Skype coaching session, 19 August 2018. 
The game (59 moves) given below with annotations and without annotations was discussed. 

Seems like a long while since written notes to the sessions have been posted. 
There is quite a backlog to be processed, hopefully these will be published in time, though unfortunately not in the correct sequence. 

Re-structuring of the sessions now offers the option for guests to bring their own games for group discussion. 
Recordings of some previous coaching sessions can be found at: 

http://www.open-aurec.com/Skype/PaulBenson/PaulBenson.htm
 
Paul Benson. 

*  *  * 

White: Z. Selkovski (Macedonia), 1900. 
Black: P. Benson (Great Britain), 1928. 
Event: IBCA World Team Championship, Sofia 2018. 
Result: 1/2-1/2 in 59 moves. 
Opening: Sicilian Defence, Hyper-Accelerated Dragon, B27. 

1. e4 c5 
2. Nf3 g6 

| A variation studied with Chris Beaumont before the event with the intention of being a surprise weapon. || 

3. d4 cxd4 
4. Qxd4 

| My opponent varies from his previous games in this system making our preparation of, 4. Nxd4, for this game of little value. 
Now black enters the internal psychological battle of wondering just how much preparation white has done on the, 4. Qxd4, system. 
Is he suspicious of a specific preparation involving one of his previous games, or is this a genuine shift in his repertoire? 
The former is nothing to fear, the latter could find black in an unmapped minefield. 
Not to worry though, pre-tournament preparation by Skype has examined this line, only have to carefully recall the general principles and get the move orders correct. || 

4. ... Nf6 
5. e5 

| The most challenging variation, the black f6 knight is pushed across the board, it will not be able to offer defensive strength to the kingside. || 

5. ... Nc6 

| An important zwischenzug, the white queen must be pushed off the d4 square to give the black f6 knight a means of moving queenside. || 

6. Qa4 

| Again white plays the most challenging move. 
Either my opponent knows exactly what should be done in this system, a possible danger for me, or he is accidentally re-creating theory at the board, a position of potential danger for him. || 

6. ... Nd5 
7. Qb3 

| This move took more than 5 minutes to be played, it seems he is making it up at the board, great, right? 
No, not really. 
While white's, 7. Qb3, is definitely known, it is an uncommon sideline. 
And what is worse, the Skype pre-tournament preparation of the system did not include this precise move order, this position is completely new to me. 
Still, if it was not mentioned during the coaching then it cannot be feared, right? 
So, just follow general principles, recall how similar lines went, and perhaps we'll transpose back into familiar territory, right? || 

7. ... Nb6 

| Well, that is where the black knight went in similar lines when the white queen was still sitting on a4. 
Unfortunately, this quite natural move is a serious positional mistake. 
One danger noted while analysing to play, 7. ... Nb6, was a possible white battery involving white playing, Bc4, hitting hard at the black f7 pawn, and this has now been prevented. 
There is however now a big problem for black over on the queenside, just how are the c8 bishop and a8 rook to be developed? 
And just how is any traditional queenside opposite-wing attack involving the expansion of pawns to be achieved? 
Theory answers this problem with the line, 7. Qb3 Nc7 8. Bc4 Ne6 9. Bxe6 dxe6, with imbalances to please both players. 
Black failed to find this knight shuffle to protect the f7 pawn and is now condemned to suffer the consequences of not having sufficient experience in a new system. || 

8. Bf4 Bg7 
9. Nc3 O-O 
10. O-O-O 

| The anticipated opposite-wing attack scenario has arisen, the usual white plans can be put into action but black has much queenside unravelling to do before the pawns can be set in motion. 
This prompts black to try a different plan involving the elimination of the cramping white e5 pawn. || 

10. ... Qc7 

| If this is the way to proceed then it should have been prepared with, 10. ... a6, keeping control of the b5 square. 
During the coaching session the suggestion of, 10. ... d5, a somewhat desperate attempt to get some activity was discussed. 
The conclusion was that, 10. ... d5 11. exd6 Be6 12. Qa3, would at least solve the development of the c8 bishop but at the cost of a pawn. 
On reflection, 10. ... d5 11. Nxd5, is a much better way for white to accept the pawn as, 11. ... Nxd5 12. Qxd5, must be good for white. || 

