Chess.com 2024, 3rd round, E05 - Draw your weapons

13^th of June

Andyh timed out in a few more games:

In the white Draqlek game, I might have finally caught a small break:

I was much more concerned about Nxg2 than Nxd3. The lines still look very similar to how I left them yesterday.

This was no break – Nxd3 is the top line for black. Nxg2 goes into this completely bizarre line which according to the comp is +0.7 for white despite having two minor pieces for a queen:

I obviously would have never played that. At this point, I can reveal that my opponent was later caught for cheating, and I wonder if he played Nxd3 because he didn’t see Nxg2, or because the above engine line was known to him.

I get another timeout late morning, from Dumblecore. Exactly what I hoped for; the load is now reduced to 15 games.

The game never left the Bogo-Indian, so nothing to comment there.

Also, there is the first result between top players, where VFK played at a very high level. I had a peek at that game, and it’s difficult to judge his strength based on that game alone. I’ll play him as slow as possible to get more insight into what we’re dealing with.

As I have only two hours left on the game below, time to make a move:

The natural move is something like Rfd1, but I don’t like the pin after Bg4, and don’t want to waste time on h3. However, on reflection, h3 has also the upside of limiting Bc8 and potentially preparing g4. Black will likely go Rfd8, but if he tries Bf5 or Na5, I might have some creative options.

h3 is a decent move, but I again put too much value on black’s Bg4 pinning and the resulting broken pawn structure after Bxf3. The comp points out that the bishop pair advantage compensates all that nicely.

In the Artem game, he is still pressing for a win:

h5 is a good move, gaining space, and limiting my future options on the kingside. I was planning Qc7 to secure the weaknesses on the queenside, and then seek counterplay. On that front, my main hope is pinned to moves like b5, and then either going after the a4 pawn, or generating an a-passer.

After a closer analysis, I see many annoying options for white, so it’s definitely not over. h6 is a strong resource, and it appears in many lines:

Both Qh8 and Qf5 looks strong after black’s Qc7, but I don’t see other good options.

I missed a trick here. Black could play 32…Ne7 and then h6, which would have created a fortress on the kingside and allowed the black queen to go on a hunt against white’s worse queenside pawn structure. The comp shows -0.6 down that line, and 0.00 on all the other sane options – including the one I’ve chosen. This was my only chance to play for a win in this entire game. I was likely so focused on getting a draw that such ambitions escaped my attention.

In the Eyeblinks white game, things are getting very interesting:

Tactics are in the air, with forks on d6, skewers on the c-file all coming to mind. After another set of calculations, I start seeing a very narrow path to victory, starting with Bxc5:

Black had two good moves here: 22…a6 and Qd7. Both led to a strong black advantage. Bc5, however, was a mistake which gave me the game. That’s how it often goes in those tactically charged positions. The calculations are not too difficult, but still happy I’ve got them right despite being unwell that day.

White’s last move is a bit weird, though I guess the idea is to force opening of the c-file by making me play Nxd3, at which point Rc1 will start firing from its hiding place. I’m thinking Qb6 could be what the doctor ordered, since b2 is no longer indirectly protected.

This shows that I wasn’t in the best of conditions, since the pawn on b2 can protect itself by going forward to b4. Rc8 or b5 were more natural non-committal moves. The comp also suggests f6, but I wouldn’t have played that without a brain firing on all cylinders.

Making a couple of moves against Artem and Eyeblinks, as they are on pre-calculated lines, skipping the analysis. It’s clear that drawing is the pinnacle of my hopes and dreams in the Artem game, which is frustrating to say the least.

14^th of June

There’ve been more timeouts since last night:

Artem started repeating moves against me; let’s see if I get a draw there later today.

The next game is the most difficult one so far:

I need to decide whether to take on e3 or d3. Taking on e3 (as I originally planned) is problematic due to Nxb2, which leads to a swift loss after the queen exchange. From further away I assumed queen exchange is beneficial, but that was a coarse and incorrect assumption.

