Preparing for Brazil: Awonder Has Adream
By Alex Betaneli & Tommy Ulrich   
November 17, 2011
Far board: Tommy Ulrich vs Adream Liang, closer board: Awonder Liang vs Aaron Jing photo Allen Becker

In recent years, there has been a healthy crop of Wisconsin chess talent. However, this talent has to be nurtured and protected carefully as winter has already arrived and will last until early April. Where can one escape to avoid the bitter cold? Why, Brazil of course! The 8-year-old Awonder Liang is doing just that as he represents the country at the World Youth Championships starting this week. Awonder comes from a true chess family. His father, Yingming (Will) Liang is a strong A class player. Will has three chess playing sons: Adream, Awonder, and Able (yes, these are really their names). Adream is an A class player, Awonder is an expert (making him the top player for his age by a wide margin) and Able is a bit lower rated, but has several years to catch up to his brothers.

Awonder is the first Wisconsinite to be invited to this prestigious event for two years back to back. A few years ago Brian Luo and Alexander Velikanov were also the top Americans in their age categories and participated at the World Youth. Although Awonder is higher rated than Brian and Alexander were at his age, there is one title –that of Wisconsin Junior Champion—that has eluded him. But he has time: both Brian and Alexander won this title when they were ten.

Even though Awonder will only face 8-year-olds (granted, strong ones), he prepared for the event by playing in a tournament with high schoolers. The Wisconsin Junior Open is traditionally held in early November, which makes it a great warm-up for the World Youth.  After defeating lower rated (but much older) opponents in the first three rounds, Awonder found himself on board 2 against another talented youngster, Aaron Jing. Meanwhile, Adream faced the top-seeded Tommy Ulrich (2167) on board one. Many speculated about the possible fifth round dream match-up between siblings, but Tommy refused to cooperate and beat the older brother. Thus it all came down to the following game (thanks to Tommy Ulrich for providing the background and annotations):

Liang,Awonder (2087) - Ulrich,Tommy (2167) [B12]

1.e4 c6

In our last game, I played the Najdorf Sicilian. Awonder played a surprising piece sacrifice novelty and some sharp complications followed, where both players missed chances to gain a clear advantage. Eventually, the game ended in a draw. This time around, I preferred a variation with more clearly defined positional themes.
 2.d4 d5 3.e5
The old main line, 3.Nc3 , is now regarded as unpromising for White, as Black seems to equalize in the main lines.
3...Bf5 4.Nc3!?

Historically, this variation became popular in the 1980s, when Vassilios Kotronias published a book recommending it and Shirov began playing it at the highest level. The move Nc3 is a natural developing move, but it blocks the c-pawn; thus, once Black plays ...c5 White's center will be destroyed as the d4-pawn cannot be defended with c3. Thus, White must play very actively and start his kingside attack at high speed in order to justify his position's strategic drawbacks. On the other hand, Black's bishop on f5 is well-placed from a positional standpoint but could be a tactical problem in the short term, as White has a target to attack with g4 and h4. Modern theory considers that White's play is too sharp and Black can at least equalize with accurate play. [The "Short Variation," 4.Nf3 e6 5.Be2 c5 6.Be3 (with the idea of taking back on d4 with the knight and later opening the position with c4); and Gata Kamsky's 4.Nd2!? e6 5.Nb3 are currently the main lines.
4...e6 5.g4 Bg6 6.Nge2 c5

Other moves are possible, but this is the most principled choice, in my opinion. Black immediately gains a better central pawn structure, which will be useful if White's attack fails.

The most aggressive response. The other main line is 7.Be3 , developing another piece before beginning the attack.

Again Black plays in the most principled way, hitting the White pawn structure where it hurts the most. Now playing g5 or taking on h5 will permanently weaken the f5-square, so White is practically forced to play

threatening to take the bishop, which would dangerously weaken the e6-pawn. In order to preserve the bishop pair, Black has to sacrifice a pawn.
8...Bh7! 9.g5?!

White needed to grab the pawn in order to have something to compensate for his already mentioned long-term problems. 9.Nxh5 Nc6 leads to unclear, well-analyzed complications.

Coincidentally, a few weeks ago, someone played the same sideline (9. g5) against me at a tournament in Minnesota. There I reacted with 9...Bf5 and soon got into trouble, although probably Black's position is also fine there.

