My last blunder happened rather late in a game, which I was winning. I had a great move planned, but the position in which it worked was not exactly the one on the board. In my mind, the opponent's queen was on another square...

I could easily have avoided this with a quick check on the board before making the move, asking myself if the move was safe. Besides, I really had a winning continuation I discarded because I think I was too focused on my solution, not paying enough attention to other options.

I have a problem with this in OTB. I used to write down my intended move before playing it, and in most of my games I ended up playing a different move at least once. Since they banned this method I haven't been able to find one that works. For three hours I remember to blundercheck every move, then I start to forget, and bang goes a piece. 

I can see that writing down your move in advance is rather like making a note of your analysis, and neither should be allowed unless both are. But why shouldn't both be allowed? In what way would this make it a worse game?

I meant in that game link, the 11. d3 perhaps White thought the c1-bishop was defended and then played the move, and then when it was opponent's turn, suddenly noticed the error. But in this particular case even if one touched the pawn, the pawn can only go to d3, so 11. d3 would have to be played if the d-pawn were touched.

13 years ago when I was in primary school and playing in a tournament for primary school students, I touched my queen and then realised that the only two squares that it could go to would leave it en prise. I only noticed this 1 second after touching the piece and raising it to play, so I resigned without moving. I suppose the issue in that game was that one of the squares I was planning to move it I thought was not being attacked, so perhaps long range piece control such as the bishop controlling a long diagonal one has to visually scan the diagonal to ensure that one does not place pieces en prise there. Unfortunately I still make this mistake not only in blitz/1-minute, but also in tournament games, also luckily quite rarely nowadays for the latter

In that case of the touch move, unfortunately moving the d-pawn and subsequently realising the blunder would probably be even worse . I suppose if one touches a piece and notices a blunder, maybe one can move the pieces somewhere else, but pawn-move blunders like 10. d3 probably I would just resign without moving if I touched the piece and missed the blunder.

I looked at that game. Unfortunately I have blundered pieces like that in tournament games as well. What usually transpires is I think I played a reasonable move, then 10 seconds later I realise that I made a serious error. In some cases I do not even see the error until my opponent plays the move (even if opponent takes a few minutes) and then I ask myself how I cannot see such obvious blunders.

In that specific game, if I made the same blunder, I would for some reason think that the queen on e2 or the rook on a1 is defending the c1-bishop.

This is a very individual thing. Leaving aside time trouble which is a different issue, some even quite strong players can have total lapses. Either they trust their "blundercheck" to work intuitively and suddenly it doesn't, or their hand simply makes the wrong move (typically one they intend to play a move or two later down the line they've calculated). 

I myself "suffer" from a harmless version of the latter: I sometimes write down the move I expect to be played next instead of the one that was just played (by me or my opponent)! This quirk doesn't cost me any points, just the occasional messed-up scoresheet 

Anyway, I think the minority who has a serious problem with this should adopt and train a version of Kotov's famous "Blumenfeld rule": Before making your move, look at the board "through the eyes of a patzer" and ask some simple questions. My favorite one is "How could this move possibly be a blunder"?, but you can also be more specific and ask "does my oppnent have any threats I haven't considered?", "will the piece really be safe there?", "did it do anything important where it stands?", etc.

The thing is to train a simple and very concrete routine, with some sort of "trigger" that should automatically evoke these thoughts. One logical trigger that is part of the process of moving: Train your hand reaching out towards a piece as a trigger for Blumenfeld's rule. It may annoy some opponents if you retract your hand often, but it's worth it to avoid blunders.

The standard way to train something like this is to just practice doing it consciously (in training games, perhaps) for as long as it takes until you eventually do it automatically (i.e the Blumenfeld question(s) pop into your mind whenever you reach for a piece).

There is also a somewhat effective "quick-fix" approach to making situations trigger responses semi-automatically without prolonged training, called Implementation Intentions (http://en.wikipedia.org/wiki/Implementation_intention). See the many papers by Peter Gollwitzer and colleagues, like:

http://kops.ub.uni-konstanz.de/bitstream/handle/urn:nbn:de:bsz:352-opus-55590/99...

http://kops.ub.uni-konstanz.de/bitstream/handle/urn:nbn:de:bsz:352-opus-57036/07...

I have used this method of "if-then-planning" a bit away from the chess board. It works some of the time - at least worth trying for someone trying to establish a good habit or get rid of a bad one.

Of course, those who virtually never blunder like this shouldn't bother with Blumenfeld's rule at all, since it's meant to solve a problem they don't have!

I think that good players need to take a lot of shortcuts and do a lot of pruning to play their best. Otherwise they might get bogged down with checking and rechecking almost every move. So from that perspective making a terrible oversight once every twenty games mightn't be so bad. You have to take a little risk when calculating, and have a little faith in your analysis. I guess it's just fingers crossed you haven't made a terrible mistake. 

I find it hard to imagine a really good player leaving a piece en prise after four minutes of looking at the position though. I mean it has happened but you would probably be more likely to get involved in a car accident on the way to the game. Usually the player is looking at one variation and then suddenly changes his mind at the last second and plays this other move without checking it properly at all (Kotov syndrome). 
   

In my last regular OTB game, it happened to me that I castled short, just allowing my opponent to fork my queen and rook (and to win a pawn in addition), after which I resigned. After that frustrating experience I came up with the idea of giving me a "receipt" before each move that I've checked all those simple tactics. With that I mean just making a note on a real or imaginery check list: "Checked all simple tactics? - Yes / No."

Does this help? My impression by playing against a computer (with physically giving me that receipt) is positive, but as in serious games one is not allowed to record anything except the moves, this practice might run into problems (of course, one can "cheat" by just putting a dot on the scoresheet, which afterwards [after recording the move or the thinking time] becomes part of the notation, or by using a ball pen with a push button - then just "activating" it is the receipt). (Could one consider this as "cheating"? I mean, one doesn't "record" anything!? What do you think?)

Best regards,

Zwischenzugzwang