Foxwoods: A Near Miss Part 1

Last year, at the last round of the traditional (for me) Philadelphia Open, it was announced that the tournament was moving to the Foxwoods Casino/Resort the next year. Before I knew it, it was Easter time again, and it was time to go play a strong 9 round norm tournament—this time in Foxwoods. My Easter tournaments have ranged from truly disastrous (aka the first time I ever withdrew from a tournament) to highly memorable. After all 2 years ago, I got my third IM norm at the Philadelphia Open, and had I won the last round I would have walked away with a GM norm to top it off. Let’s see where this year fits in…

Playing in the world’s 4th largest casino definitely added an interesting feel to the tournament as opposed to your average hotel. It was definitely a strange feeling when people in the elevator wished each other good luck—not in chess but in gambling. I stayed in an adjacent tower, and it took about five minutes of fast walking to reach the tournament hall from the hotel rooms, and that was only one small part of the resort. Who needs a gym to stay in shape when you have Foxwoods?

The tournament area was separate from the casino and provided a peaceful refuge from the crowds outside. There was, however, one special thing about the play room which we found out in round 1—it was located below a bowling alley! Imagine balls rolling and pins falling… The admittedly intermittent noise was not very loud but extremely annoying. I have to give credit to the organizers for switching the play room with the directors’ room, so that in round 2 there was much less noise, and in round 3, after they moved us even deeper into the room, there was no noise at all.

Ok, now it’s time for chess!

In round 1, I got white against Mardon Yakubov (2128 FIDE, 2159 USCF) It wasn’t the greatest game… I butchered a large advantage out of the opening, and my opponent defended well to reach this position:

Yakubov 1

Black is a pawn up, but he’s under fire. With my last move 21.f4, I was naturally trying to break down black’s center. 21… exf4 loses on the spot to 22.Ne6+, and 21… e4 runs into 22.Rxe4. Black’s best option here is to get his king out of the way with 21… Kg7!, after which white has nothing better than 22.fxe5 fxe5 23.Qe3. This wins the e5-pawn back after 23… Nf5 24.Qxe5+ Qxe5 25.Rxe5, but white’s advantage is minimal if at all existent after 25… Rhe8.

Instead of that, however, my opponent played 21… Rc7?. The idea of this move was most likely to prevent my threat of fxe5 fxe5 Rxe5 (it stops the Nd7 fork), but it gives white a chance to pounce. After 22.fxe5 fxe5 I saw that I could still play 23.Qe3—after 23… Nf5 white can still play 24.Qxe5 because after 24… Qxc5+? 25.Kh1 black is losing a rook or is getting mated. However, after 24… Qxe5 25.Rxe5 Re7, he’s not in such bad shape. Instead of doing that, I played another move which was much stronger: 23.Qa5! threatening Qxc7 Qxc7 Ne6+ winning a rook. Black’s best chance was to play 23… Kg8, but after 24.Ne4 Qd8 25.Ng5, he’s in really bad shape. My opponent instead played 23… Rc8? but after 24.Rxe5! I’m crashing through. I won a few moves later.

Not too bad for a first round I guess. In round 2, I had black against William Sedlar (2217 FIDE, 2411 USCF) and this time professional swindling was required

Sedlar 1

I had been doing fine previously, but then a silly mistake got me into a worse position. Fortunately, I wriggled my way out, and by the time we reached this position, I thought I was doing fine. Material is equal and fairly reduced, and while black’s e-pawn is under fire, black has plenty of activity to compensate for that. White could play 40.Rdxe5 (or 40.Rexe5) immediately. I saw that I had at least a draw with 40… Rxe5 41.Rxe5 Qc1+ 42.Kh2 Qf4+ 43.Kg1 Qxa4, but I wasn’t sure there was anything more, and my engine confirms that it’s indeed a draw. For a human, however, it’s not clearly obvious that there is nothing for black. Instead of taking on e5, white’s best move is actually to play 40.Kh2! simply getting off the first rank. Black doesn’t have anything that concrete, but I think that with reasonable play he should hold a draw without any real problems.

My opponent instead played 40.Rd1 and offered a draw. While this position is objectively equal, I didn’t see myself losing this one and wanted to try a little… The game went 40… Rf2 41.Qc4 Qb6 42.Qb5 Qa7

Sedlar 2

Here, my opponent played the logical move 43.a5?? and ran into more or less the only trap I had in store for him: 43… Rxg2!

Sedlar 3

If 44.Kxg2, black has 44… Qf2+ 45.Kh1 Qf3+ 46.Kh2 Qxe4, after which he’s a pawn up and is on the verge of mating white. Besides that, white is just broke. My opponent tried 44.Qxe5 but after 44… Qf2 45.Rde1 Rg3, he had to give up his queen with 46.Qxg3 and resigned a move later.

