Queen vs. Two Rooks

I’m continuing with my articles about material imbalances. This time it’s queen vs. two rooks

What can we say about queen vs. two rooks? Using the beginner material scale, two rooks are worth 10 points, and a queen is worth 9 points. Does that mean that two rooks are better than a queen? If that were a simple question to answer, then I wouldn’t be writing this article…


It depends, as usual, on coordination. Bad coordination, especially with more pieces on the board, is, in general, a recipe to disaster when facing the queen. The queen is a goddess at causing nasty cases of LPDO (loose pieces drop off). If the rooks are coordinated, however, then the two rooks can be an effective force.

Number of pieces on the board

Don’t underestimate the power of a queen and minor piece(s) combo. Those can be quite annoying, especially if the opponent’s pieces are badly-coordinated. The pieces can, of course, be cooperating with the rooks too, but usually it’s the side with the queen that is better off.

King safety

It would be criminal not to mention a thing or two about king safety, especially when we’re talking about queens. The queen has a reputation of mating unsafe kings, especially with the help of a couple pieces. And let’s not forget about perpetual checks; queens are good at that too.

I’m not saying that two rooks are not good attackers. No, they can be, but mostly when they are coordinated. “Ladder” mates exist, and two rooks on the 7th rank are true monsters. But the queen is generally better at dealing with weak kings than the rooks.

One thing that I should comment on is that, when looking up my games with the queen vs. two rook imbalance, I found that many of those were one-sided games, mainly in favor of the queen. It was not because the side with the two rooks botched it up but because the position was totally botched up to begin with.

Let’s look at a quick example of a position that isn’t one-sided.

Kopiecki, Edward (1963 USCF) – Brodsky, David (2201 USCF) Marshall Grand Prix April 2014


White has a queen for two rooks, and there are a couple minor pieces on the board – if the minor pieces weren’t there, then black would just be better. The rooks aren’t coordinated for the moment, and white is threatening mate. Here, black should go 20…Ne8, defending the g7-pawn and preventing white from infiltrating on c7. The position there is roughly equal, as neither side can do anything concrete. Instead, I went for inspired active play with 20… Ng6? and was in trouble after 21.Qc7! grabbing some queenside pawns.

The rest of the game wasn’t pretty. I basically tried to blow open the white camp with active play, but everything was under control for white, and I was objectively much worse. I managed to generate something but in the process botched up the complications and would have been lost had my opponent found a nice little tactic. Fortunately, he missed that tactic, and the position went back to objectively drawn. Then, he blundered again, and I won. Phew!

What are the conclusions from that game? Don’t overestimate the power of the two rooks and don’t give away pawns unless you have a legitimate reason to!

Last October, I got a chance to get the raw deal: two rooks vs. a queen with equal number of pawns and no other pieces on the board. In simple English, I got the raw deal. And, in simple English, the game turned into a festival of mistakes.

Tsay, Vincent (2152 USCF) – Brodsky, David (2430 USCF) Eastern Chess Congress 2016


Yes, I did use this game already in my queen vs. pieces article. This was after my messup that degraded my position from winning to better. So, what to say about this position?

The black rooks are coordinated, that’s for sure. There aren’t any loose pawns for either side, and black should be able to protect everything if necessary. The black king is defended by the rooks, though it can get a bit drafty (as it did in the game).

What’s the overall evaluation? Black is better, but he isn’t winning.

But how to try and win? The white king is a bit airy, but I didn’t see any realistic mating ideas. But how about the queenside pawns? None of them are really loose or anything, but it’s not a bad idea to eyeball them… In simple English, I needed to try to grab some queenside pawns without allowing a perpetual check. This wouldn’t be easy – I knew I had to resort to the old trade secrets of dancing around trying to make progress… AKA grinding.

Nothing particularly eventful happened for the next few moves, and we soon reached this position.


After having my rooks doubled on the f-file for the past few moves, I decided to try out the g-file with my last move 51… Kg7-h7!?, clearing the g7-square for my rook. I honestly doubt that the rook could accomplish anything effective there, but it was worth a try – I could always put my rook back on the f-file with no harm done.

Vincent had defended well up to this point, but here he cracked with 52.h4?. I thought this was a bad move but for the wrong reasons. Can you find the win for black here? I’ll come back to the solution later in this article.

I went 52… Rg7+? 53.Kh2 Rf2+ (going 52… Rf4 would have gotten there immediately, but I wanted to dance around a little) 54.Kh3 Rf3+ 55.Kh2 Rf4 56.Kh3 Rxd4


Yay! I’ve snagged a pawn! The only problem is that after 57.Qf8!, black cannot stop perpetual check. Oops… Instead, I got lucky when Vincent played 57.Qe5? which was the right idea but the wrong execution. After 57… Rd3+ 58.Kh2 Rf7, I stopped the perpetual. Still, after 59.Qe6, I had to figure out what to do. Black’s best policy is actually to give up the d5-pawn with 59… Rdf3 in some form or another to stop the perpetuals. Black retains some advantage there. I went 59… Rff3?. Here, white would have had a draw after 60.h5!, not letting black escape with the king via g6. After some thought, my engine gives triple zeros. Instead, Vincent couldn’t resist checking with 60.Qe7+?, after which my king successfully flees. The game continued 60… Kg6 61.Qe6+ Kh5 62. Qe5+ Kxh4 63.Qe7+ Kg4 64.Qe6+ Kf4 65.Qxh6+ Ke4 66.Qh4+ Ke3


All those checks may seem scary, but I was well aware that this was no perpetual. My king has run towards the queenside, and my rooks are now helping shield his majesty. This is the point where white starts running out of checks. A few moves later, I won.