11. Nb5 Qb8 

| Black is hoping to exploit the x-ray attack on the unprotected white f4 bishop, this plan is badly flawed as white is about to demonstrate. || 

12. Qe3 

| White has effectively set up a very favourable form of the well-known, 150-Attack. 
The general idea is to advance the h2 pawn to h5, place the f4 bishop on h6, trade, hxg6, then, Bxg7, and finally the crushing invasion of, Qh6+, should very quickly conclude matters. 
It is very difficult for black to prevent this without creating weaknesses elsewhere. 
In essence, black is positionally busted and is heading for the Olympiad equivalent of the field-sport early bath. 
All very disappointing, and having achieved a good position in Round 1 and thrown it away, black is now confronted with having blundered into a bad position with virtually no prospects of escape. 
Well, virtually no prospects unless one, "Punches The Chaos Button..." || 

12. ... g5 

| This took about 25 minutes of frantic thought of how to avoid the, 150-Attack, plan. 
Black is trying to randomise the position in the hope the complexities will create sufficient confusion that something might just emerge to offer chances of survival, as recommended in Chess For Tigers. 
This diagnosis of a bad position should take place as early as possible, and in order to be successful with this type of response it is necessary to admit to oneself the game is going badly and act immediately. 
Simply sitting there hoping that the opponent might similarly misplay will usually receive the reward it deserves. || 

13. Bg3 

| White declines the Chaos and by removing the bishop from attack sets up some really horrible tricks involving the advance, pawn e6. 
Some sample lines of accepting the confusion: 
(A). If, 13. Nxg5 Nxe5 14. Qh3 h6 15. Ne4 d5 16. Qg3 dxe4 17. Bxh6 Ng6, and black has snatched a piece. 
(B). Or if, 13. Nxg5 Nxe5 14. Qh3 h6 15. Nf3 Nd3+ 16. Bxd3 Qxf4+, black has eliminated the important white dark square bishop and has escaped the worst. 
(C). Or if, 13. Nxg5 Nxe5 14. Qe4 f5, black will be ready for, pawn d6, then, pawn h6, and finally after, Be6, the middlegame can begin. 
(D). Or if, 13. Bxg5 Nxe5 14. Bxe7 Re8 15. Bd6, white wins the black queen for a piece, black must improve on this disaster. 
(E). Or if, 13. Bxg5 Nxe5 14. Bxe7 Nec4 15. Bxc4 Nxc4 16. Qg5 Re8 17. Bf6, black is getting mated with, 18. Qxg7+, improvement needed. 
(F). Or if, 13. Bxg5 Nxe5 14. Bxe7 Nec4 15. Bxc4 Nxc4 16. Qg5 h6 17. Qh4 Re8 18. Bf6 d5 19. Bxg7 Kxg7, and white seems to have won a safe pawn. 

There are also lines beginning, 13. Bxg5 Bxe5, to be explored and assessed by white. 
Black does not really care if, 12. ... g5, is sound or unsound. 
It seemed the only sensible way of creating Chaos in order to deflect white from breaking through on the kingside as previously described. 
If it falls into something else equally horrible then so what, all it would mean is that after, 12. Qe3, black is lost and the players are only arguing on how it could occur. || 

13. ... g4 

| The only way to continue, the white e5 pawn must be undermined if possible. || 

14. e6 

| A very strong move, anything involving moving the attacked white f3 knight allows black the luxury of, 14. ... Nxe5, followed with, pawn d6, and black stands much better. 
Black must now close the b8 - h2 diagonal to avoid losing the b8 queen, but this also comes with tactical problems. || 

14. ... d6 
15. Ng5 

| White is fearing the Chaos too much, it seems the randomisation began with, 12. ... g4, has unnerved the opponent. 
Instead white should most definitely stick the boot in with, 15. Nxd6, some sample lines run: 