2/2 so far! Admittedly it was a shaky 2/2, but I was going to take it…

In round 3, I got white against GM Zhou Jianchao (2623 FIDE, 2702 USCF) This game would’ve been exciting had it not been entirely my preparation. My opponent found all the right moves to equalize, and I decided to repeat moves as I thought I might do in my preparation. While I’m not a fan of making “nothing draws” with white, this was fairly principled and wasn’t a bad decision, especially against the #2 seed with a 2600+ FIDE rating…

In round 4, I got black against GM Kamil Dragun (2585 FIDE, 2666 USCF). I had lost to him at the Southwest Class in February, and I was looking for revenge. This game was fortunately better than the last one. I’d say that overall it was a fairly accurate draw; maybe I was slightly worse, but it was nothing really serious and I held my own.

In round 5, I got white against GM Vladimir Belous (2520 FIDE, 2621 USCF). In a nutshell, this game got spicy pretty quickly…

Belous 1

So far, this looks like a fairly normal Sicilian, but after 12… d5! it got flashy. After my move 13.exd5 I was expecting one of two options: 13… Nxd5 or 13… Bxa3.

After 13… Nxd5, white obviously can’t play 14.Nxd5 because of Qxc2#, and 14.Rc4 Bc6 is not awe-inspiring. Instead, I was planning on sacrificing an exchange with 14.Rxd5! exd5 15.Bd3, after which white has plenty of compensation—he’ll win the d5-pawn, black’s king is still in the center and will come under fire, etc. After 13… Bxa3, white should play 14.Rc4 Qa5, after which he has options: Bd4, Bd2, Kb1, dxe6, etc. In both cases, the position appears to be rather unclear.

Instead, 13… b5? came as a big surprise to me. The idea is to prevent Rc4, but will it work…? The game went 14.Bd3 Bxa3 15.Ne4 Nxe4

Belous 2

Now… once I recapture on e4, that bishop on a3 is getting evicted. Then it’ll be time to start going after black’s king! 16.Qxe4 is an exchange sacrifice after 16… Bc5, though there’s plenty of compensation there. It’s actually best to play 16.dxe6! fxe6 17.Qxe4, after which white has a massive initiative. Instead, I played 16.Rxe4 which isn’t best but isn’t bad either. Black’s best option is to retreat with 16… Bd6 or 16… Be7, but my opponent played 16… Qc3

Belous 3

After 17.bxa3 Qxd3, white can more or less resign, but fortunately I had spotted 17.Bd4! in advance. Black can play 17… Bxb2+, but after 18.Kb1 he has nothing better than 18… Qxd4 19.Rxd4 Bxd4, after which he is probably lost. My opponent tried to defend with the creative 17… Qb3!?, but black is lost after that. While my play afterwards wasn’t the most accurate, I managed to convincingly get this job done.

4/5, 4 foreigners and 3 GMs down, performance well above 2600, and a large rating gain. What could possibly go wrong…? Stay tuned for part 2!

Advertisements

The Funny Side of Chess

Chess is a great game and more often than not rather serious, but there can be cute moments. Even very cute… Who said that chess isn’t an art?

Underpromotions

Underpromotions pop up all the time in endgame studies but not that much in OTB chess. Sometimes you have an opportunity to underpromote to a piece that is equivalent to a queen (since your opponent will take it anyway), but as IM Justin Sarkar showed in his article, there are pitfalls…

My funniest underpromotion was:

Idnani

Black clearly should’ve resigned a long time ago… There are numerous mates in one and mates in two here. I decided to end in style with 47.a8B#!

Another example:

Eigen

Black is totally winning here, but 29… Re1 doesn’t work because of 30.Rxe1 fxe1Q 31.Qxf8#, right…? Well it doesn’t, since after 30.Rxe1, I played 30… fxe1N+! winning the game immediately.

These two examples are naturally somewhat silly since the side underpromoting was completely winning anyway, but there’s one underpromotion that is important in endgame theory…

Underpromotion

Black’s pawn is almost there… but if he plays … e1Q, white has Ra1#. Therefore black has to play …e1N+! because it is a check. The ensuing rook vs. knight endgame is a draw. The same trick works with an f- or a g-pawn, but not with an h-pawn, since the knight gets trapped in the corner.

If you haven’t seen this one before, I’d recommend you study it a little. I myself don’t actually think I’ve ever had this one on the board, but I or you probably will someday…

Stalemate

Oh boy, this is fun if you are on the correct side of the board, it really is.

There are legendary stalemate tricks in online rapid/blitz/bullet due to the existence of premoving (making a move of your own before your opponent makes their move in order to save time on your clock). And yes, this move is played as long as it’s legal, even if your opponent played something completely different than what you expected. You may think you know your opponent’s move, but you may be for a rude surprise.

Rosen

Here is Eric Rosen’s trick. Black is obviously completely lost in the pawn endgame after Kxf7, so black tried his last trick with 69… Kh8!!, and he was in luck since white premoved 70.h5 after which it is stalemate.

I recently saw another brilliant trick in the chess.com Bullet Championship:

Bortnyk

White is obviously completely winning here as well. Both players had only seconds left at this point, but black set a genius trap with 66…Bb8!!, and white, none other than Grischuk, fell for it by playing/premoving the natural 67.a8Q?? after which it is stalemate.

Unfortunately, these brilliant techniques don’t really work in OTB chess, though I have witnessed a couple examples of diabolic stalemates that were entertaining for spectators like me. There have even been a couple stalemate tricks in high-profile games like Jakovenko-Gelfand. But who says that stalemates have to be diabolic tricks…?

