nc6 , white is forced to move his bishop away and you can start to go to work advancing those pawns -- also it blocks the c-file from infiltration by white's queen.

I think 38...b3 is a strong move -- so what about e4 attacking your knight continue with b2 :) after Bxb2 Qxb2 he takes your knight but you have Qa1+ and now you're playing Qxe5 with check and winning that Q endgame.

41...Nb5 is a pretty move I like how the bishop is helpless after that, good find.

Like you said 43. Kf1 loses immediately, but even after 43.Bb2 white has to prove there is a draw here -- his problem is all his light square holes on his K-side just begging the black king to run through the formation and gobble up pawns :) meanwhile the bishop can never be an active piece, all black's pawns on light squares and all white's pawns on dark squares is the fantasy position for such a N vs B endgame.

Thank you for info on both topics! The first post has been examined and will be reread seval times later to make sure I understand what you're trying to say. I have a terrible habbit of reading something and having no idea what it said.

To the second post, Thank you VERY much for your corrections of my analysis. Perhaps next time I'll use an engine afterwords. I agree 33. Bd4 (in my first line) would have been a mistake after farther review.

I'm glad you think Qc5 was better, I just didn't find it with 5 minutes left on my clock.

Thanks for the correction about 38...b3, I missed the obvious b2 push.

Could you help me out with the 39. e4 line though?

I didn't mean to be so long winded in the first post :)  I'm basically saying to compare the difference in king safety first, then look at each side's queen -- is she active/in the middle of the board, then last look for pawns that are potential targets.  Of course if you can win a few pawns or immediately create a passer do that first, these are just the general considerations for this endgame when the position looks equal.  And that's how I'd come up with the moves h4 followed by h5 (hurting king safety of opponent) or Qe4 (helping queen activity for white).

Oh I see that e4 line now, I somehow missed it before.  At the end of it anyway just like before push the pawn and then have the queen escort it without getting your king into a perpetual check.

Considering that fork to trade the bishop off maybe white would rather play 40 Be3 because then he really does threaten a perpetual for example after Qd8+ white can't block with Qe8 like before because you have Bh3+ -- also black doesn't need to play 39...Nc3, he can win the pawn that's attacking his knight with Qc1+ Kh2 Qf4+ and Qxe4.

With the passers black should be winning in any case as long as he doesn't slip up... but of course during a live game that's far from easy... I didn't realize this was a live game, thought it was on the turn based side -- I was wondering you found the Qc5 and others so easily why you didn't just play them instead :) but of course it's not so easy when the clock is ticking away!

It's impressive you're willing to look at one of your won games so hard, many times the only thing I remember about my wins is the very end where I won

---

Some nice posts in this thread, orangehonda. Two things to look out for if you are the inferior side in such an ending.

a) watch out for possibilities for forcing a perpetual. Checking is annoying :D

And especially

b) Passed pawns are absolute gold in queen endings, because one passer and the queen can escort it home. So it can often be worth shedding a few pawns to get a moving passed pawn to help you draw the game. No matter how many pawns up the superior side are, they can't let you queen.

That's good advice marvellosity -- passers are always great to have in this type of endgame, in fact it was partly that bit of wisdom, and for taking part in this thread, that made the following endgame really stand out to me.  I happened to be playing through some games from the 1953 Zurich tournament and came across it.  It's interesting because optically it seems like white should be dead lost with his king in the center and black's passed pawn.

In fact Bronstein's annotation after 33. Qf2 was: 
"A quick glance at this position might leave one wondering how even Keres, with all his skill in queen endings, could save this game.  However, there are two peculiarities in this position which ease white's task a bit: the black king's scanty cover, which means White is always threatening a perpetual check, and the solid position of White's queenside pawns -- after all, they are three against two!  The meaning of this latter circumstance will soon become clear."

Taking white in this position at my club against an expert we also couldn't find a way for black to win -- an instructive drawn position for sure.  I include Bronstein's annotations in the game below.

 

---

I recently posted in another endgame forum topic, starting out , where chessmaster102 was asking for some different endgame training exercises, I give a few.  As for finding examples of specific types of endgames, I'm not sure how.  Chessbase may have a way to search for games by endgame theme, someone familiar with chessbase would have to answer that for you though.

Here's me playing against myself, black should easily win in any case as long as he's careful to make sure white's checks will run out.  These may not be the best moves, but this is generally what it would look like.

[Edit.  Ok, there are already mistakes I can see :) but if you want it played perfect slap it in an engine, the point is black has a crushing advantage ]