I want to go back to the moment after 52.h4?, where I had a win.


Black should be concerned that his king doesn’t get into a perpetual check, but the winning move here is ironically 52… Kg6!. Black’s plan is simple: play Kg6-h5xh4. Then, white will be forced to trade his queen for black’s two rooks after Rf2+, resulting in a completely winning pawn endgame for black. How does white stop this? Well, he can’t! His checks are useless! I completely missed this remarkable idea, and kudos if you found it.

So yeah, queen vs. two rooks is not an easy imbalance to play… What’s the conclusion? I guess it is not to underestimate the queen – in their raw form, two rooks are better than a queen, but in many other situations they are not.


Upsets Galore at the World Cup

Midway through the quarterfinals at the World Cup in Tbilisi, only three of the world’s top ten players remain in the circuit, and Magnus Carlsen isn’t even one of them.  Did we want it to happen like this? No, not really.  Did we predict such a thing?  Never.  Was something like this bound to happen?  You could say that.

So why do we always believe that all of the top players will have a deep run and that the last few rounds will provide more fireworks than a July 4th celebration?  Moreover, why do we dismiss the possibility that some (or even most) of the top seeds will be knocked out relatively early?  Now, this won’t be an article about the psychology of such predictions, but we can look at some of the reasons for why the World Cup has played out the way that it has this year.

  1. Schedule

The World Cup circuit is arguably the toughest tournament in the modern era.  In order to reach the finals, a player has to play a minimum of 12 games over a span of 18 days, which is already tiresome – and that’s only if you’re extremely lucky.  If you’re on the opposite side of the spectrum, you might have to play 54 games over a span of 18 days.  Overall, it is a lot of games to play at one time, and players would not have as much time to rest and prepare in between games.  And, of course, the entire event is about 2.5 times longer than most supertournaments these days.  Put it all together, and you have one of the toughest schedules on the planet.

  1. Stress

If you’re one of the top seeds in a tournament, especially a knockout tournament where you play relatively easy competition at first, you’re supposed to do well.  It’s as simple as that.  You’re supposed to win most of your games and keep moving on to the next round.  So, what happens when you try to force the issue too much and end up losing as a result?  You risk being closed out prematurely.  Depending on the color situation, it may be even harder to come back from a loss – just ask Anand in round 2 or Carlsen in round 3.  The stress to play well and not make mistakes can be overwhelming when combined with the fact that a top-seeded player must face it every day of the tournament.

On top of that, there are the added stakes to the entire event.  Finishing at the very top would guarantee a spot in the upcoming Candidates tournament, taking participants one step closer to the chance of playing current World Champion Magnus Carlsen in the World Championship Match.  Obviously, all players want to try and gain that opportunity, so the competitiveness in regards to the stakes definitely adds to the pressure on the top seeds.

  1. Time Control

The last point examined how trying to win too early in a match can just as quickly backfire.  However, there’s also the flip side to that – not winning enough early can lead to muddied waters in tiebreaks.  The lower-rated opponent may play exceptionally well in faster time controls, or perhaps the top-seeded player may not play as well in the faster time controls.  In general, games in faster time controls are much more “up in the air” in terms of possible results, and one blunder in a blitz game may be enough to knock someone out.

We looked at a few possible explanations for the struggles of some of the top seeded players in this year’s edition of the World Cup.  While there are much more possible reasons, especially, “it just happened,” these were some that definitely could have played a factor.  Looking forward, we have the chance of seeing an Aronian-MVL or Aronian-Svidler semifinal matchup on one side of the bracket, and a possible So-Liren semifinal matchup on the other side.  While anything could happen, the chess world can be excited for the fireworks to come, no matter who moves on.  And, as always, thanks for reading and I’ll see you next time.

Strategic Concepts: The Wedge on d6

Tabling for Pitt Chess Club at the Peterson Events Center

Prior to flying to St Louis for the Rapid and Blitz, I took a quick road trip to the Cleveland Open where I had a somewhat lackluster performance. I got reasonable positions out of each game, but my form was simply off and the momentum I built in Columbus simply didn’t translate into results.

With the fall term starting and my rating bouncing between 2130 and 2160, the road to National Master seemed to get getting longer, not shorter. With one game at the Pittsburgh Chess Club left for the summer, I decided to put all this stress aside and just play chess – a mindset I’ve brought up several times here on Chess^Summit.  How did I do this with the White pieces? 1. e4!

Chess with a New Face

Drawing inspiration from my final round in Reykjavik last April, I opted for Fischer’s “best-by-test” move and forced myself to think from move 1. I actually won a really nice game (which I will attach below), but I wanted to focus on a critical moment where I executed a strategic idea I had studied from the recent FIDE Grand Prix in Geneva:

Screen Shot 2017-09-05 at 13.09.27
Steincamp–Kostyak, position after 10…Be7

After ten moves, White is clearly better. I have an extra pawn, and I’ve managed to kick Black’s knight to a5 where it has no mobility. When my opponent played 10…Be7, I immediately wanted to break open the position and stop Black from castling with 11. d4, with ideas of dropping a knight on d6 and wreaking havoc in the center. However, Black’s superior development makes it near impossible to breakthrough – in fact, in some variations, Black is even better if White takes too many risks (thanks to the White king still being on e1).