(A). If, 15. Nxd6 exd6 16. Bxd6 fxg3 17. Bxb8 Rxb8 18. gxf3, intending, Rg1, and, Qg5, black does not have sufficient compensation for the lost queen. 
(B). Or if, 15. Nxd6 gxf3 16. Nc4 Nxc4 17. Bxb8 Rxb8 18. exf7+ Rxf7 19. Bxc4, black is at least a piece down and probably losing an exchange as well. 
(C). Or if, 15. Nxd6 gxf3 16. Nxf7 Ne5 17. Nxe5 Bxe6 18. Ng6 Qc8 19. Nxf8, black has lost an exchange without any obvious compensation. || 
15. ... fxe6 

| Perhaps, 15. ... Ne5, was more accurate here, finally eliminating the white tricks arising from, Nxd6. 
With the game move black was intent on scaring white into thinking the semi-open f-file might carry threats and keep thinking in terms of defending rather than trying to exploit the weakened black kingside. || 

16. Qe4 

| Threatening mate with, 17. Qxh7+, but this is easily blocked. 
Instead, 16. h3, forcing open the h-file for the white h1 rook looked much more dangerous. || 

16. ... Rf5 
17. Nxe6 Rf6 
18. Nxg7 

| White chooses a safety-first approach and eliminates the potentially dangerous Dragon bishop. 
Instead, 18. Nec7, wins an exchange which should be quite a safe snatch as black is in no position to launch a violent attack against the white king as punishment for such time-expenditure. 
Black cannot exploit the mutual protection of the tandem white knights, 18. Nec7 a6 19. Nxa8 axb5 20. Nxb6, and the wandering white knight escapes intact. || 

18. ... Kxg7 
19. Bd3 Bf5 
20. Qe3 

| White keeps up the pressure on the dark squares, perhaps planning an awkward, Qg5+, if appropriate. 
Instead, 20. Qe2, intending, Bxf5, and then, pawn h3, should give white a comfortable kingside initiative. || 

20. ... Qc8 

| Trying to make it look as if something could be happening up the c-file while keeping an extra guard on the advanced g4 pawn. 
Instead, 20. ... Qf8, with pressure up the f-file, offering much-needed protection to the e7 pawn and permitting the a8 rook the chance to centralise was perhaps more logical. || 

21. Bh4 Rf7 
22. Rhe1 

| White prefers to build slow pressure against the central black pawns rather than force open the kingside with, 22. h3. || 

22. ... h6 

| Designed simply to prevent any annoying, Qg5+, invasions. || 

23. Nd4 

| White is threatening both the black f5 bishop and to trade off a defender of the e7 pawn with, 24. Nxc6, black had better find something good here or the position is going to fall apart. || 

23. ... Nd5 

| The knight misplaced as long ago as move 7 finally contributes to the game, there is a useful tempo attack on the white queen while backwardly defending the e7 pawn. || 

24. Nxf5+ 

| Ouch! 
The black f5 bishop falls off with a check and the only defenders are a rook and a queen, time to resign, right? 
Instead, 24. Bxf5 Nxe3 25. Bxc8 Nxd1 26. Bxb7 Rab8 27. Bxc6 Nxb2 28. Rxe7, has the sort of material imbalance which offers each player many chances to go wrong. || 

24. ... Qxf5 

| No need to resign, the tactics save black. || 

25. Bxf5 Nxe3 
26. Rxe3 Rxf5 
27. Bxe7 

| White allows black to make gains, at the end of the captures black will have a 7th rank rook and a kingside pawn majority. 
These plusses are not enough for a decisive advantage but are enough to keep white on the defensive. 
Instead, 27. Rd2 Kf7, and black is holding the position. || 

27. ... Rxf2 
28. Bxd6 Rxg2 
29. Bg3 Rf8 

| Probably forced, the coming white rook invasion of the 7th rank needs neutralising. || 
30. Rd7+ Rf7 
31. Rxf7+ Kxf7 
32. c3 

| Accompanied with an offer of a draw, quite reasonable under the circumstances. 
White is using the c-pawn to blunt the activity of the black c6 knight, black is not going to be put off so easily. || 

32. ... Ne7 

| A bad decision to reject the draw is compounded with a bad move. 
How many times has this happened in your own chess games? 
So what encouraged black to decline a very fair offer of peace? 
Black plans kingside activity, hoping the presence of the knight over there will soon create a passed pawn which will ultimately cost white a bishop. 
Fine in theory, but in practical terms the white pieces have too much activity, in essence, black should have taken the draw and feel relieved to have avoided a loss. || 

