Game of the Day 4

I have more than 1 book on Rubinstein, a player I'm fascinated with. Rubinstein learned the game relatively late in life compared to most great players at the age of 16. But already 10 years later he was beginning to establish himself as one the top players in the world. He was strong enough to be considered for a World Championship match against Lasker, but never got the chance to play because of the outbreak of the First World War.

Rubinstein's style was fairly universal, but he tended more toward the safe side of chess. He was a great innovator int he opening, a fantastic attacking player when he had the chance, great with the initiative, and solid in defence, but his true forte was the endgame, and especially rook endgames.

I have 3 books about Rubinstein and others that contain short parts about him. The most basic of the 3 biographies is "100 Selected games" by Hans Kmoch, a master strength contemporary of Rubinstein. The notes in this book are fairly simple to follow, and it provides a great introduction to Rubinstein. Much of this book is available to browse in google books.




The game I have chosen today is Spielmann-Rubinstein St Petersburg 1909. This tournament was a great triumph for Rubinstein who shared equal first with World Champion Lasker, and winning their individual encounter. Rubinstein also played a masterful pawn ending against Cohn in this tournament that everyone should know. Against Spielmann, Rubinstein came back from a bad position to win a rook endgame that seemed impossible to win.

Rubinstein was Black in this game, but if his opponent Rudolf Spielmann, a famed attacking player, could have found the best continuation here, we would have been talking about him rather than Rubinstein. See what you can find, and then check out the full game at the end of the article.

The main brilliance of this game starts from the following position:

White has more pawn islands than Black which makes Black's position preferable, but if White can swap his a and d-pawns for Black's d-pawn, the endgame with 3 vs 2 on the king side is almost certainly drawn. Rubinstein was aware of this and his play was directed against this. Have a look at the game with notes by Kmoch, and I will post a more detailed analysis of this endgame, which is annotated in Kasparov's "My Great Predecessors" volume 1 among others.

Anyway, enjoy the game!

Loading embedded chess game...

Endgame Technique

So how did you go with this position from my last blog post? It is black to play and win! The first question is which pawn should black give up, as any move Black makes drops a pawn. I guess most people would look at dumping the least advanced pawn, so a bishop move backwards or forwards calls out. 1..Bh4 2.Kxf4 Bd8 3.Kf3 Bc7

This is a zugzwang position. Any White move will lead to a worsening of the position. White's bishop is stuck preventing Black's king from coming to h2. So White must make a king move. As Kf2 allows Black's bishop to skewer White's remaining 2 pieces by Bb6+, the only move is 4.Ke2, though this allows Black's king to improve. 4..Kg3

Now if White's king moves, Black will play Kf3 and then Bg3-f2, flushing White's bishop from it's guard post. So White has to let Black's king in. 5.Bc5 Kh2 6.Bd4 Kh1 7.Bc5

White can only play a wait and see policy. But Black now has the winning plan of Bh2-g1 forcing White's bishop away from defending Black's pawn's promotion.

This is a very instructive technique. And even without this technique, I think my junior opponent should keep playing and making me suffer while he had even a 1 pawn advantage. It is something we all have to learn, and taking inspiration from the current World Champion, Magnus Carlsen, in this respect is the way to go. If there is any play left in the position, keep fighting!

Start of Season Rust

The summer holidays give a well deserved rest to many people. Taking a break, a step back from normality, getting away and recharging are all worthwhile endeavours. The summer break comes to an end and we all have to get back to reality. Schools start back, holidays finish and it's back to work, and for chess players, the close season is over and we start our year.

Of course not all players take a summer break from chess. The summer can be a busy time with national championships and junior championships taking place. What that means, is when the new season of club chess begins, some players are decidedly rusty, while others are hotter than ever. There have been a few poor results by pretty good players in the past week, but I'll stick to myself!

I have changed clubs and now play at Box Hill Chess Club. For my sins I was the top seed at the Box Hill Autumn Cup (though I think I should have been second seed to Luis Chan as his FIDE rating is higher than mine) and had to play a junior about 600-700 points below me. Unfortunately, Shawn Zillmann is going to rise over 100 points from his excellent performance in the Australian Junior Championship, bringing him closer to 1700, and his rise is probably not stopping there, so he might be closer to 1800-1900 in real terms. While I'm close to 2200, and should still put players like this away most of the time, in real terms I was playing more like 1700 for this game, and really should have lost.