Colas

I was black in this game, and I had survived the worst. My queen had been harassing the white king for quite a while, but here black has one drawing move: 76… Qf6+!, since if white plays 77.Qxf6 it’s stalemate. My opponent played 77.Kc7, and since we had already had the position twice before, I claimed a threefold repetition after 77… Qc3+ and the game was a draw.

(Note: tabelbases say that 76… Qf6+ is not the only drawing move; 76… Qa1 is also supposedly a draw. I won’t pretend that I can explain why…)

Neat, right? In all seriousness, moves like 76… Qf6+ are easy to miss, especially towards the end of a long game, but they could save you half a point!

Any diabolical experiences?

Third Time’s the Charm: The GM Norm Endgame

After a tough stretch in the middle of the Southwest Class, I had 4/6 but still had a 2600+ FIDE performance. I have played enough GMs, titled players, and foreigners to satisfy the technical requirements. “All” that was left was good play and some luck. 2/3 or 2.5/3 in the last three rounds would get me a GM Norm. Deep breath.

In round 7, I got white against IM Zurab Javakhadze (2430 FIDE, 2504 USCF). This game was, in a nutshell, not what I was expecting…

Javakhadze 1

Up to this point, everything was relatively normal. 11… f5 doesn’t work on account of 12.exf5 Rxf5 13.g4, but black could choose to play 11… Bxf3 12.Qxf3 Nd4 13.Qd1 c6 14.Ba2 Bg5 or 11… Nd4 12.g4 Nxf3+ 13.Qxf3 c6!, after which white doesn’t have much. Instead, my opponent played the seemingly natural 11… Qd7?? running into the Puzzle Rush tactic 12.Nxe5!

Javakhadze 2

I had seen this one coming for a while. Black is losing a pawn no matter what he does. On top of that, it’s a strong central pawn, and white will also have the bishop pair afterwards. I think it’s totally reasonable to say that black is just lost here.

I can’t remember the last time I had a winning position after only 12 moves, and especially against an IM… How much more luck could go my way??

CHESS SUPERSTITIONS

If I didn’t venture into the world of chess superstitions, this article would be a scam. Before this tournament, my personal chess superstitions didn’t really go beyond the realm of lucky pens. The pen I was using this tournament hadn’t served me so well before but was now doing an excellent job. I so wasn’t going to change it. There was, however, the possibility of using my lucky shirt.

The World U16 Olympiad in November was a rough tournament for me to say the least, but there was one highlight: beating the top seeds Uzbekistan in a huge upset. I therefore used the same logic I have with lucky pens to wear the same shirt I was wearing that match for the last day, when I sure wanted a bit of luck…

In round 8, I got black against GM Kamil Dragun (2589 FIDE, 2668 USCF). I didn’t back down at all, and I’m really glad I did that. Unfortunately, my nerves weren’t the problem—my brain was…

Dragun 1

I had just scraped out of some complications—and a bit of trouble—to reach this position, which really didn’t seem that dangerous. However, white a) has the d-file, b) can get his pawns moving more easily on the kingside, and c) can possibly tie black down to the a-pawn. Still, after a reasonable plan of action like 26… f5 27.Ra4 Re7, black should be holding his own.

Instead of playing on the kingside, however, I decided to play on the queenside and tie white’s pieces down there. That really didn’t work out. The game went 26… Rb8 27.Kc2 a5 28.Ra4 Rcb5 29.Rb1

Dragun 2

Ok, what next…? White will simply play c4 and start pushing black back. The best course of action here is still to play 29… f5! 30.c4 Rc5 31.Kc3 Kf7 32.Rba1 Ra8, after which black is a bit worse but things aren’t too bad. Instead, I went for more counterplay with 29… Rd5, which just turned out not to work. After 30.c4 Rd4 31.Kc3, I realized that 31… Rbd8 32.Rxa5 Rd3+ 33.Kb4 isn’t so great for me, though after 33… f6! black can still put up resistance. I backed down with 31… Ra8, being under the illusion that I could save the a-pawn. After 32.b3 Re4 33.Ra1 Re2 34.Rxa5, I found myself totally busted. I threw in the towel a few moves later.

Did I just ruin my GM norm? I should’ve crawled into a hole after this, but I didn’t. I just… brushed this off.

Now for some drama. My performance dropped below 2600, but the norm wasn’t gone. I needed to beat a 2400+ opponent to get my performance back above 2600. But would I even get to play such an opponent? I honestly didn’t really care. Yes, I did want to get such a pairing, but I wasn’t stressed out. Here are a few reasons…

  • My rating is still far away from reaching 2500; if I were 2490, it would have been a totally different story…
  • There’s no doubt in my mind that on the way to 2500, I’ll have my fair share of norm chances if I keep playing like this
  • GM won’t be the end of my chess journey—it’s just a step on the way up, though a huge step at that.