33. Re4 

| The white rook is about to demonstrate the black king is overloaded, it will be needed to defend the e7 knight and the h-pawn, it cannot do both. || 

33. ... h5 
34. Re5 h4 

| This tactic doesn't work providing white is awake. || 

35. Bxh4 

| The temporary abandonment of the h2 pawn is irrelevant as white now has a double attack on the black e7 knight. || 

35. ... Ng6 

| Looks good but white can sneak out of the tactics a pawn up. 
Instead, 35. ... Nc6 36. Rf5+ Kg6 37. Rf2, and white is a comfortable pawn up, black will struggle for the rest of the game. || 

36. Rf5+ Kg7 

| Instead, 36. ... Ke6 37. Rf6+, and black loses the g6 knight. || 

37. Bg3 Kh6 

| Black is running out of ideas. 
Without a kingside pawn majority the only chance now is to force a knight for bishop trade on g3 and hope to win the white g3 pawn making the black g4 pawn passed. 
In the meantime the white rook will be munching down all the black queenside pawns making the 3 white queenside pawns into a mass of passers, all the winning chances are with white, even if the advancing black g-pawn costs white his rook. || 

38. Rf7 Kg5 
39. Rxb7 Nf4 

| Totally disastrous. 
When turning down the draw black had assessed the position as offering winning chances due to the 7th rank rook, the mobile kingside pawn majority, which when combined with an invading king and knight should win material. 
Now having lost a very important pawn black still tries to make the optimistic plan work but instead hands white the opportunity to win on the move. || 

40. Rg7+ 

| Returning the favour. 
Instead switching the rook in a different direction with, 40. Rb5+, forces the black king backward, 40. ... Kg6, when, 41. Bxf4, white wins both the black knight and the game. || 

40. ... Kf6 
41. Rxa7 Nd3+ 

| Black must now throw everything up the board hoping the active pieces are sufficiently well-placed to punish any errors from white. 
However since both players have about 10 minutes each remaining on their clock plus a 30 second minute increment returned per move this seems rather unlikely. 
A different plan was suggested during the coaching session, 41. ... Ne2+ 42. Kb1 Nxg3 43. hxg3 Rxg3, and try to promote the passed g4 pawn to cost white the rook. 
For this plan to succeed black will need the king to be on f2, a rook somewhere on the f-file, perhaps f8 is sensible, and the pawn on g2, with white having much earlier placed the rook on the g-file behind the black passed pawn. 
Promotion of the black g-pawn will cost white a rook but the black king will then be on g1. 
Now just how many moves by black does this take? 
And while black is organising this, just what will white be doing on the queenside? 
Will the black rook and distant g1 king be able to cope with the white mass of connected passed pawns as they sprint up the board? 

Instinct suggests white will comfortably win this sort of race. 
So black in live play at the board adopts a plan of winning 1 of the white queenside pawns in the hope the split pawns will be difficult to advance safely. 
Another important factor to consider is the status of the white king. 
This unit must not be allowed to escape off his back rank, if he gets out to support the advance of the queenside pawns then black will find it virtually impossible to avoid losing material due to pawn-promotion. || 

42. Kd1 

| Perhaps the counter-intuitive, 42. Kb1 Rxb2+ 43. Ka1, seemingly placing the white king into a possible mating net offers white an easier plan of promoting the a-pawn. 
The black rook would be denied access to the a2 square, and if the black rook retreats with, Rb6, the white king can start advancing behind the a-pawn as it proceeds up the board. || 

42. ... Rxb2 

| Instead, 42. ... Nxb2+ 43. Ke1 Nd3+, gives the white king a little test. 
It can shuffle kingside to defend the h2 pawn which would release the g3 bishop to support the advance of the queenside pawns. 
Alternately, he might stay in the centre with, Kd1, guarding the c2 square against a possible black, Rc2. 
With the game move of, 42. ... Rxb2,the white king is given no choice, the combined efforts of black rook and knight have him imprisoned, black now has the extremely remote dream of a possible mating net by placing a king on e3 to support, Rd2+ mate, this will of course never be permitted. || 