Shawn played excellently, though a little passively, but he defended and maneuvered well and the game was pretty much equal for much of it.Then towards the endgame, I blundered allowing Shawn a winning advantage of the chance to make an outside passed pawn which had to be dealt with, and while I was doing that, he cleaned up my pawns and went into a same coloured bishop ending with 2 connected pawns while I had none. We finally came to the following position.

Obviously, Black can't play 65..g1=Q as White plays 66.Bxg1 and after 66..Bxg1 Kf4 winning the final pawn. The easiest win (in my opinion) is simply to play 65..Bd6 (dropping it back to a safe square anywhere along the diagonal works) 66.Bg1 (to stop black playing Kh2)

Now Black can simply lose a move with his bishop when White will be in Zugzwang having to give way with either bishop or king. Ok, so I saw this over the bard, and was relieved when Shawn played 65..f4.

I played 66.Bf2 and 66..Bg3 is forced after which I played 67.Bg1 preventing his king from coming to h2.

At this point Shawn's head fell a bit and I could see he was resigned to a draw. I even felt a little sorry for him, as he truly deserved to win. But at this point he offered a draw, and I accepted thinking I'd make a miraculous save. Much to my surprise, this position is still winning for black, and I'm sure Shawn's IM coach will have told him how.

The question is, can you work out how to win this position for black, without using a tablebase or engine (as you're not allowed to use them in the game? I'll post the answer next time!

Pawn Endgames 2

Yesterday, I introduced a pawn endgame between Csom-Suttles Indonesia 1982.

White had played for this position and must have thought the position was better or winning for him. For example, how will Black defend the king side pawns? 1..Ke5 is met by the opposition move 2.Ke3 and Black's king will have to give way as White has extra tempo moves with pawns. The only Black pawn that can move safely is the a-pawn, and that is, in fact, the move Suttles played, 1..a5. However, this doesn't solve Black's problem as after White advances their king 2.Kf3 Black has to prevent it coming to f4, 2..Ke5 and now 3.Ke3 gives white an opposition.

It is easy to see from this diagram that Black's king will be forced to move back. The only waiting move is 3..a4 but then the ball bounces back into Black's court after 4.a3 when Black will have to let white's king into f4. And even though Black can oppose the White king when it gets to f4, by moving to f6, White has spare tempo moves with the h-pawn. 4..Ke6 5.Kf4 Kf6

Now it really is clear to see that White will move their h-pawn and Black's king will have to move giving White's king a way forward. This is exactly what happened in the game, and White won this endgame.

So let's return to the original position:

the lines that I previously considered aren't too difficult to work out with just a basic understanding of pawn endings. So the only other measure is to counter attack. Black should look to run to the queen side with his king and try to promote simultaneously with White. So we come to the stage where we must count how many moves it will take to promote. Black has c6-b6-a5-b4-xc4-d3 and 4 pawn moves on 10 moves. White has f3-f4-xf5-h3-g4 (a trade on g4) then 4 moves with the pawn which equals 9 moves. As Black moves first that means that White will promote and then Black promotes straight after. 1..Kc6 2.Kf3 Kb6 3.Kf4 Ka5 4.Kxf5 Kb5 5.h3 Kxc4 6.g4 hxg4 7.hxg4

Now Black moves the king and both sides will promote simultaneously.However, where black moves their king is important. It might look as if b4 is a good square for black's king aiming at White's a-pawn as well as clearing the way for Black's c-pawn. But then White's king will be able to come back and defend.

7..Kb4 8.g5 c4 9.Ke4! and now if Black advances the c-pawn, White will simply move their king to d3 and that will be the end of promotion.

Black will have to advance his king to c3 which loses valuable time in the pawn race and White will promote a full move ahead of Black.

The correct move here is Kd3 (d4) shepherding the pawn through and promoting at the same time as White securing a draw.

Hopefully some of these variations will help you think about some key ideas in pawn endgames:

- should I trade into a pawn endgame?
- who has spare tempo moves?
- can either side win a crucial opposition?
- are there counter attacking ideas?
- counting accurately the moves to promotion is a good calculation skill.

Like I said, thse players are not world beaters, and the ending wasn't one of the greatest, but I found it interesting nevertheless.

Loading embedded chess game...