As the wallcharts for round 8 were incomplete, it was impossible to figure out whom I would play; the best I could do was to make educated guesses with the incomplete results at hand. And as it is often the case with the last round, the pairings didn’t come up until after the round was scheduled to start. I did the only reasonable thing—I prepared for a couple players whose ratings would give me a shot for the norm. And in the end… I got lucky with the pairings: I was white against IM Kacper Drozdowski (2490 FIDE, 2562 USCF). This was the third time when I was in a situation where I needed to win to get a GM Norm, the first time was at the Philly Open in April 2017 and the second time at the Washington International in August 2018. I had lost both times before, but this game was different…

Drozdowski 1

The opening had gone well for me, as I had established a solid bind in the center and was already going on the offensive with 14.g4. Black is already in a tight spot here. Attempts to lash out with 14… g5 won’t work: white has 15.Qh3 or even 15.Nd5! (with the idea 15… exd5 16.Nf5 with a strong attack). 14… h6 15.h4 g5 is even riskier. Black’s best chance may be to sit tight with 14… g6, but that doesn’t look pleasant…

Instead, my opponent played 14… Rfd8?, which looks logical but won’t be stopping white’s attack. After 15.g5 Ne8 16.Qg3, the game became one-way traffic: I played Rac1, reinforcing my position, played f5, got my knights on f5 and d5 after he recaptured exf5, won a pawn, etc. I won.

I honestly wasn’t so nervous this game and only started getting a bit shaky when I knew I was winning and choosing between plenty of very good options. Fortunately, this only meant that I double checked my calculations more often than usual and played very accurately.

And that’s how I became a really happy camper. Yes, the third time was the charm with getting a GM Norm—and what was officially the highest FIDE performance in my life (2610)! My fall-winter slump was finally over.

Image result for minions yay gif

Let’s go!!!

As for the shirt, it’s a keeper!

Third Time’s the Charm: The Middlegame

After the first 3 rounds of the Southwest Class, I had 3/3, which included wins against 2 GMs. This was obviously a perfect start. However, in the next three rounds, aka the “middlegame” of the tournament, I got hit with some serious turbulence.

In round 4, I got black against GM Razvan Preotu (2522 FIDE, 2590 USCF). I lost a pawn out of a bad opening, but fortunately I had serious compensation. I scraped my way back to equality, and then…

Preotu 1

This knight endgame is slightly more pleasant for white, but black shouldn’t be in any trouble here, right…? I decided to activate my knight here with 45… Nf7, though 45… a5, preventing white from playing b4, may have been easier. After 46.b4, I made an inexplicable decision: I played 46… Kb6? allowing white to play 47.b5. I instead should’ve played 46… a6, and black really has no problems after that one. For some reason, however, I thought I should also be holding easily after my move.

Preotu 2

Now black is starting to get a bit cramped. Over the next few moves I continued to drift: 47… Kc7 48.Kc5 Ng5 49.Nf4 Ne4+ 50.Kb4 Ng5 51.Ka5 Kb7 52.g3 Kc7 53.Ka6 Kb8 54.Nd3

Preotu 3

Here I played 54… Ne4? which is officially a mistake. Instead, I had to play 54… Nf7, which ties the white knight down to the e5-pawn and takes the d6-square away from the white king if he plays like he did in the game. Black should be holding here, but after my move he’s in huge trouble. My opponent correctly played 55.g4! Ka8 56.b6 axb6 57.Kxb6 Kb8.

Preotu 4

After 58.Kc6 Kc8 59.Nf4, black is probably lost, since after 59… Ng5 60.Kd6, white is winning the e6-pawn. Instead, my opponent played it the other way around with 58.Nf4? Ng5 59.Kc6, walking into 59… Nf7! hitting the e5-pawn. If white retreats with 60.Nd3, black can simply play 60… Kc8 and white can’t get through. My opponent played 60.Nxe6 and offered a draw, which I accepted.

Phew! One of these days I’ll lose a game for playing like this, and it’ll serve me damn right.

3.5/4, 3 GMs and 3 foreigners down. Not bad!

In round 5, I got white against the top seed GM Jeffery Xiong (2666 FIDE, 2750 USCF). Rating-wise this game would be a tough order, but I had white and was in good form…

Xiong 1

So far, everything was all right. I didn’t get any advantage with white, but I wasn’t worse either. Still, there were a lot of pieces on the board, and anything could happen. After the best move 20.Be4!, improving my bishop and attacking the b7-pawn, the positon is around equal. I instead played 20.Bf5?! and missed the tricky move 20… Nh4

Xiong 2

I should’ve played 21.Bxb6, but I was worried about 21… Qxd1 22.Raxd1 Nxf3+ 23.gxf3 fxe6.

Xiong 3

This isn’t so pleasant for white, but in reality, the position isn’t far from equal after 24.Kh2 or 24.Re4 Nxh3+ 25.Kg2 Ng5 26.Rb4. Look what I did instead:

21.Nxh4? Bxd4 22.Qg4? Nd3 23.Red1 Bxf2+ 24.Kh2

Xiong 4

I admittedly missed the very strong 24… Rc4!, but even after a move like 24… Ne5 black is much better. After 25.Qf3 Qxh4 26.Bxe6 fxe6 27.Qxd3 Qf4, I went down pretty quickly.

Okay, what was that…? It was my first setback, and considering that I still had a 2600+ performance, I managed to brush this game off pretty easily.