I had an inaccuracy in calculations. Qxe3 is a slightly better option – since Nxb2 does not lead exactly to a loss. The position looks horrible, but it is not as bad as it looks, as the engine evaluates it at -1. On the other hand, black had the Bxg5 option after Qxd3, which I simply didn’t see. If not for that, my choice would have been absolutely correct. Fortunately, black didn’t see it either.

I do get the threefold repetition against Artem, which is a bittersweet moment. On one hand, I’m very happy about how I conducted the defense. On another, this is a player I should win against.

The analysis shows a high level of play on both sides – I made just one mistake with Qg4:

It’s a tough one to swallow – will see how the same player fares in his other games. I’m very temporarily in first place, with 14 games left.

To put it simply, I just didn’t set white with enough challenges and opportunities to go astray. The French Exchange is not noted for its counterattack chances, but I should look for more double-edged lines against lower rated opposition even there. Easier said than done, but the main lesson is being more deliberate with choosing the right opening for the right person.

The next game for this morning’s session is:

I feel like e5 is mandatory here. Otherwise, white will go e5, exchange my bishop on b4 and grab the outpost on d6. His other plan is to go d4-d5 and generate a d-passer. This is what caused some strong players to lose for black here, so I’ll need to find an antidote.

Back to the symmetric game:

I still don’t have a good plan here to create imbalances. Rfd1 is the natural move, but it might be too natural. If black goes Bf5, should I just charge with g2-g4-g5? On a closer look, g4-g5 seems like a step too far, but g4 is in the right direction.

My fortunes have changed when it comes to raffles. Got “Best Dad Ever” present at work’s picnic today, which I was sure would go to some childless student or one of the women in the office. Sometimes there’s some minor poetic justice to be found in this world. 

The drive to Yosemite was alright (bar long traffic jams in Los Banos), up until we got to the unpaved road to the farm. The instructions in Airbnb were in direct contradiction to the messages sent by the host, and I unfortunately followed the former. We got there eventually, and managed even to sneak to Mariposa, where the only place still open was a pizza parlor with long benches and pictures of local coaches on the walls. Wheel of Fortune was playing on the TVs to complete the country vibe.

Oh, and Mavs finally won – I didn’t see the game, as it was during the drive, but looks like I’ll get to watch one last game on Monday. Most likely it will be the last one for the season.

Back at the farm, I have time to look at a couple of games. The most interesting one is the Eyeblinks white, where I’m seeking that elusive win:

Taking the rook may take me to a win, but it won’t be easy. This is the line I calculated a few days ago with a similar conclusion.

Found a few improvements, but it still does not seem decisive to me. I have an extra exchange, but it’s not a slam dunk from here.

Qxa7 is a creative alternative, but black can hunker down with Kg8 and Qf8, stopping all creativity:

Looks like I’ll need to go after the exchange, with a long uncertain conversion ahead.

The comp has no illusions and rates the Qxh8 line with +3.5. The key point is that I made the right choice by a method of exclusion, which is something I should continue doing – it tends to work even if I miss calculation subtleties.

15^th of June

The first game is in the Semi-Slav, which I never played before:

I suspect he’s using a top level predecessor game here. In that game, Erigasi played Re8, perhaps to make exd4 a more viable option and force white to play d5. I don’t like that move as much, and more leaning toward something like Rd8, however that loses the pawn on e5. Next option is just exchange on d4 and hang on to the pawn on c4 with Nb6. Then I see white’s potential plan after h3 – continue with g3-g4, perhaps Nh4 and f2-f4. So, the conclusion is to play h7-h6 to counteract that plan and let him clarify his intentions more.

The engine backs exd4 up, since it does not believe in mirages and believes black’s extra pawn and a knight on d3 are compensating whatever white’s cooking up on the kingside. However, on the positive side, it thinks my choice is still slightly better than the GM’s, so I’ll take a small retroactive victory lap.

Next game is with Draqlek:

Here, things are much simpler. cxb4 is forced, and I see a draw-ish ending with equal material and B vs. N in my future.