Pawn Endgames

I saw on the MCC Facebook Page an interesting post by FM Michael Baron. Michael said "I have my own list of greatest endgames" but then directed others to a chessgames site called the Greatest  Ever Chess Endgames. It is important for players of all levels to have some knowledge of endgames and some guiding principles to lead one's play. Theoretical endgames become more important as a player gets stronger, but as in opening study, it is probably better to work on general understanding rather than theoretical knowledge. In this respect, looking at great games where great endgame ideas were played is fantastic advice, that you'd expect from an FM like Michael Baron. When an endgame particularly catches your interest, then deeper investigation can be undertaken with a book like Dvoretsky's Endgame Manual, or Muller's Fundamental Chess Endings.

Here's an interesting endgame that I saw recently. It isn't brilliant, and the players aren't overly famous, but it is still interesting. The game was between Csom-Suttles Indonesia 1982.

White is much better in this position. The d5 rook is dominant, while black's king is a bit intimidated. For instance, white can win a pawn straight off with 1.exd6+ Bxd6 2.Rxf5. However, both players were in the first time control and would have been seeing their clock hand turning toward the flag, so the Hungarian Grand Master decided to head for the pawn ending which he must have considered to be significantly better for him. The game saw all the pieces traded over the next few moves:

1.exd6+ Bxd6 2.Rxd6 Rcxd6 3.Rd1 Kc6 4.Bxd6 Rxd6 5.Rxd6 Kxd6 6.Kf2

A big consideration in endgames is whether it is good to trade into pawn endgames. On the more general theme of trading pieces, it usually isn't good to trade your good pieces for your opponent's less good pieces, so one could argue that generally speaking, white has given away the advantage he held by the comparative strength of his pieces. However, if by simplifying into a pawn ending we guarantee a win, then it is an acceptable thing to do.

The question is whether white can win this? What would be black's best try here to defend?

I'll carry this on tomorrow....

Those Annoying Time Zones

The worst thing about living in Australia is the time zone differences with Europe and America where many important events happen. I'm not going to stay up into the middle of the night to watch Federer win Wimbledon, or Usain Bolt run the 100 metres in the World Championship. I can see the replays the next day, but it just isn't the same. Similarly with chess, I generally play catch up with events the day after they happen, which isn't bad in terms of keeping up to date, but it still isn't the same as watching some live chess.

While this is usually the case, I have been rather fortunate with the Sinquefield Cup. While I can't see the start of the games, I am able to catch the end of the games when I wake up in the mornings. For me this is far more preferable than catching a few opening moves before having to leave the games as they start to get interesting. So far, I've eagerly followed 3 endgames from the event. The amazing finish to Aronian-Caruana in round 2 had me spellbound. Anand's defence against Carlsen in the following round had me thinking abut my own technique. And then this morning, I wake to find the game Vachier Lagrave-Carlsen in full swing when Carlsen blunders to give MVL a winning endgame.

During the endgame this morning I had one of those moments where you realise that your understanding of the game is just not on a level with other players.

Now in my primitive way of thinking, black is 2 pawns up though white could win the f-pawn. However, winning the f-pawn involves trading bishop for knight which would leave a lost pawn ending where black just forsakes the h-pawn and marches the king to the queen side. So, this is an easy win and there doesn't seem much that white can do? Isn't it just time to resign?

Magnus continued with 63.b4. I was sitting at my computer watching the game, thinking a trade on b4 would probably be ok, or just advance the king to g5. If black's king can get to g3, it's game over. However, neither of these "obvious" moves would have been good enough to force a win. The only move here which leads to victory is the far from obvious 63..c4 and amazingly, that is what MVL played! Would I have played this move, or even thought of it as an option? Probably not. But a deeper look at the position makes it clear why my candidate moves aren't good enough.

63..cxb4 64.cxb4 Kg5 65.Kf2 [blockading the pawn and the g3 sqaure for black's king] 65..Kf4 reaching the following position

So the question is, how does black progress? At least one of black's pieces needs to protect the f-pawn which means only one of them can try to win white's b-pawn. But that won't happen because white's bishop will sit on c6 and eventually the b-pawn will advance to b5. Even worse, from c6 the bishop can go to e8 and win white's h-pawn!