It didn’t take me long to realize that Black has some compensation in his development for the pawn – not enough for equality, but certainly enough to stay in the game. Needing to catch up in development, I played 11. 0-0 0-0 12. d6!. Here’s an excerpt from my game notes I wrote after the game concluded:

“20 minute think – This was a hard move to make because this pawn does look weak and it lets the a5 knight back into the game [via c6]. The problem for White is that despite the material advantage, I am lacking in development, so a slower approach lets Black back into the game. This move is aimed to help me catch up and activate my pieces. I wasn’t 100% sure I was keeping the pawn, but the amount of tempi it would take to win it should leave Black with a worse position”

Screen Shot 2017-09-05 at 13.22.35.png
Steincamp-Kostyak, position after 12. d6!

To summarize, I’m basically throwing a wrench into Black’s position. While Black tries to play around this pawn, I’ll get the time I need to create a harmonious set-up. This is an important move in the game because it stops Black’s plan of …f7-f5, and without this, Black really doesn’t have much. After 12…Bf6, the position turns static, where I hold an extra pawn and Black has two misplaced pieces – the knight on a5 and the bishop on f6. Much to my satisfaction, this d6 pawn not only helped me develop, but also helped me conduct an attack in the center and on the kingside. You can play through the game in full here with my notes and analysis.

Pondering this 11. 0-0, 12. d6 idea, Photo Credit: Franklin Chen

Theoretical Notes

So what is it about this d5-d6 push that makes it so powerful? Surely in principle this is a hyper-extension! Let’s not rush to conclusions. The idea revolves around Black’s ability to make a blockade. Here’s a game of mine from the Columbus Open last June:

Screen Shot 2017-09-05 at 16.12.52
Wright–Steincamp, position after 16…Nd6

Here I just played 16…Nd6 following the standard knight blockade idea in the King’s Indian. Even though my knight’s primary function is defensive, it hits a lot of critical squares (c4, b5), and thus is an active piece. This knight is a strong enough piece that my higher rated opponent decided to play 17. Nxd6 cxd6 18. Rfd1, and with the center closed, the game eventually petered out to a draw.

There are two distinct differences between a pawn on d5 and one on d6:

  1. A Black knight on d6 is an active blockader. From behind the d5 pawn, it can reach c4 and e4, and pressure various points in White’s camp.
  2. A Black knight on d6 can be supported by a pawn, whereas a knight on d7 cannot. This is an important distinction, because when White has a pawn on d6, he can play to undermine the blockade. In the example above, we saw that the pawn on d5 was neutralized after 17…cxd6.

If we return to my game where I played 12. d6!, we quickly see that the knight on d7 is a poor blockader and can’t contribute to the central fight:

Screen Shot 2017-09-05 at 13.22.35
Steincamp-Kostyak, position after 12. d6!

Black is by no means obligated to blockade this pawn, but if he isn’t careful, the threat of promotion can become overwhelming – thus is the power of a pawn on d6. Of course, I can’t take credit for this strategic idea, and for that I have to thank my modern predecessor, Peter Svidler, for bringing it to my attention.

Svidler Pushes Delroy

As I mentioned, I was inspired to push my d-pawn from an earlier example during the FIDE Grand Prix in Geneva earlier this summer. Peter Svidler, in a last round clash with former Women’s World Champ Hou Yifan, pushed the d-pawn and got a nice win:

Screen Shot 2017-09-05 at 23.25.47
Svidler–Hou Yifan, position after 16. d6!

In this game, Hou Yifan opted not to create a blockade, and was swiftly punished with a further d6-d7 push, as her rook was tied down to d8. Even though the computer evaluates the above position as equal, Svidler showed just how easy it was to smash through Black’s defensive resources in this win.

I watched the commentary for this game live, and I distinctly remember GM Evgeny Miroshnichenko saying “this is the way to punish these positions”, and that just stuck with me…

Beat the Youngster with Old School Chess!

With the World Cup starting last Sunday, I’ve found myself spoiled for choice when it comes to analyzing games. One fixture that caught my attention was 2016 World Champion Challenger Sergey Karjakin against International Master Anton Smirnov. Even though Smirnov held his own at the Match of the Millennials earlier this summer, I would not have predicted that Karjakin would need tiebreaks to outlast the youngster.

In their first tiebreak, Smirnov had out-prepared Karjakin from the Black side of a Petrov and established equality out of the opening with a massive time advantage. Surely  Smirnov could keep the course and head into the next tiebreaker on an even score, right? Unfortunately, the youngster from Australia got tricked by a mirage when he materialistically played 23…Bxf3?:

Screen Shot 2017-09-05 at 23.47.13
Karjakin–Smirnov, position after 23…Bxf3?

Thinking he could snack on h3, Smirnov must have been surprised after 24. gxf3 Qxh3 25. d5! where he realized this pawn was going to d6, restricting Black’s pieces in the center. Even with Karjakin’s damaged kingside, this proved too strong for Smirnov, and he went down in the endgame. I really like this example, because it really showcases the strength of the central passed pawn. Again, the computer gives 23…Bxf3 an equal evaluation, but just like Svidler–Hou Yifan, we are starting to see we need to treat these positions differently.