43. a4 Kf5 

| Black chooses the plan of trying to create a passed g-pawn supported with an active king, this should not work but something must be tried, and as they often say, "Attack is the best form of Defence". 
The alternate plan of sending the black king over to the queenside is defensive, this would not put white to any tests regarding calculating how to stop a black passed g-pawn while trying to promote a queenside pawn of his own. 
Giving your opponent the task of coping with an opposing attack while carrying out his own will chew up more time on the clock, just what is needed as he might become frustrated when a clearly won position is proving difficult to convert. || 

44. a5 Ke4 
45. Re7+ 

| A double purpose move. 
Firstly, the black king is not allowed to settle on e3 when, Rd2+ mate, would be possible. 
Secondly, the a7 square is vacated for the passed a-pawn to occupy. || 

45. ... Kf3 
46. a6 Ra2 

| The black rook must now permanently remain on the a-file to prevent promotion. || 

47. a7 Nf2+ 

| Black has the right idea but executes it incorrectly, white now has a clear win to be found. 
Instead, 47. ... Nc5, is the correct route to take to play, 48. Ne4. || 

48. Kc1 

| Without much time to invest white chooses the safety-first option which is to avoid exchanging material. 
With more time it is likely that the winning idea of, 48. Bxf2, could be successfully analysed, there are only 2 short variations to be found: 
(A). If, 48. Bxf2 Rxf2 49. Rf7+ Kg2 50. Rxf2+ Kxf2 51. a8=Q, white is easily winning, however the capture with the king on f2 is trickier to refute. 
(B). Or if, 48. Bxf2 Kxf2 49. Re2+ (forcing a deflection) Rxe2 50. a8=Q, white is clearly winning with careful technique. 

48. ... Ne4 

| Forking the white g3 bishop and the c3 pawn, again time shortage for white is going to influence the game. || 

49. c4 

| We all know about the unwritten law: "Passed pawns must be pushed", fine, but unfortunately white is focusing on the wrong pawn. 
There is a nice win with, 49. Be5, the idea is to place the bishop on d4, defending the a7 pawn, after which, Re8, supports the material-winning promotion. 
Black will have some moves to splice into this plan, perhaps the most logical line could have been, 49. Be5 Nd2 50. Bd4 Nf1 51. Re8 Nxh2 52. a8=Q Rxa8 53. Rxa8, and white has a decisive material advantage. || 

49. ... Nxg3 
50. hxg3 Kxg3 

| The optimistic black plan of king-activation is beginning to bear fruit. 
Sure, black is still losing but finding the winning plan for white with little time on the clock becomes a difficult task. || 

51. c5 Kf2 

| This is superior to, 51. ... Kh2, which would allow white a timely, Rh7+, forcing the black king back onto the g-file to block the advance of the g-pawn. || 

52. Rf7+ 

| Well-motivated but flawed, time shortage again. 
White is flicking in a rook shuffle intending to get the rook behind the passed black g-pawn without costing a tempo. 
This plan actually assists black, white has a great opportunity available if the black king remains on the f2 square where it is vulnerable to a shock move. 
The idea is, 52. c6 g3 53. Rg7 g2 54. Rxg2+ Kxg2 55. c7 Rxa7 56. c8=Q, is a theoretical queen versus rook ending win requiring a little technique. 
Explaining how to win such an ending would involve a separate article, however the basics for both sides are offered here. 

The Defender: 
Keep the rook in defensive contact with the king, this is creating a king-rook cluster, do not try to play actively with the rook, it will soon fall foul of a forking queen check if undefended, this should happen very quickly if it is on the opposite side of the board from the king. 
It is simpler to avoid placing the rook on the same diagonal as the king, in this formation they are vulnerable to a queen pin, very risky if the opposing king is close by, so having the rook and king on the same file or rank if possible is much safer. 

The Attacker: 
The overall aim is to slowly force the opposing king into a corner. 
This is done by advancing both the king and queen toward the defensive king-rook cluster, in this way the rook will not be permitted to give a sequence of checks to the attacking king. 

The Ideal Position for the Attacker: 
There is a 4-fold symmetry as the defending king can be forced into any of 4 corners. 
All the pieces must be lined up on the long diagonal with the defender to move. 
e.g. Place the defending king on h1, defending rook on g2, attacking king on f3, attacking queen on e4. 