63..Kg5 64.bxc5 bxc5 65.Bd5 Kg4 66.Kf2 Kf4

Very similar to the last position, black can make no progress. In fact, with white to move there is already a repetition likely by 67.Bf7 Kg5 68.Bd5,

So this all goes with the need for strong calculation at all phases of the game. If it is possible to see that these 2 moves lead nowhere, then the next thing to do is look for other moves. MVL's 63..c4 just loses that pawn, putting on the same colour sqaure as controlled by white's bishop. But in winning the pawn, white gives black time to mobilise their pieces, and black's knight especially, moves from its depressing defence from the the edge of the board to an attacking piece in the centre. 64.Bd5 Kf5 65.Bxc4 Kg4 66.Kf2

So far, all seems fairly natural, but what now? Black's knight has 3 squares to move to but they all appear to lose a pawn. 66..Ng2 67.Bd5 wins the f-pawn or black's knight has to return to h4. This must be bad as white's queenside pawns will start marching. 66..Nf5 67.Be6 pinning black's knight after which white will advance the queen side pawns forcing black's king to defend which allows white to win the f-pawn with and the h-pawn. So 66..Ng6! but this also loses a pawn to 67.Be6+ Kf4 68.Bf7

Black's knight is skewered to the h-pawn and black's king has taken the f4 square from it. But amazingly this position is winning, thanks to the activity of black's pieces and the advanced passes f-pawn which is being nursed to promotion. 68..Ne5 69.Bxh5 Nd3+ 70.Kf1

What a turn around in position. White has levelled the game materially, but white's king has suffered an indignity in being pushed to the back rank. White's bishop is also somewhat askew. Meanwhile, black's king is in great shape and can infiltrate further into e3 or g3 (MVL chose g3) while black's knight has transformed itself. Carlsen resigned a few moves later when the knight further improved it's position by the maneuvre Nd3-f2-e4-d2/c3 or Nd3-f2-d1-e3/c3. White would have to part with his bishop for the f-pawn nad cannot force a trade on the queen side.

Lessons learned? First, we all need to calculate stronger in all phases of the game. Second, it is wrong to make assumptions based on general concepts such as material levels. While mostly material is of primary importance, there are times when other factors need to supplant this. I know that I am overly materialistic in my games, so seeing more examples like this and trying to adopt similar ideas when appropriate can only improve my chess.

The Most Instructive....

I'm guessing that many chess players have been following some of the top class action in the northern hemisphere's summer season. For me, the British Championship is an interesting event to follow as I know some of the players. But like many, I've been keeping an eye on the Sinquefield Cup in America with Carlsen and a group of the world's elite. The first 3 rounds have brought about much fighting chess and a joy for those players who include 1.e4 e5 in their repertoires.

However, if we mere mortals want to improve our games and learn from the best what would be the 1 thing that we should study from this tournament so far? Well, any interesting position is good to study, and non standard types of manouvres, like Aronian's 10.Rh4 from his first round game vs Nepomniachtchi is a fun move.

Aronian-Nepomniachtchi Sinquefield 2017

In answer to black's 9..Qa5 Aronian here didn't defend his a3 bishop, or retreat it, but played 10.Rh4!? which protects the bishop due to the threat of Ra4 trapping a queen. Neat!

However, openings, tricks, fanciful ideas are to my mind beautiful to see but unlikely to bring many long term benefits to the game of an average club player. No, in my opinion, the best position to learn was Anand's defence of a rook ending a pawn down against Carlsen.

I'm sure that many people reading this blog will know that this is supposed to be a draw, but I wonder how many would be confident of holding this position with white against Magnus Carlsen? Anand did it comfortably and to be honest, white is starting with the best possible pawn structure for the defence and it is impossible for black to get his rook behind the pawn before white does. But the black a-pawn will force it's way down to a2 or a3. In fact, in just another 4 moves this position was reached.

White still just has to sit and wait for black to try something and then react, but what can black do? Advance on the king side? Bring the king to the queen side? With the pawn on a3, black's king has a hiding square on a2, but white's rook will be able to take king side pawns as black won't be threatening to promote. So the other option is to advance the pawn to a2, but what then? black's rook is as immobile as white's, and if black's king comes to the queen side, it will be subject to checks that it won't be able to escape from.

Carlsen didn't give up trying, and eventually, this position was reached with black pushing his g-pawn. So what would you play here as white? What would be your candidate moves? Perhaps hxg5, or Ra7+, or even Kf3? Let's look at trading a pair of pawns as that's what we're told we should do as defenders. After 1.hxg5 fxg5

Now what? White's king is becoming more open, and what black would like to do is have his rook on a1, pawn and a2 and swing the rook over to do a check. So imagine doing nothing like 2.Kf3 Ra1 3.Kg2 

Now the position is critical for white after 3..g4! If white shuffles the king, black's king will come to the queen side, while if white aims for more trades with 4.f4 exf3 Kxf3, then white's king becomes more exposed.