Of course pushing your pawn all the way to d6 alone won’t win you games, but it is certainly a viable option when considering how to restrict your opponent’s counterplay. Its important to notice how in each of these games, the pawn on d6 acts as a wedge, and makes it easier to play in both the center and on the flanks. So next time the opportunity presents itself, strongly consider d5-d6! (or …d4-d3!) – it might just win you the game!

NY State Championship

It’s Labor Day, and I’m sitting in a hotel lobby in Albany. I’ve tied for first at the NY State Championship. In the past NY recognized all the winners as state co-champions. Not anymore; there is only that much space on the trophy. Whoever has the best tiebreaks gets the state title. That means I have to wait for every single game in the Open section to finish. Last year, I waited until the end and was second by half a point on tiebreaks losing to IM Alex Ostrovskiy. Hopefully, I’ll have better luck this year.

NYS trophy

Will my name get carved onto this trophy?

Long story short, I have a few hours to kill and an article due tomorrow.

The NY State Championship seems to be my annual redemption tournament; after botching things up towards the end of summer, I get my revenge and some of my rating points back…

So how did it go this year?

This year’s field was much stronger than last year’s field. Last year there was only one GM, this year there were three… Tying for first did not seem like an easy task at all.

In round 1, I got black against Abhimanyu Banerjee (2155 USCF). It was a fairly smooth victory, though there was one unusual moment…


Black to move

Black’s queen and knight are both under attack, and one will fall. Instead of just moving my queen out of the way, I decided to go 18… Ne1!? after a long think. The point is that if 19.Rxc2 Nxc2 20.Rb1 Nxd4, black has a rook a piece and a pawn for the queen. Still, I think white should be OK there, though nothing more. Instead, my opponent played 19.Qf1? after which I soon won the d4-pawn, and white’s position fell apart.

In round 2, I got white against Steven Taylor (2117 USCF). That game was also a pretty smooth victory where I basically got a winning position out of the opening and managed not to botch it up xD.

So far, so good!

In round 3, I got black against FM Ethan Li (2360 USCF). The game was a fairly quick draw; I got a little worse out of the opening but never let Ethan get anything serious. Drawing this game was not a big deal; the rating difference wasn’t so large and besides, I was still tied for first (it was a 9-way (!!!) tie at that point).

In round 4, I got white against GM Sergei Azarov (2643 USCF), and the game was a quick 15-move draw. He offered a repetition out of the opening, and I decided to take it instead of playing on. Was it a good idea? In retrospect, it probably was. I was half-a-point behind the four leaders, so I still had a good shot.

In round 5, I got black against Jacob Chen (2226 USCF). Things went very well for me out of the opening, and I was just better with black. A nice little trick netted a pawn, but then I had to convert it. Jacob decided to give me a second pawn to get into a rook endgame where it wasn’t so clear if I was winning.


Black to move

What’s going on here? Black is two pawns up, but those are doubled h-pawns. That’s inconvenient. What’s black’s winning plan? I wasn’t quite sure what exactly it was, but I knew I had to try to create a passed e-pawn. How to create the e-pawn? Well, I want white to push f2-f3, so that I can create the passer on e4 instead of all the way down on e2.

First of all, I want to keep the h4-pawn on the board. The game went 29… Ra4 30.Kh3 Rf4 31.f3 e5 32.Rb5 f6 33.Ra5 Kg6 34.Rb5


Black to move

I’ve advanced my pawns, but what to do now? Black can’t go f5, and my rook appears to need to babysit the h4-pawn. Therefore, I decided that the next order of business was to defend the h4-pawn with my king to free the rook. I went 34… h6 to prevent white from being able to easily attack it. The game went 35.Rb6 Kg5 36.Rb5 Kh5 37.Rb6 Kg5 (repeating once) 38.Rb5 Ra4! 39.Rc5 Ra1 40.Kh2 Re1


White to move

Black has made a lot of progress! His rook has gotten active, and the white king is confined. Now for the e-pawn push…

Jacob decided to go active with 41.Rc8 f5 42.Rg8+ Kf4 43.Rh8 but it’s too late.


Black to move

Can you find the knockout blow for black? Here’s how the game ended.

All in all, I’m not sure the endgame was objectively winning, but the game looks fairly convincing. Still, I think white could have defended better somehow.

Going into the last round, it was time to take a look at the tournament situation. GMs Mark Paragua and Bryan Smith had 4.5/5 and were playing each other. GM Sergei Azarov and I were the only players with 4/5. Since we had already played each other, we got bumped down to play the 3.5-pointers.

I got white against IM Jay Bonin (2361 USCF), who was at 3.5/6. IM Bonin tried to create some chaos, but it backfired. I got a near-winning position, which I won without any real problems.

The results are in.

Back to Albany. Both GM Paragua drew and GM Azarov drew; that puts me in a 3-way tie for first. GM Bryan Smith lives in Pennsylvania, so he’s not competing for the state title. It’s only me and GM Mark Paragua. There is only one game going on, and the tiebreaks will be the same no matter the result. They are calculating them…

First tiebreak: the same!
Second tiebreak: the same!
Third tiebreak: well, have a look!

NYS Final Standings

GM Paragua wins! It’s no big deal. I still had a good tournament and a good result. I’ve redeemed myself, like I do year after year the NY State Championship.

Plus it’s obvious that I am improving! Last year I lost on the 1st tiebreak. This year, it took 3 tiebreaks. Next year, I plan on having them all even and getting NYS recognize all the winners as NYS Co-Champions :).