In round 6, I got black against IM Omer Reshef (2491 FIDE, 2566 USCF). I didn’t hold back at all in this wildly complicated game. My silicon friend points that white did have a big advantage at a few moments, but to my human eyes, the position was just unclear.

Reshsef 1

Material is technically equal here, but white has a central pawn mass versus black’s b-pawns. There’s quite a commotion in the center of the board. And this was after the position calmed down a bit! I honestly wasn’t sure which result I was playing for, but I knew I had to act fast. I played the logical 23… Re6 but missed some details after 24.Qf4, after which I came to the conclusion that I was in trouble. I actually had 24… b5!, a move which I don’t think I even considered, at my disposal. Black has threats including Nxc3 Bxc3 Bd6 and Bb2. White is actually the one who has to play for equality with 25.Qd4! Bb2 26.Qxd8 Rxd4 27.Bd4.

Instead of that, I made a serious mistake with 24… Nxc3. The game continued 25.Bxc3 Bd6 (25… Rxe2 26.Qg4 g6 27.Rd7 Qc8 28.Qd4 is also unpleasant for black) 26.Qd4 Qg5

Reshef 2

27.Rd7! attacking the bishop would’ve given me a run for my money, since I simply won’t have time to take the e2-pawn. My opponent played 27.Rxb7 instead, which is strong but not best. After 27… Bc5 28.Qf4 Qxf4 29.gxf4 Rxe2 30.d4 Bd6, the dust settled.

Reshef 3

White is going to be a pawn up once he collects the b3-pawn, and he’ll have a dangerous passed d-pawn. This is far from easy for black, but it could’ve been worse. Though my play wasn’t the best, after defending for 55 moves, I managed to make a draw.

1/3 in this phase of the tournament wasn’t ideal, but it was a decent result given my opposition. I still had a 2600+ performance, was gaining plenty of rating, and was having a good time. Now all that was needed for a GM Norm was to maintain a 2600+ performance. By my estimates, I would need to score 2/3 or 2.5/3 in the last 3 rounds, which is obviously much easier said than done.

Next up, the endgame!

Third Time’s the Charm: The Opening

Last fall was far from great. My play was off, and my rating progress looked like a mudslide. The U16 Olympiad in Turkey was a breath of fresh air for everything… except my chess. With my schedule getting even more hectic than usual, the winter wasn’t looking very good for chess, but failure is not an option.

Over Presidents’ Day weekend, I flew to the Southwest Class Championships in Dallas, TX, to play a strong 9 round tournament, have a good time, and hopefully do well or at least not badly. OK, OK, let’s not pretend that getting a GM Norm wasn’t at the end of that list. And a great tournament it was!

There was just so much content in the tournament that I’m splitting this article into three parts: the opening, the middlegame, and the endgame. Yes, even tournaments have an opening, a middle, and an end.

As in a real game, a good opening is a good sign. The “opening” of a tournament, aka the first three rounds, is where many norm chances—and tournaments—go downhill. A bad start could easily mean spending most of the tournament playing lower-rated opponents which can tank your average beyond repair. Even if you score well and get back up to the top boards or even win the tournament, you may come nowhere close to a norm. Been there, done that!

In round 1, I was black against Austen Green (2080 FIDE, 2278 USCF), a local TX player. Out of the opening, we reached this position:

Green 1

This position looks fairly normal where white has tried to open the center with c4. With my last move 13… Bd7-e8, I wanted to bring my bishop into the game by moving it to h5 or g6. White has to decide what to do here.

A natural move like 14.Rfe1 will be met with 14… Bh5, leaving white in an awkward situation with his knight on f3. 14.Nf4 can be met with 14… g5!? 15.Ne2 (15.cxd5 Qxf4 16.dxc6 Bxc6 should be slightly better for black) 15… dxc4! (15… Bh5 16.c5 is OK for white) 16.Bxc4 Bh5, where white is in an unpleasant position. White’s best option is most likely to play directly with 14.c5! Qd7 15.b4!, intending to play b5 next. Black can naturally strike with 15… e5, but after 16.dxe5 fxe5 17.b5 e4 18.bxc6 Nxc6 19.Ned4, white is actually doing all right, though his position does look somewhat suspect.

Instead, my opponent played 14.Qf4?! offering a queen trade. 14… Qxf4 15.Nxf4 didn’t appeal to me, though black actually can stay afloat with some …Nb4 tricks. 14… Rd8!? was a strong possibility, but I chose 14… Qd7 instead, because I felt that the white queen was misplaced on f4. It blocks the e2-knight from reaching its natural destination and could get attacked in lines after …Bg6 or even …e5. White should play 15.c5 or 15.b3, though black is already somewhat better after 15… Bg6. However, my opponent played 15.Qh4?, which played right into my hands. I replied with 15… Bg6

Green 2

White is in a tough spot here. If 16.Bxg6 Nxg6 17.Qg4, black wins a clean pawn with 17… dxc4. 16.cxd5 Bxd3 17.dxc6 Nxc6 doesn’t look good for white, but it was probably the lesser evil. My opponent played 16.Nf4, and after 16… Bxd3 17.Nxd3 Nf5 18.Qg4 dxc4 19.Nc5 Qd5, I found myself a pawn up, which I went on to cash in in a rook endgame.