10.Nb5 is more testing, but after 10...Be4! (forcing White to block the d1-h5 diagonal) 11.f3 Bf5 12.Nxd4 Ne7 Black is doing well positionally thanks to his absolute control of the f5-square. Also, the e5-pawn is a little weak and White's king is exposed.

Black is still much better after this move, but there was an even better alternative. [I should have just taken the pawn with 10...Bxc2µ Normally, in an open position White should have enough compensation, but here Black's solid central structure prevents White from taking advantage of his lead in development. It's worth noting that 11.g6 Nc6 12.gxf7+ Kxf7 is actually just what Black is hoping for. The king is safe on f7, and the h4-pawn and g4-square are convenient targets for Black's pieces. Incidentally, the fact that White already stands much worse on move 10 illustrates how difficult to play this variation is for both sides.

11.Bb5 does not present any inconvenience to Black, as he can respond with 11...Nge7 followed by ...a6 to eliminate the pin.

The bishop goes to the wrong square. 11...Nge7! was correct, and Black can follow up with ...Nf5 and ...Be7 or, even better, ...a6 and ...b5 winning the e5-pawn. 12.Nb5 Nf5
White tries to put pressure on e6, as this point can be easily undermined with the pawn break g6. Objectively better was 12.Bd3, although this trade favors Black from a positional perspective as White's pawns are mostly on dark squares.


A serious blunder. While the possible punishment is mainly tactical, it should be noted that blunders are often slightly unnatural moves that violate the rules of development. This move is a perfect example. Black should have simply continued developing with 12...Nge7! , after which the sacrifice 13.Bxe6 fxe6 14.Nxe6 is not really dangerous; Black is much better after 14...Qb6 15.Nxg7+ Kd7 when f2 is hanging.

White could have taken advantage of Black's error with the tactic 13.g6! dxc3 14.gxf7+! (14.gxh7? Qxh4! 15.hxg8Q+ Rxg8 16.Rh2 cxb2 17.Bxb2 g5µ) 14...Kxf7 15.Nxe6÷

Black is much better again.
14.0-0 Bf5?

This time Black errs by focusing on positional gains at the expense of tactics. [After 14...0-0!µ White would simply lose a pawn for nothing.

White should have avoided the unfavorable bishop trade with 15.Bg2!
15...Bxh3 16.Nxh3 Qd5?


Even though Black is about to win a pawn, he should not trade queens as White's king is more exposed. Also, this move has a tactical problem. 16...Qb6 once again would have won a pawn for free.
17.Qxd5 Nxd5 18.Nhf4! Nxe5

Black has to allow the doubled pawns. [18...Nxf4 19.Bxf4 conveniently defends the e5-pawn; and White gets a pleasant blockading square after 18...Nde7 19.Nd3]
19.Nxd5 exd5

19...Nf3+?? 20.Kg2 Nxh4+ 21.Kh3+-
20.Kg2 d3
The clever knight retreat 20...Ng6! would have been rather annoying for White. During the game, I was afraid that I would lose the forward d-pawn, but White's h-pawn is weaker. For example, the straightforward 21.Kh3 0-0 22.Rd1 doesn't work because of 22...Rfe8 23.Nxd4 Re4!
21.cxd3 Nxd3 22.Bf4?!

White should have traded the f2-pawn for the dangerous passed d5-pawn. 22.Rd1! would have achieved this: 22...Nxf2 23.Rxd5 Bb6 and Black is only slightly better as he will lose the h5-pawn.

Black also underestimated the value of the d-pawn. Also, the king belongs on the queenside, even though it looks dangerous to put it there. More principled was 22...0-0-0! when White cannot win the bishop with 23.Rfc1!? because of 23...Rhe8! 24.Be3 (24.Kf3 b6µ) 24...Kb8!µ
23.Rad1 Nxb2 24.Rxd5

 An interesting endgame has been reached. White is down a pawn and Black has a queenside pawn majority, but the weakness of the h5-pawn and White's more active pieces should allow him to hold a draw.


The knight doesn't really belong on the edge of the board, of course, but I wanted to preserve the option of ...Nc3 at some point, once White's knight moves away from e2. 24...Bb6 removes Black's means of defending the b7-pawn, so White should respond by attacking it with 25.Rd7; 24...b6! was the best choice.
25.Ng3 Rfd8

Black can't hold the h-pawn, as 25...g6 26.Ne4 gives White a dangerous attack.
26.Rfd1 Bb6 27.Be5!