P.S. My opponents who lost the last round are on probation. Just kidding…

My Toughest Tournament Ever

Norm tournaments are mostly about present and future IMs/GMs, but I thought it would be interesting to see what it’s like for a random master to play a big norm tournament like the U.S. Masters. Long story short, it’s really hard – evidently, I can only handle about half a tournament (Meanwhile, David plows through 10 of these things in a year.).

Admittedly, it seemed ridiculous given my recent results that I should jump right into the strongest edition of one of the strongest tournaments in the country. But the tournament has been on my bucket list since I became a master, and I ultimately had little choice in the matter, having locked myself in on the whim of a Chess.com Titled Tuesday event.

Things didn’t look any easier on site, where I found myself the 4th-lowest seed in a field of IMs, GMs, and norm hopefuls. But to my amazement, I held my own in this field, defeating two IMs in the first four rounds! Unfortunately, I learned the hard way that these events are marathons and not sprints, but from the beginning this was meant to be for the experience than anything else, so there was a lot to learn and a lot of memorable moments. Aside from chess, it was good to see David and Vanessa again!

After earning the National Master title in April, I found it hard to stay motivated given my busy school and summer work life and the lack of uncertainty over what I wanted to achieve. The few times I did play chess were disasters; for example, in my last tournament before the Masters, I had to fight uphill to draw both an 1800 and 1500, and somehow managed to mess up a R+2 vs. R ending.

At this point, I wasn’t going for anything in particular other than staying solid as much as possible against whoever came my way. There was no warmup period, as I was immediately dealt Black against IM and 2-time U.S. Open champion Michael Mulyar. At first, the game looked like what you’d expect of this type of matchup:

Screen Shot 2017-08-30 at 12.21.00 AM
Mulyar – Li: After 20…Qc7

I’d played the opening reasonably, but after wasting time with dubious knight maneuvers, I found myself facing an imminent e4-e5. To make matters worse, I was down 40 minutes on the clock.

White surprised me by playing 21. f5!?. 21…f6 was the only “permanent” answer, but that would invite an eventual Ne6, which looked extremely unpleasant. Unsure of what to do, I stalled with 21…cxd5, which was immediately met by 22. f6!?. Clearly, moving anything on the kingside is a disaster, so 22…dxe4 followed. Instead of simply playing 23. Ng5 with fxg7 soon to follow, White attempted the more immediate 23. Qg5, and after the forced 23…Ne6 24. Rxe4 Nac5, erred with 25. Rfe1? Nxe4 26. Rxe4.

Screen Shot 2017-08-30 at 12.46.46 AM

I’m guessing White just overlooked 26…h6! (probably the simplest way to resolve matters), and after 27. Qh4 Qc5 28. Qg3 g6 29. Kh1 Bd7 30. Rh4 Qe3 31. Ng4 Qc1+ 32. Kh2 h5, I was basically out of the woods with an extra Exchange. Still, even though Black is completely winning, this is completely possible to mess up, especially with only 5 minutes to make move 40. Fortunately, I closed out the win without too much trouble.

Beating an IM for the first time was a huge accomplishment for me, but looking deeper, I didn’t do anything too fancy here; I just tried to make the most of a bad position and took the chances I got. This is not to say that you can just simply for chances and win, but even very strong opponents make simple-looking mistakes.

On the other hand, I was definitely a little too confident going into the next two rounds, even though my opponents were better than almost everyone I’ve ever played. Since I had White in both games, I hoped I could sustain a solid position without too much effort. Alas, there’s a lot more to the chess than that, and I was soundly outplayed by IM Guillermo Vazquez and Deepak Aaron, a strong master rated over 2300 FIDE.

One of the more questionable decisions could have been avoided pretty easily:

Screen Shot 2017-08-30 at 1.22.59 AM
Li – Aaron: after 17…Rg8

The most straightforward continuation is 18. Qf2, even if it requires Ke1. I believe White should be somewhat better with the king relatively safe and the h6-pawn likely to fall.

Instead, I charged ahead with 18. Qxh6?I’m not sure how to explain why I chose this. I even calculated everything pretty much correctly; I just completely misjudged the resulting position. After 18…Bg5+ 19. Qxg5 Rxg5 20. Rh8+ Ke7 21. Rxa8 Bc8 trapping my rook, I assumed playing g4/Ng3-f5+ would be enough compensation for losing the Exchange. But after 22. g4 b6 23. Ng3 Qb7 24. Rxc8 Qxc8 25. Nf5+ Rxf5! 26. exf5 Qh8, Black’s queen proved too powerful due to my weak f3 and c2 pawns.

Screen Shot 2017-08-30 at 1.28.55 AM
Li – Aaron: after 26…Qh8

Overconfidence with White would prove to be a recurring problem – I didn’t score with White for the whole tournament, even against lower rated players (more on that later).

With a 1/3 score, I had to wonder if I was headed for a bye as one of the lowest seeds (not a disaster by any means, but I wanted to play as many games as possible and there was little else to do around the area). It’s not often that someone at my level finds themselves in a must-win situation as Black against an IM, but that seemed to be the case here!

Since my opponent was a 1. Nf3 player, I decided it was important to try some preparation – otherwise, I’d be simply trying to match a much stronger player. As I’ve mentioned before, I’m not the best at openings or preparation – in fact, I had never successfully prepared for a specific opponent before. But I noticed my opponent had played a pretty specific line against two GMs before, who had both gotten great positions due to a temporary pawn sacrifice. I also didn’t think my opponent would suspect I’d prepped for him (mostly due to my strength), so I gave it a try. Long story short, it worked great!