All in all, it was a fairly smooth game, especially for a first round…

In round 2, I was white against GM Bartek Macieja (2527 FIDE, 2615 USCF). Already playing up in round 2 was a treat! I was by no means going all-in this game, but it got flashy pretty quickly…

Macieja 1

So far, so good. White has a nice positional grip on the position, and I also had a serious lead on the clock. Black’s last move 21… g5 was a practical necessity to prevent me from simply playing Bxf6, occupying the d5-square, and positionally squeezing black. Now I had to decide what to do.

22.Bg3 is a perfectly reasonable move. I wasn’t quite sure what white’s plan would be after it, but it probably involves some kind of play on the kingside with h4 while maintaining a bind in the center. However, I couldn’t resist the temptation of playing 22.Bxg5!?. The game continued 22… hxg5 23.Qxg5+ Kh8 24.Re3 Nh7 25.Qh5

Macieja 2

White is obviously planning to play Rh3 next threatening mate, and stopping that threat is easier said than done. 25… Rg8 is possible, but after 26.Rf3!?, with the idea of capturing on f7 with the rook, black’s position looks very dangerous. Black’s other option is to bring the c7-rook into the game, and this is actually best accomplished by 25… Bd8! 26.Rh3 f5! 27.exf5 Rg7, where black is getting some counterplay of his own. My opponent chose the reasonable looking 25… Bg5 26.Rh3 f6, which I had somehow missed when I played 22.Bxg5, and now I had to decide how to proceed.

Macieja 3

This is the point where I messed up. Black is still pretty tangled up here. 27.Be6!, with the idea of bringing the bishop into the attack with Bf5, is the strongest move here. Black can double rooks on the 7th rank to protect the h7-knight, but it will take him forever to untangle after that. Meanwhile, white might even bring his a4-rook into play with Rc4, causing even more headaches from black, who could easily make a fatal mistake.

My move 27.Nd5 was fine, but after 27… Bxd5 I recaptured the wrong way with 28.exd5?. My goal was to win the a6-pawn, but this move allows black to get some much-needed privacy and counterplay of his own with …f5. 28.Bxd5, with the idea of installing a positional bind, was much stronger. The game continued 28… Rg7 29.Rxa6 Qd8 30.Rg3 f5 31.h4! Bxh4 32.Rxg7 Kxg7 33.Ra7+ Be7 34.Bb5

Macieja 4

Over the past few moves, I sacrificed a pawn with 31.h4! to renew play against the black king. In this position, black is pinned on the 7th rank, and I’m threatening the powerful Rd7. 31… Rf7! was the best way to stop this, and white has nothing better than a draw, for instance after 32.Bd7 Ng5 33.Be6 Nxe6 34.dxe6 Rf6 35.Qg5+ Kh8 36.Qh4+ Kg8 37.Qg5+ Kh8. My opponent played 34… Rf6? instead, and this turns out to be a serious mistake. 35.Rd7 doesn’t work on account of 35… Qf8, but I found another idea: 35.Be8!, threatening Rd7 again. 35… Nf8 is more or less the only move for black, but he’s really tied up now. This was as good a moment as any to get my queenside pawns marching with 36.b4!. The game continued 36… f4 37.b5 Ng6?! (37… e4 would have provided better resistance, though white is probably still winning there). 38.Bxg6 Nxg6 39.b6

Macieja 5

Black has almost succeeded in untangling, but he’s too late. The b-pawn is out of black’s control. Combining the b-pawn and threats against the black king, I won in a few moves.

Not bad! This was far from a clean game, but chess is still… a game!

In round 3, I got a double white against GM Angel Arribas (2454 FIDE, 2518 USCF). Again, I had no intentions of going all-in; I just wanted to play and see how it’d go…

Arribas 1

This appears to be a fairly normal Sicilian position, except that a) black has a pawn on h5 and b) black hasn’t castled yet. However, he’s gotten the bishop pair, and my pieces aren’t that impressive. Still, if black simply plays 17… 0-0 here, he’ll run into 18.Rd4! Qc6 19.Nd2!, after which his queen is simply getting harassed. He has to play the rather ugly move 19… Be8, and after 20.Rd1 white is much better. Therefore, to prevent Rd4, my opponent played 17… e5, which I decided to directly counter with 18.f4.

18… 0-0 is a bad idea on account of 19.fxe5. 19… dxe5 loses to 20.Rxf6!, and 19… Nxg4 20.Qd2 Nxe5 21.Nd5 is very bad for black. Instead, black should play 18… Be6 or 18… Bc6, and my opponent chose the latter. After 19.fxe5, however, the game took an unexpected turn.

Arribas 2

Black should play 19… dxe5. White does win a pawn after 20.Qg3 0-0 21.Qxe5, but after 21… Rfe8 black has reasonable compensation. Instead, my opponent played 19… Ng4? which more or less loses (!) after 20.Qg3!. After 20… dxe5 21.h3 Nf6 22.Qxe5 (22.Rf5! is even stronger), black is just lost. After 20… Nxe5, as was played in the game, I played 21.Qxg7 Ng6 22.Nd5 (22.Nd4 may have been even stronger) 22… Bxd5 23.Rxd5

Arribas 3

Black’s position is truly busted, and my opponent resigned.