General strategy for the Attacker from the Ideal Position: 
If the rook moves then the attacking king can give a discovered check and mate will quickly follow. 
Instead if the defending king moves then the attacking queen moves either up the file or across the rank to the edge of the board without giving check, this queen move is designed to prevent the defending king from returning to the corner square. 
This restrictive attacking queen location inflicts a zugzwang forcing the rook to break out of the defensive king-rook cluster, after which it will quickly fall to a fork from an attacking queen check. 

Some sample variations from the Ideal Position with the defender to move: 
(A). If, 1. ... Ra2 2. Kg3+ Kg1 3. Qe1+ mate. 
(B). Or if, 1. Rh2 2. Qe1+ mate. 
There will be similar rotated position mates if the defending rook chose to move along the g-file. 
(C). Or if, 1. ... Kg1 2. Qh4 (zugzwang, controlling the corner square) Rh2 3. Qe1+ mate. 
(D). Or if, 1. ... Kg1 2. Qh4 (zugzwang, controlling the corner square) and the rook has 7 other moves to make it more difficult, horizontally with, Ra2, Rb2, Rc2, Rd2, or vertical with, Rg6, Rg7, Rg8. 
This would imply one must learn how to handle 7 replies, not so. 
The symmetric nature of the chess board means the same method can be used once the defending king and now-distant and unprotected rook are not on the same file or rank, here it means they must not both be on either the g-file or white 2nd rank. 
A different way of assessing the position is to examine which way the defending rook runs, it can approach the attacking queen or distance itself from the attacking queen. 
So if the defending rook chose to approach the h4 attacking queen with any of, Rg6, Rg7, Rg8, then the queen must push the defending king off the g-file with, Qe1+, making the defending rook distant from herself as well. 
Then after, Kh2, the attacking queen centralises with a check, Qe5+, and soon wins the distant rook. 
If instead the rook ran into the distance with, Ra2, Rb2, Rc2, then the attacking queen is also distanced and now has the correct formation of non-alignment of file or rank and centralises with, Qd4+, and soon the rook will fall. 
Finally after the queen zugzwangs the defending king with, Qh4, and the defending rook played, Rd2, then, Qe1+, picks it up immediately. 

Apologies for not showing how to finish it in all variations, please consult any of a variety of sources of endgame texts, or even better work it out for yourself, you'll learn a lot more this way. 
Clue: The critical queen moves tend to be either playing up/down a diagonal to give check, or a sideways check to hit the undefended rook up/down a diagonal, happy hunting. || 

52. ... Ke3 

| The black king might have a different job in mind now, that is to get over to the queenside. || 

53. c6 

| Strictly speaking this is not a mistake. 
However, white is missing an opportunity to gain a tempo over the game continuation with, 53. Rg7 Kf3, and the white rook has landed on the g-file in 2 moves at the expense of the black king making 2 moves. 
As it happens this does not matter here, white is still winning, but in another game missing such a shuffle could prove costly. || 

53. ... g3 
54. Rg7 

| The white rook must get behind the passed black g-pawn as mutual promotion favours black shown with, 54. c7 g2 55. c8=Q g1=Q+, forcing mate. || 

54. ... g2 
55. Rg3+ 

Time pressure finally takes it's toll. 
Black has had little to think about since entering the rook ending and has been accumulating time on the clock as those 30 seconds are added on. 
White however has been spending time trying to find a win, it is here but is quite a tricky one to find when short of time. 
It goes, 55. c7 Rxa7 56. c8=Q Rxg7 57. Qc3+ Kf2, and now a very important zwischenzug, 58. Qf6+. 
This queen check is forcing the black king either to move away from supporting pawn promotion or to occupy the promotion square and so prevent pawn promotion, in both cases white then captures the rook to achieve a winning position of queen versus knight's pawn on the 7th rank. 
The process runs, 58. ... Ke2 59. Qxg7 Kf2 60. Qf6+ Ke2 61. Qg5 Kf2 62. Qf4+ Ke2 63. Qg3 Kf1 64. Qf3+ Kg1, the black king occupies the promotion square giving the white king a chance to weave a mating net. 
Play would conclude with, 65. Kd2 Kh2 66. Qf4+ Kh3 67. Qf2 Kh2 68. Qh4+ Kg1 69. Ke3 Kf1 70. Qf2+ mate. 
Finally, this technique of gaining tempi permitting the attacking king to approach is also useable against an e-pawn or a d-pawn. 
The rules are different for a bishop's pawn or rook's pawn, the attacking king must be close enough to the opposing 7th rank pawn to weave a mating net, consult those all-important endgame texts for the precise circumstances determining if a win is possible. || 