Screen Shot 2017-08-30 at 4.31.58 PM
Saravanan – Li: after 8…c6

White deviated from the games I’d looked at, playing 9. Nb3 instead of 9. Nc2. This was probably an improvement, but I suspected that the GMs would not have played 8…c6 if it was just hope chess. This gave me the confidence to play 9…d5 anyway, and the game continued similarly with 10. cxd5 cxd5 11. exd5 Qb6 12. Qd4 Nbd7.

Screen Shot 2017-08-30 at 4.36.34 PM
Saravanan – Li: after 12…Nbd7

This is when I realized the point of 9. Nb3: protecting b2 and giving White more options, e.g. 13. Bg5. Black has good compensation for the pawn with a powerful dark-squared bishop and White’s king still in the center, but I didn’t see how exactly I would get the pawn back and White has ideas of Nb5.

So I played 13…Qxd4 14. Nxd4 Nb6 15. d6 Nfd5, intending to open up the a1-h8 diagonal and develop my other bishop as soon as possible. After 16. Nxd5 Nxd5 White had a choice between the game move 17. Rd1 and 17. O-O-O. I think White missed a chance here even though it looks like the e2-bishop would be hanging, since the d5-knight is loose and White has possible Bb5 and Bc4 counterattacks.

Screen Shot 2017-08-30 at 4.46.28 PM
Saravanan – Li: after 17. Rd1

Instead, after 17…h6, 18. Bc1 was forced and I played 18…Rd8 intending 19. Nb5 Be6 20. Bc4 Rac8. But instead of simply blocking off the c-file with 21. b3 (saving the extra d6-pawn for the moment) White tried to force matters with 21. Bxd5?! Bxd5 22. O-O?

Screen Shot 2017-08-30 at 4.51.31 PM
Saravanan – Li: after 22. O-O

21. Bxd5 already makes White’s position fairly uncomfortable, and 22. O-O is simply squashed by 22…Bc4; White resigned a few moves later.

Unfortunately, the overconfidence with White kicked in again when I played FM Hans Niemann; I overextended quickly and blundered a pawn on move 15, and later the Exchange while trying to win the pawn back. Although I hadn’t been favored to win any of my games as White, I definitely felt like I could have put up more resistance in those games.

My next game against WFM Apurva Virkud was only marginally better, as I essentially played into a bad version of the Bogo-Indian and got pretty passive early on.

Screen Shot 2017-08-30 at 10.26.39 PM
Virkud – Li: after 5…Nc6

In this position, I expected White to transpose into the mainline Bogo-Indian with 6. Nf3 (having opened with 3. g3 instead of 3. Nf3), where 6…Bxd2+ 7. Qxd2?! is met with 7…Ne4 8. Qc2 Qb4+. Instead, I was surprised by 6. e3! giving White the ideal setup after 6…Bxd2+ (necessary to play …d6/…e5) 7. Qxd2 d6 8. Nc3. Soon after 8…e5 9. d5 Nb8 10. Nge2 a5?! 11. Nb5, I got tied down defending c7. Later, I blundered in an already difficult position and lost soon after.

In the next round, I bailed with a 9-move draw against NM Sam Copeland after accepting an interesting gambit in the Two Knights Caro-Kann. Early draws are admittedly not something I would encourage in general, but since he wasn’t feeling great and I was having second thoughts about the line, I guess we thought it would be prudent to call it off early. Furthermore, I wanted to save energy for my last round (I had to skip Round 9 to fly back to Pittsburgh) especially as White, and I think that was the right decision.

Unfortunately, even though my opponent was the lowest-rated (although by no means a weak player) I’d faced all weekend, I managed to overextend on the kingside in an Exchange Berlin and ultimately did not come close to winning. This was disappointing, although it goes to show how each game is really different; I definitely relied on my intuition more, perhaps due to beating strong players early on. This is definitely a game I’ll have to analyze more, but I’ll show the (effective) ending, since it was pretty surprising.

Screen Shot 2017-09-02 at 1.47.39 AM

24…Bxd3! 25. Qd1 (otherwise White will be down at least 2 pawns) 25…Rxd4! 26. cxd4 Bxd4+ 27. Kh2 Qxb2+ and I played on a bit longer than I should have, but with best play White is down too many pawns for the Exchange.

That’s a lot to take in for one’s first norm tournament. I wish I could have been a bit more consistent throughout, but for someone just trying to make some noise in the tournament, I am pretty satisfied with what I got out of it. In fact, my lifetime record against International Masters is now a curious 2.5 – 2.5 (with all my scores coming as Black).

Last but not least, I’d like to thank the tournament staff, especially organizer Walter High, for such a strong and smoothly run tournament. I definitely hope to be back someday!

IM Kostya’s Predictions For the 2017 World Cup

Like most chess fans one of my absolute favorite tournaments to follow is the FIDE World Cup! The upsets, the tiebreaks, the must-win situations, the armaggedon, it’s truly one of the most exciting events in chess. This year seems especially hard to predict as the tournament is stronger than most years, with most of the top-10 playing, including a rare sighting of Magnus Carlsen & Viswanathan Anand!

If you’d like, you can check out my full bracket here. I typically went with experience wherever I could, and picked only a few upsets as I feel like many of the favorites are going to bring their A-game. Especially Magnus — assuming he’s in good health I just don’t see anyone getting in his way to reach the top. I mean when was the last time Magnus even lost a tiebreak match?

