From concrete to abstract and back.

Since I'm not an Idiot Savant (the Savant part is missing) I can only remember concrete topics when I have build an abstract framework to retrieve them from memory. By bringing all topics in the middlegame under the very same denominator piece activity, I have reached a pretty high level of abstraction. In order to make it practical, I must define a structure which is somewhat more concrete. A thoughtprocess for positional chess so to speak.

If you must decide to a move in the middlegame and there are no concrete tactics around, I use the following structure of events. I work backwards, from the target to the initial position.

The targets.
There are 3 possible targets after which you can go in the middlegame.

The king.
These are the preconditions for a king attack:

• The center must be secure. It may not collapse in a counterattack. Realize that when you sac a piece to open the kingside, your opponent is free to sac a piece back in the center to start a counter attack. If your center can not withstand this, a kingside attack will fail. That is why you often have to overprotect the center.
• Your opponent must not have serious counterplay at the queenside.
• You have to outnumber your opponent at the kingside. For instance because his pieces are deflected to the queenside. You need at average 3 extra pieces. 1 extra piece is needed to be sacced in order to open the kings position. 2 extra pieces are needed to mate the king. So you must be sure that when your attack peters out, you can manoeuvre easily pieces from the queenside to the kingside. Usually the queen is needed to deliver the actual mate.
  
• The enemy king must have no escaperoute to the queenside. That's why an attack on focal point g7 is stronger than an attack on focal point h7.
• In case of a pawnstorm there must be a "hook" in the form of an advanced pawn. Otherwise you can't open a file. It saves you a piece to sac.
  
• A wedge in the form of a pawn on e5 is welcome.
  

A weak pawn.
A pawn is weak when it can't be defended by another pawn without nasty consequences. It must be possible to attack the pawn.

• Induce a weakness.
• Fixate the weakness.
  
• Attack 2 weaknesses alternating.
• Attack where your opponent is weak, meaning where he has the least defending pieces.
• Solve your own weaknesses.
  

A passer.

• Genesis of a passer. There where you have a majority of pawns you can create a passer.
• Free the road to promotion from blockaders.
• If you can choose where to promote, choose the square with a color that is out of reach of an enemy bishop.
• Block enemy passers
  

The road to the targets.
This is where piece activity comes in.

• Mobilization or developement is the transportation of a piece from its initial square to its place at the frontline.
• Pawnplay is needed to open up lines into the enemy camp. If you move a pawn, you must have piece activity in mind. Special attention is needed for freeing moves. Besides for the opening of lines pawns can be used to gain space.
  
• Strongpoints or focal points or invasion squares all share the same idea: it is in the enemy camp, you can place a piece on it that can't be chased away by enemy pawns and you outnumber the defenders of that square.
• Attackers and defenders of a strongpoint often have to battle.
• Regrouping. When there appear weaknesses or when new lines become open, you often have to regroup your pieces as a consequence.
• When your opponent has an active piece, trade it off.

These are the positional ingredients I have found sofar. There are two ommisions I'm aware of:

• Exchanges. I haven't investigated this yet, but it is a very powerfull weapon.
• Leftovers of the middlegame, favourable for the endgame. This must be treated as a byproduct of moves that are aiming at something in the middlegame. Playing moves solely targeting the endgame is no good. Since I'm a novice to positional play, this is not the moment to treat them.

The coming time I will put this list to the test and refine it.