While not a bad move, cxb4 is not forced. Black also had the option of Bd7 and Rc8 to recapture the c5 pawn. This would have left a tiny advantage due to the dynamic pawn duo in the centre.

Staying with the same player:

Here, things are also both simple and complicated. Kxe3 is forced and is also what I calculated a couple of days back when I saw this entire line. That’s the simple part. The complicated part is that I’m likely to lose this game. While I avoided immediate destruction, white’s position is in shambles, with unprotected king, poor pawn structure, and a worse minor piece. This one will be very tough to defend.

My eval was extremely pessimistic, as the engine reports +0.12, wondering what I’m moaning about. Perhaps I was biased because this is just not the type of position I enjoy. That being said, I got a lucky break due to black playing 16…Qxb2+ rather than Bxg5, which would have left him a pawn ahead.

Enough for this morning, time to enjoy nature.

Yosemite is one of the most fascinating places I’ve been to (and I’ve been around the world). Of course, having another 6500.5 people sharing my opinion by populating the same trails does not help, but that’s the tax we all pay for nature’s wonders. On the way, the local souvenir store had the best horror story preamble I’ve ever encountered. A creepy music box playing the theme to Pink’s Panther, talking fortune teller machine, rotating large deer head commanded by Alexa – you name it.

Time to make a couple of chess moves. In the tournament, andyh lost another point – this time an honest loss, rather than a timeout.

I was expecting Bf5, and my plan was simple – hit him with g4. Without queens on the board, I don’t see a large downside to opening the kingside up, while I see benefits with getting some space, and potentially getting g4-g5 on the agenda. Other ideas are going for f2-f4, or exchanging a pair of knights with Nd4 or Nd5:

g4 was the right move against the wrong player. If I knew that I’m playing a 2400 IM, I would have played as conservatively as possible. At that time, I only knew that this chap has lost a game and is likely to lose more (spoiler alert – he didn’t). Both Nd4 and Nb5 were more conservative and appropriate options.

To the black game:

We are out of the opening book, and since it’s an unfamiliar opening, I have to engage the brain cells, as much as I’d like not to.

The d1 rook is clearly eyeing the black queen with a malicious gaze. I can do the same for him with Rc8. Qb6 is also a tempting move. The third option is cxd4.

I prefer closed positions, but here it’s not going to happen. Since I’m running out of time, I’ll go with cxd4, as that leaves white fewer options.

This was a rare mistake from white, with the eval going all the way down to -1.5. I stumbled upon the top computer suggestion. It’s hard for me to see why black has such a generous assessment; perhaps the much better activity of the pieces gets converted into a won pawn down some complex line.

Next game is with the cautious person:

As long as white doesn’t castle, I shouldn’t castle myself, what with the rook sitting there in ambush on h1. There’s also no sense in playing h6 until white castles or plays h5. The tempting approach here is to play Na5 + Rc8, operating on the queenside. Another idea is Na7, to exchange the poor white-squared bishop, but that seems too slow – even in such a closed position. Na5 it is.

Correct on all points. Again, a top computer line. I was doing well that day!

Now getting into the difficult part:

I earmarked Rhb1 here, to start creating a semblance of counterplay. I definitely want to get black’s queen out of a3. Black’s ideas are:

• h7-h6, to open the h-file and go after my kingside.
• Pile up the pawn on c3 by doubling the rooks on the c-file.
• Use the compromised king’s position with moves like Qc5+, a6, Bb5.

Queens' exchange is very much in my favour considering how bad my king’s position is, and the fact that the pawns stand on the right colour. Black’s bishop is constrained, and I should try keeping him like that, perhaps playing a2-a4. If black goes h6, I can try g5-g6, opening his king’s position up.

Last game for today is this endgame:

To put it simply, I don’t see alternatives to Rb8-b6 here. Will go with that, since the hour is getting late.

Continued here: https://www.chess.com/blog/RomanKleiner/chess-com-2024-3rd-round-e06-unfounded-optimism