A few highlights:

  • I think Magnus vs. MVL is nearly a lock for the Quarterfinals. Both players are exceptionally strong in rapid, and only Grischuk may pose problems for MVL.
  • I have Andreikin defeating Aronian in Round 3, but I’ll be rooting for Levon regardless. In fact, this could easily be his year.
  • Although I picked Ding Liren to defeat Kovalenko in Round 2, this could easily be a big upset, as Kovalenko crossed 2700 previously and is known to be an excellent rapid/blitz player.
  • A likely Round 3 match-up between Kramnik and Ivanchuk will likely bring the house down, I almost picked ‘Chucky’ here but Kramnik tends to bring it during the World Cup!
  • It was hard to choose Nakamura over both So and Caruana, but he is the better rapid/blitz player and when he’s in form, he seems unstoppable.

And here are my answers to the Chess^Summit Sweepstakes:

Who wins the 2017 FIDE World Cup? 
Magnus Carlsen

Who is the Runner Up?
Hikaru Nakamura

Which American player goes the farthest?
Hikaru Nakamura. Caruana and So are obviously reasonable picks here as well.

Which Russian player goes the farthest?
Sergey Karjakin. This is a bit of a hedge, since I tend to underestimate Karjakin in most events, but he seems to have an iron grip on doing well in World Cups. It was very hard to choose him over Kramnik!

Which Chinese player goes the farthest?
Ding Liren. Wei Yi or Li Chao could easily take this spot, but I feel that Ding is the most solid of the group. Yu Yangyi is another contender.

Do all the top 8 seeds win 2-0 in the first round?
Yes. I mean maybe they won’t, but I have a feeling we’re going to see some very determined players at the top.

Which player will score the most draws?
Viswanathan Anand. I’m really not sure how Anand will perform here, but it seems likely that he’s going to be risk-averse and take his chances as they come in the rapid/blitz.

Which player will score the most wins?
Maxime Vachier-Lagrave. One of the hottest players right now, I can see MVL being in a lot of must-win situations and pulling them out!.

Top rated player below 2700?
Ruslan Ponomariov. A great value, Pono has been 2700+ for most of his career and is a fantastic rapid player. Look for him to potentially make a deep run. Although I picked Harikrishna to advance after their encounter in Round 2, I wouldn’t be surprised to see Pono pull the upset.

Top Junior
Wei Yi. He seems like a heavy favorite, since other juniors will have to play up in Round 3 (or before) while he plays down until Round 4.

Do any former World Champions qualify for Candidates 2018 through the World Cup?
No. Although Kramnik is one of the favorites and always performs well, somehow I doubt it with So, Caruana, Nakamura, MVL, Aronian, Karjakin and of course Carlsen all in the mix.

Does Carlsen make the top 4?

Enter the sweepstakes now with your 2017 World Cup predictions!

A Look Back at the 2017 U.S. Girls’ Junior Championship

Last month, I competed in the 2017 U.S. Girls’ Junior Championship from July 7-18 at the Chess Club and Scholastic Center of Saint Louis (CCSCSL). This was my fourth time participating in this tournament, so I was one of the few returning veterans looking to win. Going into the tournament, I had high expectations to win to earn a chance to go to the 2018 U.S. Women’s Championship and 2018 World Girls’ Junior Championship. I had come very close to winning the U.S. Girls’ Junior Championship before, but I fell short due to heartbreaking moments. In 2014, I ended up tying for first with 5.5/9, but I got third on tiebreaks. In 2015, I had to settle for third place with 5.0/9 after losing a won position in the last round. Then in 2016, I had my worst performance by only scoring 4.5/9 points and getting 5th place. This was why I wanted to redeem myself at the 2017 U.S. Girls’ Junior Championship.

The 2017 U.S. Girls’ Junior Championship was an invitational only tournament between the top 10 girls under 20. I started the tournament as a second seed and was one of the favorites picked to win the tournament. Unfortunately, I wasn’t able to win the tournament, mainly due to missing a lot of chances. Nevertheless, I was still able to finish in clear second which is always a good result. Now, let’s take a round by round look at my games at the 2017 U.S. Girls’ Junior Championship

Round 1

In the first round, I played one of my well-known friends. We have played each other a number of times, so we were very familiar with each other. In the following game, we reached this important position.


White to move

In this position, I am clearly better with my better placed pieces and more space. I just played 25.Nh4 with the idea of of attacking the Bishop on g6 and having plans of f4 sometimes. My opponent responded 25…Ne5 which was a good move to centralize her pieces. My next move was not the best, and it was the move that turned the game around. I should have played a move like 26.Kg2 to improve the king and have ideas of f4. Instead, I played 26.Bxe5?!. It’s not necessarily a blunder, but it’s also a too forcing move. In better positions, you don’t need to force the position. Instead, you can slowly improve all your pieces and win more slowly and smoothly.  My opponent responded with 26…Rxe5, and the game continued 27.Nxg6 hxg6 28.Qg3 g5 29.Kg2 g6 30.h4 Kg7 31.h5 Rh8 32.Rh1 gxh5 33.gxh5 


We reached this opposite color bishops position, with my advantage slowly disappearing. The game had become more complex with the opposite colored bishops than the position I had before. Later, I would blunder into a worse position, but I was able to escape with a draw, luckily.