3/3. My performance was well over 2600, and I had already played 2 GMs and 2 foreigners. Oh boy. This was looking good. Onward!

Positional Exchange Sacrifices

Upon hearing the word sacrifice, most of us think about brutal sacrificial mating attacks, but that’s not always the case. Exchange sacrifices can be based purely on positional reasons in endgames. They can be a way out of a bad position, or they may be the best way to get winning chances.

How do you know that your exchange sacrifice is worth it? Obviously it all depends on your position, but I’d say that the following factors would be good indicators

  • Afterwards your pieces will be well-placed (that’s obviously a plus!)
  • Your opponent’s rook(s) is/are not active or will not have an easy time infiltrating
  • Your opponent’s pieces aren’t so active in general
  • Your position is relatively secure
  • You have passed pawn(s) that your opponent needs to worry about
  • Last but not least, the more pawns you have in return for the exchange, the better

Now I’m obviously not claiming that ordinary exchange-down positions are ok for you. No, no, no! A rook is better than a minor piece in most circumstances. And of course you have to calculate accurately, since you really can’t justify losing by force. Let’s look at a few examples.

At the recent U16 Olympiad, I did make a (good) exchange sacrifice.

isik

I was black in this strange position. White has passed pawns on g5 and f6 that are blockaded but do tie up black’s pieces. Meanwhile, white’s bishop does a much better job blockading the g5-pawn. Black doesn’t have a clear plan of action here, and if white can get a rook to the h-file (say after Kf2 and Rh1), black’s position won’t be pleasant.

Taking everything into account, I decided to play 24… Re5! here, with the idea of sacrificing an exchange on g5. This is a good idea from a practical perspective. White’s f6-pawn will still be a thorn, but I’ll be able to take care of it by playing Kd7-e6. Besides that, the only realistic problem with black’s position is that white invades with a rook and takes my queenside pawns, which will become an issue but doesn’t seem to be too concerning.

While white is still the one pressing, this is a better scenario for black than if he waited around and let white proceed with his plans. In the game, I was able to successfully hold a draw after some adventures.

While the idea of sacrificing an exchange came to me naturally there, I’ve had some mishaps in the past. Take this example from 2014, when I was ~2250 USCF:

davtyan1

Again, I was black in another fairly strange position. Black has a pawn lodged on d3, but white has his a knight lodged on d6 in return. How to handle this? Black is in check and obviously has limited options. A tempting possibility here is to remove the knight from d6 by playing 19… Rxd6! 20.exd6 Nf5.

davtyan2

White’s life is far from easy here. If black could simply play Nxd6-e4, he’d be dominating. Since 20.Rc1 attacking the c5-pawn is simply met with 20… Kb6, white will probably play 20.e4 Nxd6 21.e5, where black has lots of compensation after 21… Ne4, Re8, Nf5, etc. This is because black has a solid blockade on the light squares, his d3-pawn is strong, and white’s rooks simply don’t have open files to exploit. While it’s not so bad for white, black is for choice.

Instead of that, I played 19… Kb6, which isn’t a bad move. It’s after 20.a5+ Kc6 21.Rc1 that I made my howler.

davtyan3

Here I should have also gone 21… Rxd6!. After 22.exd6 Kxd6, black will have a lot of compensation for similar reasons like above. Instead of that, I played 21… Rb8??, preventing Bxb4, but after 22.e4 I found myself in a lost position, since I won’t be able to save my c5-pawn. I went on to lose.

Looking at this game now, I’m totally shocked I didn’t sacrifice the exchange. The first time is ok, but the second time!?

Those two games had some similarities. Both were fairly strange positions full of imbalances, though they were also fairly closed positions in which rooks weren’t that powerful. It was also easier for me to establish a blockade after sacrificing the exchange than to play the positon “normally” in both situations.

Scanning through my games, I’m surprised how rare these positional exchange sacrifices are in my practice. This goes to show that yes, being an exchange up is usually a good thing, but there are situations in which a minor piece and a pawn are more useful than a rook.

This painful experience of mine from a couple years ago shows that a rook can truly be dominated.

exchsac1

I thought I was doing all right in this position, but not after I got hit with the strong sacrifice 33.Rxc8!. After 33… Rxc8 34.Nxf5+ Kf7 35.Ne4, black is in big trouble.

exchsac2

White’s knights are quite well-placed and powerful, especially compared with their black counterparts. Moreover, they’re attacking the d6-pawn. If it falls, black’s position will be in ruins. I therefore played 35… Rd8, but after 36.Ba5 Rd7, my rook is literally stuck. To be more precise, it’s totally dominated. I went on to lose this position.

Long story short: positional exchange sacrifices do exist and can be quite good in various situations. If it looks like you have a lot of purely positional compensation after an exchange sacrifice, it’s worth a shot!