55. ... Kf4 

| The black king could also have chosen either e4 or d4 to escape the check. 
However the apparently attacking, 55. ... Kf2, is a blunder as, 56. Rxg2+ Kxg2 57. c7 Rxa7 58. c8=Q, is a won ending for white. || 

56. Rxg2 Rxa7 
57. Rc2 Rc7 
58. Kb2 

| Attempting to cut off the black king from reaching the c6 pawn fails by just a tempo. 
It goes, 58. Rc5 Ke4 59. Kb2 Kd4, and since the white king cannot land on the b4 square the white rook must retreat with, 60. Rc1, after which, 60. ... Kd5, the white c-pawn falls. || 

58. ... Ke5 
59. Kb3 Kd6 
Draw agreed, 1/2-1/2 

| A demonstration of a couple of chess maxims, please forgive any paraphrasing: 
"The hardest thing in chess is to win a won game." 
"The king is a strong piece". 

In closing, a successful English Grand Master commented that people tend to select short games for study because they enjoy searching for the quick kill, perhaps hoping to use such methods in their own games. 
Fine, but these short games are short because errors are being made and then punished. 
What is one supposed to do if the opponent resolutely refuses to make a mistake? 
He concluded that studying long games will also bring rewards, especially those ending in a draw, just the games people tend to ignore. || 

*  *  * 

White: Z. Selkovski (Macedonia), 1900. 
Black: P. Benson (Great Britain), 1928. 
Event: IBCA World Team Championship, Sofia 2018. 
Result: 1/2-1/2 in 59 moves. 
Opening: Sicilian Defence, Hyper-Accelerated Dragon, B27. 

1. e4 c5 
2. Nf3 g6 
3. d4 cxd4 
4. Qxd4 Nf6 
5. e5 Nc6 
6. Qa4 Nd5 
7. Qb3 Nb6 
8. Bf4 Bg7 
9. Nc3 O-O 
10. O-O-O Qc7 
11. Nb5 Qb8 
12. Qe3 g5 
13. Bg3 g4 
14. e6 d6 
15. Ng5 fxe6 
16. Qe4 Rf5 
17. Nxe6 Rf6 
18. Nxg7 Kxg7 
19. Bd3 Bf5 
20. Qe3 Qc8 
21. Bh4 Rf7 
22. Rhe1 h6 
23. Nd4 Nd5 
24. Nxf5+ Qxf5 
25. Bxf5 Nxe3 
26. Rxe3 Rxf5 
27. Bxe7 Rxf2 
28. Bxd6 Rxg2 
29. Bg3 Rf8 
30. Rd7+ Rf7 
31. Rxf7+ Kxf7 
32. c3 Ne7 
33. Re4 h5 
34. Re5 h4 
35. Bxh4 Ng6 
36. Rf5+ Kg7 
37. Bg3 Kh6 
38. Rf7 Kg5 
39. Rxb7 Nf4 
40. Rg7+ Kf6 
41. Rxa7 Nd3+ 
42. Kd1 Rxb2 
43. a4 Kf5 
44. a5 Ke4 
45. Re7+ Kf3 
46. a6 Ra2 
47. a7 Nf2+ 
48. Kc1 Ne4 
49. c4 Nxg3 
50. hxg3 Kxg3 
51. c5 Kf2 
52. Rf7+ Ke3 
53. c6 g3 
54. Rg7 g2 
55. Rg3+ Kf4 
56. Rxg2 Rxa7 
57. Rc2 Rc7 
58. Kb2 Ke5 
59. Kb3 Kd6 
Draw agreed, 1/2-1/2 

*  *  *