Round 2

In this round, I also played a well-known friend, and we played each other a lot too. Our game featured a long opening variation with a lot of heavy theory, and, unfortunately, I ended up on the wrong side of the theory. I was the one that didn’t know the opening well, so I tried my best to play the right moves at the board and I did alright. We reached the following position.

Black to move

I’m up a pawn in this position, but my king is a little bit weaker. With 21.Bd2, White wanted to have ideas to play g4 to get rid of my strong knight on f5 and open up the king. I didn’t want to give up my h4 pawn that easily since the h4 pawn closes down the h-file, so I played the interesting move 21…g5. This move exposes my king more, but I thought that I could bring my queen over, and my king would still be protected. The game continued 22.Qg4 Ng7 23.Ne2 and I played 23…Qg6 sacrificing the h4 pawn. I just wanted to trade queens so my king wouldn’t be attacked anymore. The game would peter out to a bishop of opposite color endgame and we drew quickly.

Round 3

I scored my first win this round, although it was kind of lucky. I had white against the number one seed. We played a fairly even game, but I had slight pressure on my opponent. Then, my opponent made a very strange decision giving me a free pawn and a huge advantage. Take a look at the following position.


In this position my opponent played the horrible blunder 26…Bc6?? which gives me the f7 pawn. Every one makes these moves once in a while, usually when we’re really tired. Anyway, after this blunder I won the game fairly easily.

Round 4

This was a special day for me. It was my birthday and I was turning 17! A big shout out to the CCSCSL for wishing me happy birthday on that day! Anyway, in this game I reached a slightly better position in the endgame. However, I didn’t push very hard in the position and ended up settling for a draw. The ending position was still slightly better for me, but I thought it looked like a draw, so I didn’t continue playing. This was one of the games where I could have gotten a half point more.

Round 5

This was my best game of the tournament. I was playing the lowest rated player as white, but I dominated the game after my opponent made a few mistakes.

Can you find the nice tactic here that finished the game immediately?

White to move

Round 6

I played the reigning 2016 U.S. Girls’ Junior Champion this round. In last years tournament, I suffered an early loss to her, so I wanted to redeem myself this year. This game was evenly balanced from the beginning, and I didn’t have huge chances to gain an advantage. I also didn’t take much risks. We agreed to draw on move 33.

Round 7

This was an important round as I was only one point behind the leader. It was essentially a must win game to try to catch up to the leader. I was white against a strong opponent that I have played many times before. Here is the first important moment in my game.

Black to move

In this position, I just played 27.Rab1 which is a mistake. I missed a nice tactic that my opponent also missed too. Can you find it?



Black should play 27…Nxg3+!! (27…Rxe5!! works too) 28.hxg3 Rxe5! threatening mate on h5 29.fxe5 Bxf1 30.Qxc5 Bxg2+ 31.Kxg2 Rxc5 and black is up 2 pawns

Instead the game continued 27…Rb6? 28.Qxa7 f6 29.Ba1 Ra6 30.Qb7 Bc4 31.a4 Rac6 32.Rfe1 Ba6 33.Qb2


In time pressure, my opponent made a big mistake and played 33…d4?? This opened the door for me to play 34.Rxe2 Bxe2 35.Qxe2 giving me two bishops for the rook, although my bishop on a1 was badly placed. After maneuvering my pieces around to better squares I was able to win the d4 pawn. Then we reached this position.

White to move and win


This position was completely winning. I just had to use the fact that black’s king had no defenders, and I could attack it. I missed the winning move here as I opted for the move 49.Bc3? Black played 49…Rxd3 50.Bxb4 Rxd2+ 51.Bxd2 Rxa4. This is a theoretical drawn endgame of 2 pieces vs a rook with pawns on the same side. I knew about this endgame, but I wanted to see if my opponent would make a mistake. I tried to make progress in the endgame, but my opponent defended well and we agreed to a draw.  The winning move in the above diagram was 49.Qe3! This was a relatively simple move as it moves the queen out of the pin and threatens Qe7+. Black can’t take the pawn with 49…Rxa4 because of 50.Qe7+ Kg8 51.Rc2 threatening Rc8#.

Round 8

I played the leader as black this round, and I was down one and a half points. This was a must win game for me if I wanted to contend for first place. I approached this game like a regular game, and I didn’t put too much pressure on myself to win. I just wanted to play good chess. Unfortunately, I didn’t play good chess, and I got a worse position out of the opening by mixing up my opening moves.


This position is terrible. All of my pieces are horribly placed, and my bishop on c8 and both rooks will not see life for a long time. I fought hard and played good moves afterwards, but it was too late. My opponent only needed a draw, and she got a draw rather easily. I wish I could have played a better game, but everyone has their bad days, unfortunately.

Round 9

We already knew the first place winner before this round, so I was fighting for clear second place and $2000. I was white this game, and I walked into some nasty preparation from my well-prepared opponent. At one moment, my opponent had more than 90 minutes while I had to spent 30 minutes navigating the opening. After the opening, I was not in the good position, so I decided to simplify into the drawn rook endgame. We drew quickly then.


I ended up scoring 5.5/9 with 2 wins, 7 draws, and 0 losses. I got clear second, a nice trophy, and $2000. Overall, this was a good experience and fun tournament for me. I could have done better, but I’m still happy with my result. Finally, I want to thank the Chess Club and Scholastic Center of Saint Louis for organizing a top notch tournament in the chess capital of the U.S., and I also want to thank everyone for taking the time to read my first chess article here at chess-summit!