Revisiting the Past

It’s always interesting to go back in my personal games database and look at some old games of mine. It brings back good and bad memories and highlights how much things have changed. The other day I revisited a game from 2014, when I was rated around 2150. At the time, I thought it had been a very nice game, except for a small theoretical slip-up. Upon taking a closer look, I found that that wasn’t exactly the case…

pascetta 1

In this French, I had sacrificed an exchange on f3 as black to reach this position. Back in those days, I was a bit of a chicken, and sacrificing an exchange was about as “wild” as I’d generally go. In this position, black has two pawns for the exchange in the form of a central pawn mass. Black’s rook on c8 is nicely placed, his knight on g6 is temporarily guarding the h7-pawn, and his queen can easily hop into the action. Meanwhile, white’s queen is fairly active, white’s h3-rook looks strange but it’s useful, and his a1-rook will join the game pretty quickly. Black’s plan is to play …e5, with ideas of e4, Nf4, piling up on the c2-pawn, etc. If white sits back and does nothing, things could turn sour for him very quickly.

Looking at this position today, it seems pretty natural that white should generate counterplay. He doesn’t have to perform an all-out attack on the king; he can just keep an eye on it. This could be accomplished with a move like 18.Rf1. 18… e5 could quickly turn into a disaster after 19.Qf5!. Black will probably play 18… Qb6 instead, but after a move like Rf2, Kh1, Rg3, etc. white is doing all right. Black will most likely not get away with 19… Qxb2, and it really isn’t clear what he will do next.

My opponent played 18.Re1, which is a pretty natural move, though it isn’t best. I replied with 18… Qb6. Then came 19.Rb1?.

pascetta 2

This is a move that sets alarm bells off in my IM brain. What really surprises me is that I didn’t make any comment about this move in my notes. 19.Rb1 defends the b2-pawn, but is that a serious issue in the first place? if white say plays 19.Kh1? If 19… Qxb2, white has 20.Qe2 Nf8 21.Rf3, generating counterplay against the black king. One critical resource to spot here is that 21… Qxc2?? loses to 22.Rxf8+! Kxf8 23.Qxe6 Qc4 24.Kg1. White doesn’t even have to play like this. Alternatives include 19.Rf3 and 19.Rf1 (yes, this does waste a tempo compared to 18.Rf1, but it’s still fine for white), after both of which 19… Qxb2 is actually bad for black. Black could play 19… e5 instead, but after 20.Qf5!, white’s counterplay is coming just in time. Another thing to point out is that playing 18.Re1 on the previous move, white is moving his rook right back to the awe-inspiring square of b1 in response to a reasonable move from black (18… Qb6). Wasting a tempo can’t be good, and the white rook will be doomed to babysitting the pawn.

To summarize 19.Rb1: no, no, no, and NO!!!

I naturally wanted to play 19… e5, but I was worried about 20.Rxh7 Kxh7 21.Qh3+, hitting the rook on c8. What I missed was that I actually don’t have to recapture on h7; I can go, for instance, 20… Nf4 21.Qf5 d3+ 22.Kh1 Rxc2, with a totally winning position. My d-pawn is close to queening, white’s king is suddenly shaky, and white’s attack is nonexistent.

Instead, I played 19… Qc5?. Now I’m protecting the c8-rook in those variations and am “attacking” the c2-pawn. However, this is a mistake that gives white a second chance to activate his rook. If white simply moves the rook away with 20.Rf1!, 20… Qxc2 is no longer a threat on account of 21.Qxd4. And if black plays 20… e5 instead, he’ll be met with 21.Qf5! where white is clearly getting serious counterplay. My opponent instead played 20.Rc1?, another bad move. I replied with 20… e5 and got my pawn mass rolling. After 21.Kh1 e4 22.Qd2 d3 black is already winning. My opponent tried to generate counterplay with 23.Qg5, but I played 23… d4 24.Qg4 d2 25.Qe6+ Kh8 26.Rf1 Qc4 27.Qf5

pascetta 3

Wow, those pawns really are rolling. Comparing this to the starting position, it’s fairly clear that white has lost a gigantic amount of ground without getting anything real in return. I finished the game off with an elegant trick: 27… Qxf1+! 28.Qxf1 Rxc2. White has a queen for a knight and three pawns, but he is helpless in preventing 29… Rc1. Not bad! My opponent resigned.

After the game, I really liked my play. True, the ending where my pawns were rolling down the board was picturesque, but it shouldn’t have gotten to that point. One thing which strikes me now is how wrong my 2014 notes to the game were. Yes, it was nice to live in the bubble that this game had been a masterpiece and that my position had been good all along. I think I didn’t understand that the position is, in reality, around equal. I’ve had a few other such “masterpieces,” where my play was far from brilliant and where my opponents greatly helped my cause.

Takeaways:

  • Don’t judge a position by its cover. Yes, that position was easier to play for black, but white wasn’t helpless against black’s great plans.
  • Don’t just sit there and wait for your opponent to execute his plan. Try to mix things up. If it looks like you will get steamrolled if you do nothing, you should try to generate some counterplay ASAP.
  • Try not to be passive. In this case, white should have tried to keep his rook active instead of dooming it to eternal babysitting with Rb1 and Rc1.
  • Don’t automatically recapture pieces. We all do it, but once you figure out that things aren’t so good after recapturing, look for alternatives. When I was looking at lines with Rxh7, I was always recapturing Kxh7 and didn’t realize that I’m winning after Nf4.