The whole endgame is a nightmare, and there are simplifications to other endgames that need to be understood as well. Anand's solution was excellent. If we go back a few moves:

Anand chose to play 60.g4! here. I have to admit, it wouldn't have been first among my candidate moves, but the resulting positions are all level.

If black captures hxg4, then white can play hxg5 fxg5 and Kg3 picking up one of the g-pawns before retreating the king back to the corner. Carlsen took the other way 60..gxh4 but after 61.gxh5, this h-pawn gives white sufficient counterplay. Look at the final position when the game was agreed drawn.

It is black to move and although he has an extra pawn, and 2 passed pawns, black has virtually no moves. Playing h3 will allow Kh2 when the only move is Kg8 but a repetition will occur after Rg7+ and the rook will fly back to a7.

While opening knowledge and tactical and imaginative flair are essential ingredients in a players arsenal, learning technique can help us save valuable half points, or like with Carlsen, squeeze out victories from unlikely positions. Remember that very often, the defence in the endgame is being carried out after 4, 5, 6 hours of intense concentration so it is important to keep trying and to keep putting pressure on opponent's. Here's the endgame with some notes by me. I strongly urge anyone with any chess ambition to learn the technique from Anand's play so that you can use it your own games.

Endgame at a School

I was at a Primary School this morning and the following position appeared in one of the games.

Even though the boys were 2 of the better players in the group I was still impressed and astounded that they had reached this position. From here on, though, the game was full of blunders, but each one was instructive for the kids. It was black to play and he started with the shocker 1..Kc5?? This immediately moves the king outside the square of the pawn and shows that kings can't catch pawns when they are behind them. In the above position, black has the undermining 1..a6! when his king is closer to the pawns and able to take both white pawns. Thus the kids are being taught about king activity and placement. The nearer their king is to the main action, the better.

After white's blunder, came a mutual shocker 2.Kd3?? allowing me to explain that passed pawns must be pushed. At this point the black player saw that he should have lost the game and he retreated his king rather than take on b5. 2..Kb6 and both kings manouvered 3.Kc4 Kc7 4.Kc5

At this point, black played the inspirational 4..Kc8! going straight back and was rewarded with a draw after 5.b6? axb6 6.Kxb6 Kb8 (with opposition) 7.c7+ Kc8 8.Kc6 stalemate.

While congratulating white on his fine defence I showed both boys how using the king actively is an important issue in endgames. Instead of 5.b6, white should have tried 5.Kd6! when the opposition move 5..Kd8 fails because after 6.c7+ Kc8 7.Kc6, it isn't stalemate as black has a pawn left.

Now black has to move their a-pawn. The best try for a kid is to play 7..a6, say loudly "oh no, can I take that back", and then mumble something about losing their last hope! But seriously, the majority of children that I teach would find it hard to resist the temptation of taking a free pawn. But the win is simply 8.b6 a5 9.b7 checkmate, while the tempting 8.bxa6 is stalemate again.

Of course this endgame isn't perfect, and is very simple, but it shows a number of endgame principles that everyone, not just kids, should know. And when kids play these positions and then receive instruction it sticks better in their mind than when they are learning endgames theoretically. By the way, all the advice I gave them came after the game, as I didn't want to interfere while the game was ongoing. Both these boys are yet to compete in events outside their school, but like most kids, the are learning fast.

Endgame of the Week

I haven't been playing much chess, or studying much chess recently for one reason or another. So when I do look at things, they tend to be about things I enjoy. And I must admit that as I get older, I become more interested in endgames. Downloading TWIC each week and then sorting the file according to number of moves lets one see which endgames are being played.

This week's longer games included a pawnless rook vs bishop ending which finished in a draw. The defender knew the technique of heading the king to the corner not controlled by the bishop. There was a knight and bishop checkmate, and some interesting queen and pawn and rook and pawn endings, including R + e and f pawns vs R which ended in a draw. I must admit, I don't know the theory of these endings, but roughly know that f and h pawns are mostly draws, and a and h pawns are mostly draws. It is something worth looking at to build my technique.

I was most interested in the following position, not least because a GM was getting bashed by someone 300 rating points lower.

I would imagine time was a factor here as white is about to play his 61st move. White will definitely lose their a-pawn, but can win material on the king side, which should be enough for victory. The best first move is 61.Ng7, when black's f-pawn falls. However, white played 61.Nd4+ and black traded on d4 going into a lost king and pawn endgame. 61..Bxd4 62.exd4 (to be fair, keeping the minor pieces on didn't hold out much hope either).