A strong defensive move. 27.Nxh5 Nc3! 28.Rxd8+ Rxd8 29.Rxd8+ Bxd8 30.Nxg7! Nxa2! and Black is better thanks to the queenside passed pawns.(30...Kxg7? 31.Be5+)
27...Rxd5 28.Rxd5 Rd8

I thought that trading rooks benefits Black as his pawn majority is more mobile and less easily blocked by the White king.
29.Rxd8+ Bxd8 30.Nxh5 g6 31.Nf6+ Kf8 32.Kf3

32.f4!? was an interesting alternative, as then 32...Ke7 runs into 33.f5!
32...Ke7 33.Ke4 Ke6

The endgame is approximately equal, but the game remains interestingly unbalanced with rival pawn majorities on opposite wings.
34.Bd4 Bb6?

This is the wrong trade, as surprisingly, White's knight is more important to him than his bishop. A better choice was 34...Bxf6! 35.Bxf6 Nc5+ 36.Kd4 Nd7 kicking the bishop away and preparing to advance the king to g4 - note that in the game, this important light square was guarded by White's valuable knight on f6.
35.Bxb6 Nxb6 36.f4!

White is doing well now as his pawn majority is farther advanced than Black's.

37.f5+! gxf5+?
I suspected that this move wasn't the best during the game, but at the same time, I saw that Derek had a somewhat better position on the next board, so I assumed that I needed to win in order to have a chance to qualify for the Denker. Thus, I aimed to unbalance the game, even at the expense of playing good moves...

Also very promising was the simple 38.Kf4! when White will win the f5-pawn next. 38...Nc5 39.h5 Nd3+ 40.Ke3 Ne5 41.Ne8!±
38...f4 39.Ke4!

Preventing possible Black counterplay. White could have immediately begun pushing the pawns: 39.h5!? This works very well tactically; for example, 39...Kf5 40.Ne4! b6 41.g6! Nc5! (41...fxg6? 42.h6+-) 42.g7! Ne6+ 43.Kd3 Nxg7 44.h6! A typical tactical theme: the knight is dominated by the outside passed pawn. 44...Kg6 45.hxg7 Kxg7 However, although White wins a piece, I doubt that he can win the game against accurate defense, as there is only one White pawn left on the board.
This may not be objectively best, but at least it's annoying for White and keeps the position unclear.
40.Kxf4 Nxa2 41.h5 Nb4!

The only route for the knight to return and stop the pawn is by d3 (or c6) and e5.

White was almost certainly winning after the precise 42.Ke4! Nc6 43.Nd5! with the idea of playing h6 and controlling the g6-square with Nf4. For example, if Black tries 43...Ne5 then 44.Nf4+ Kd6 45.h6+- wins outright as Black can't stop the pawn.
42...Nd5+ 43.Kf3

White's king needs to step carefully: 43.Kg4?? f5+! 44.gxf6 Nxf6+-+
43...Kf5 44.Nd6+ Kxg5 45.Nxf7+ Kxh5 46.Nd6 b6 47.Ke4 is easy for White to defend. I wanted to keep more tension in the position.

After this move, White loses his last two pawns unnecessarily. 44.h6 Ne7 45.h7 Ng6 46.Kg4= draws, as White has enough counterplay on the kingside to keep Black's king and knight stuck there while White's knight blockades the queenside pawns.
44...fxg6 45.hxg6 Ne7 46.g7 Kf7 47.Nd6+ Kxg7 48.Nb5 a5 49.Ke4 Kf6

Such a natural-looking move, and yet it's only here that White conclusively throws away the draw. 50.Nd4= and White can win the queenside pawns as his king is closer to the action.
50...Ke6 51.Kc4 Kd7 52.Nd4 Kc7 53.Kb5 Kb7

It's not too late for Black to miss the win: 53...Nd5?? 54.Nb3!= allows White to sacrifice his knight for the two remaining pawns.
Now 54.Nb3 is easily met by 54...Nc6

54...Nc8 55.Nd4 Nd6+ 56.Ka4 Ka6 57.Ne6 b5+ 58.Kb3 Kb6 59.Nf4 Ne4 60.Nd3 Nc5+ 0-1


Awonder lost this time, but let us not forget that his opponent (a talented junior in his own right) was twice as old. In the end, Tommy Ulrich and Derek Sachs shared the championship title and will soon have playoffs for the Denker spot.

Good luck to Awonder and the rest of the US delegation in the World Youth.   

Find more information on the official site, and follow the entire US delegation on Also see Andrea Rosen's article on the journey to Brazil with her son Eric.