This installment uses an illustrative game to explore how to bring about favorable resolution of pawn tension in the middlegame.
A recently concluded Red Hot Pawn game between IBeatScurvey (1556) and shouresh (1659) (Game 14698832) began as depicted in the following chess movie (which I've lightly annotated).
The above chess movie (which ended with 18. a4) reached the position
To begin formulating a middlegame plan, we first focus on the pawn structure:
Does Black own any half-open files? No, because none of Black's pawns has been captured or left its original file.
Does Black own a mobile pawn majority, that is, one that could advance to create a passed pawn if neither side's pieces intervened? To answer this, let's consider how Black might support the advance of the e-pawn, which would be the "candidate" (that is, the pawn that doesn't face an enemy pawn on that file and therefore could in principle advance further than its peers). Black's f-pawn can't support the advance of Black's e-pawn, because the former would be captured by White's g-pawn once it reached the f5-square. Therefore, with regard to the ability to autonomously form a passed pawn, this pawn formation would be equivalent to
If Black were to play ...d5 and ...e4, White's f-pawn would prevent Black's e-pawn from safely advancing further. From this discussion, we can conclude that Black does not own a mobile pawn majority.
Here again is the position reached by 18. a4:
Returning to the pawn structure,
is there a file on which Black has more space than White does? Yes, the c-file, on which Black has two ranks before encountering a pawn, whereas White has only one rank before encountering a pawn. But White could acquire just as much space by advancing his own c-pawn to the third rank, because the two sides' c-pawns are not yet in contact with each other. So, Black's space advantage on the c-file doesn't justify trying to create pawn tension in that sector.
The remaining possibility for creating pawn tension would be where the two sides' respective pawns on a file obstruct each other. That would be on the g-file, on which each side's pawn is blocked by the enemy's pawn. Black could create pawn tension there by playing ...f5 or ...h5. For each of these pawn breaks, we'll need to consider the consequences of either side's initiating a pawn trade.
BLACK CONSIDERS ...f5
The move ...f5 would bring about the following pawn structure:
White Answers ...f5 by gxf5...
If White played gxf5... and Black recaptured (necessarily with a piece), this would result in the following pawn structure:
This pawn trade gives Black a half-open f-file against an isolated White pawn, whereas the Black e- and g-pawns, although also on half-open files, are protected by pawns. Black would have two pawn islands instead of one, but White would have three pawn islands instead of two.
In the resulting full position after gxf5...,
Black would recapture at f5 with the bishop, as keeping the queens on the board increases Black's pressure against the h-pawn and aids Black's prospects for exploiting White's more exposed king.
White Retains the Tension after ...f5
Supposing White refrained from playing gxf5..., how should Black handle the pawn tension?
Black plays ...f5 followed by ...fxg4
If Black played ...fxg4, this would bring about the following position:
Here, Qxg4... would transpose (after Black plays ...Bf5) into the line where White trades pawns at f5. But hxg4..., would allow White to meet ...h5 (to further open the kingside) by the surprisingly annoying Bh3... (see next diagram).
White threatens gxh5... to pin Black's queen. Moving the Black king or queen off the h3/c8 diagonal would still allow White to discover on whichever Black piece remains on that diagonal. However, Black has ...Rg8, intending to meet gxh5... by ...g4! (see next diagram).
Now Bxg4... would lose a piece after ...Bxh5; Qxg4... would lose the queen after ...Bf5; hxg6... would expose White's king further after ...Rxg6 and taking the bishop next move.
But these tactical pyrotechnics became necessary only because Black exchanged pawns at g4, which made the h3-square available to White's bishop. Black would be advised to instead retain the pawn tension until Black could benefit from releasing it.
Here is the full position after the contemplated 18...f5.
Assuming that White "passed," how might Black make progress?
Black plays ...f5 followed by ...h5
We've seen that after playing ...f5, White's replying gxf5... would invite ...Bxf5, increasing Black's pressure along the h3/c8 diagonal. How might White be induced to initiate a pawn trade? One way would be if Black threatened to win material by taking at g4, such as if Black were to also advance the h-pawn. If White doesn't deploy a third unit to control the g4-square, then Black's ...h5 would result in a position like
where White would either have to trade pawns or lose the g-pawn. Playing 20. gxh5 would open the h-file against his own king and would give White three pawn islands to Black's one. Counterattacking by 20. Qe3 would fail as shown in the next chess movie.
If after 18...f5, White were to retreat the queen to e2 (to clear the f3-square for a third defender of the g4-square), the continuation might be as follows:
BLACK CONSIDERS ...h5
The move ...h5 (instead of ...f5) would bring about the following pawn structure:
For reasons similar to those examined after ...f5, White should strive to retain the pawn tension after ...h5. Then Black would probably play ...f5, transposing to the position examined earlier where White was compelled to exchange the g-pawn or lose a pawn.
So, it would seem that either 18...f5 or 18...h5 would have helped realize Black's material advantage.
In the position after 18. a4,
Black actually played 18...d5 and followed it by 19...Bb4. But this "active" development of the bishop accomplished nothing. Black ought to have focused instead on how to open a file for the rooks, as we've just seen.
(A list of the threads I've initiated at this forum is available at http://www.davidlevinchess.com/chess/RHP_my_threads.htm .)