However, this game ended in a draw after both sides promoted.  From the diagram the game continued 62..Kb5 63.Ke2 Kxa5 64.Ke3 Kb5 65.Kf4 Kc6 when we get to the key position.

White now played the natural 66.Kxf5? missing the finesse 66.Ke5! when black's king will be forced away from d5 meaning that white will be able to capture both f and e pawns (and g and h if he wants). After the text move black promotes  the b-pawn in 8 while white promotes  the g-pawn in 7, but it is black to move so they promote at the same time. This leaves a dreaded queen and pawn ending which are notoriously difficult to win.

So once again, a natural looking move, a capture, turns out to be not the best move. A simple finesse of the king would have led to victory for white against a player 300 points higher rated. Still, a draw was not a bad result for the white player.

Chess Diaries or Notebooks

Do you keep a chess diary, or chess notebook? How do you store your chess knowledge? Do you bother to store your chess knowledge? And is it advantageous to keep a notebook or diary? And if you were to keep a diary/notebook would it be a paper one or an online one?

It's an interesting topic, especially as much of what we play is stored either online, or in databases electronically. But what about keeping a notebook, or set of notebooks divided into different topics. Apparently super GM Yasser Seirawan filled over 30 notebooks in his career. And top coaches from around the world suggest the use of a notebook in your chess study. Here's an excellent article on exactly that subject, from UK GM Nigel Davies.

So let's assume for a second that keeping a notebook is a good thing. What would or should your notebook contain? Davies seems to talk in terms of individually tailored chess wisdom, while Seirawan seems to prefer different themes specifically documented. I guess, in the end, it would be down to the individual, with some guidance from a stronger player or coach. I don't think there is anything wrong with making it whatever you want:

a series of games that have impressed you

positions that interest you or that have pushed your ideas of thinking

a record of your games and ideas and progress

opening notes

tactical themes

endgame ideas

strategic or positional ideas

things that interest you about chess ie history, news, biography etc

In fact, writing a blog is a bit like writing a journal. However, I must admit to missing using pen and paper, while a gain a shine to my fingertips from the typing.

Anyway, if you do keep a notebook or journal or diary about chess, please let me know.

This was a position I posted yesterday. It is white to play and win. The key idea is the pawn break f3-f4-f5 which will eventually clear the way for the f7 pawn to promote. So white wants to move the king, but to where, or does it make a difference? The winning move here is 1.Ke3. The idea is to head white's king to the defence of black's queen side pawn majority as well as clearing the way for the f-pawn to advance. 1.Ke4 is also possible, but it is far less easy, and will rely on a winning queen ending which can be a bit messy. After 1..b5 2.cxb5 Kxb5 white's king is well placed to cut off black's c-pawn so the f-pawn can already be moved.

3.f4 now leaves black's king having to fall back to stop immediate promotion, but that will leave the white king to grab the c-pawn, and then the game.

This would be a nice position to put in a notebook under the heading of pawn endings, with perhaps a breakthrough theme.

Study - Answer

A couple of days ago I posted a study by Bron that I found quite cute.

It was white to play and draw, and the theme of the puzzle is about trading bishops to leave black with 2 knights, insufficient mating material to force checkmate.

1.Bc8+ Kg3 2.Be6 A fork, forcing one of the knights to move, but they can both give check which allows a defence.

2..Nb4+ [2..Ne7 will fail to 3.Kd7 when there are no more useful checks, so one of the knights must fall] 3.Kc5 Na6+

Now that the a6 knight is protected, the other knight can be moved. So white gains time by attacking black's bishop, which must then move to a poor square. 4.Kd4 Bh7

At least both knights are now safe! However, the poor placement of black's bishop allows white to trade it now. 5.Bc4! gaining time by attacking black's unguarded knight on a6 5..Nb4 and the knight responds by protecting the d3 square which is where white's bishop wants to go.

And finally, white succeeds with the plan, by playing 6.Kc3! attacking the knight, but simultaneously eyeing the d3 square. 6..Nc6 The only safe square.

7.Bd3 now forces the trade of bishops, as black's bishop has nowhere to run, leaving the game with 2 knights vs a lone king, a draw.

While this might not seem that difficult, there is something of a beauty to the way it was composed, with white's bishop and king both able to gain time every move to improve their positions while black's bishop proved useful to start with in aiding the knight on the queen side, but ended up being the weak link in black's position.