All posts by admin

Drawmeister vacation

I don’t know what it is, maybe because it is summertime, but there is not much going on at the moment, at least in terms of forced draws.

If you want to watch exciting chess with decisive results, then check out the matches MC vs. Grischuk and Naka vs. MVL on chess.com. Just kidding! Nobody wants to watch exciting chess with lots of decisive results of course. We all love to watch computer analysis ending in forced draws, don’t we? Well, the Olympiad is coming up, so things could become “interesting” again very soon.

Expectation

If you ever wondered what Elo-difference means in expectation, here are some key odds*:

50 points difference = 5:4 odds
100 points difference = 2:1 odds
200 points difference = 3:1 odds
300 points difference = 6:1 odds
400 points difference = 10:1 odds

When Carlsen plays Caruana, it would be like pocketpair vs. overcards in Poker. Caruana on the other hand would be a 2:1 favorite against Leko, which can be quite a burden. Maybe that is the reason why Peter doesn’t get invited to such tournaments anymore.

*These are not the exact numbers, but they are close.

Coincidence or method?

When we look back, Leko was a child prodigy, Bacrot was a child prodigy, Polgar was a child prodigy, Radjabov was a child prodigy, Ponomariov was a child prodigy, Giri was a child prodigy, So was a child prodigy and Wei Yi is actually still a child prodigy. They all have something in common and that is their high percentage of short draws without a fight. The only difference is how they do it. Some prefer repetitions while others play 15 moves of mainline theory and offer a draw. Only Giri is a bit different, because he plays long games, but the overall theme with him is zero risk tolerance too.

Maybe this whole phenomenon has something to do with how child prodigies are brought up these days. They are the sensational kids who are playing in a field of strong grandmasters where a draw is sufficient to boost their rating. Their occasional wins usually come from situations when the strong GMs are pushing too hard in order to save their face. Kasparov-Radjabov, Linares 2003, would be such an example.

Giri’s career is especially interesting, because he only has a few wins against players from the top ten in classical chess, yet some people believe that he will be a challenger for the title in the not so distant future. Same goes for the chinese youngster Wei Yi. He played about 600 games to reach 2700 while mostly beating weaker opponents also. Neither Giri nor Wei Yi have won a top major tournament and yet they are some of the strongest players of the world by “definition”. This shows that you can reach the highest level by rather consistently beating 25xx-players and drawing everyone, rated 26xx and above. It is not surprising, because the result will be a 2700+ performance.

Believe it or not, but I have the feeling that the Elo-rating is the root of all evil. Being able to win Elo-points by making a draw kills the game, because there is too much incentive for not fighting it out. Chess is essentially a drawn game, so maybe the reward for reaching this result is too high in comparison for what you get for successfully disturbing the balance.

That’s one part of it, another one is the way too high k-factor for young players. Even with the k-factor 40 being capped to Elo 2300 for players under 18, it doesn’t prevent such “overrated” youngers losing points to other youngers who are above 2400 already.

In the end there is a reason why it takes  so many young players only a couple of hundred rated games to become world class after all, and it’s certainly not that their Stockfish or Komodo, whatever, finds better lines than the engines of all the other players. There is probably a lot of stat-padding going on at the tables and behind the curtain and their overall strategy is just perfectly adjusted to the rating-system.

So successful

The kid with an engine won the Sinquefield Cup 2016!

You don’t need chess books, you don’t need a coach, in fact you don’t need anything that costs money. All you need is an engine, calculation skill, good memory and a bit of luck to get your preparation in.

So stylish

Today Wesley just shut it down, but Fabi obviously didn’t mind either. With this draw So is well on his way to tournament victory. I don’t see MLV beating his Berlin in the last round and I can’t imagine Aronjan beating Topalov neither in the Ruy Lopez nor in the Italian. Maybe Svidler will blow up against Anand, which is unlikely given their history, but even in that case Wesley is a favorite to win it all.

The surprising thing is that he actually admitted killing any play in the position in the interview. Usually listening to his interviews feels like a total waste of time to me, just like watching his games, but he was honest and one has to give him credit for it. So far Wesley has shown the modern blueprint for winning tournaments. Pleasing the spectators is not part of it and nobody can blame him.

Tough luck

Svidler and Topalov both lost in the same round and I forgot to make predictions. Well, I guess I would have predicted all draws anyways.

Wesley is just amazing. His tournament strategy works out to perfection. He started with a well deserved surprise-win against Nakamura and since then he makes draws against equal or higher rated opponents and goes after the perceived targets in the field, Topalov and Svidler. Nevertheless I expect him to draw the rest. He doesn’t have to take any risks, because Anand will not be able to catch up.

Gimme dat Ding

Went 4/5 this time, predictions are getting sharper. Still come on, I should have been 5/5! What was Ding doing there? The whole idea with Rc8 and following it up with b4 and c5, basically spitting a pawn for nothing, was totally weird. Why not simply play c6 and treat it like a Breyer? I guess we all have a pretty good idea how Topalov must have felt about it.

 


Source: AustralianAds on Youtube

Playing the percentages

You can’t go much wrong by predicting draws. That’s how games between players of similar strength usually end. At least I got three predictions right this time. Nevertheless my score could have been better since MVL came close after Topalov fell into an opening trap, but unfortunately it turned out to be too difficult to convert. To be fair, I didn’t expect such a battle between Giri and Ding and I didn’t expect Caruana to allow Svidler to set up a position with a symmetrical pawn structure either. Come on, how can you play the Lasker against a guy who is struggling that much? Actually Fabi actually got lucky that Peter spoils all his good positions too.

Round 4 predictions:
Aronjan-Caruana 1/2
Nakamura-Topalov 1-0
Ding-Anand 1/2
So-Giri 1/2
MVL-Svidler 1-0

 

P.S.: Congratulations to Stockfish for finally beating Komodo in TCEC again.

So much about predictions…

How many games did I get right? It looks like one. Ding and So did indeed draw as expected. I shall not make any further predictions 😉

I am still surprised that Naka effortlessly beat the unbeatable Giri and that Anand played the Caro Kann and beat MVL. It’s actually not that surprising that Svidler lost again, because it seems that he overlooks tactics and mishandles techinical positions just as well. In his current form even the type of position doesn’t matter anymore.

One game was very interesting from a educational standpoint and that is Caruana-Topalov. In particular it was Fabi’s decision on move 19, because it is a nice example for the move-search-algorithm.

Let’s go through the process of finding candidate moves:

  1. Are both kings safe? Yes, there is no shade of an attack.
  2. Can we win material? Not by force. We can threaten to win the two bishops with h3, but it belongs into the next cathegory.
  3. Can we improve our pawn-structure? From left to right, we can play b3, d5, dxe5 and h3 of course. The first choice b3 gives up the bishop pair and white can’t even capture back in the end. Closing the center with d5 looks playable. Taking on e5 looks playable also. Finally there is the move h3, which is already on the list.*
  4. Can we improve our pieces? The queen is placed well on d1. The rooks are ok too. The bishop a2 looks weird and could be improved. There is also the option of taking on f6 and playing Nd5 similar to the outpost in the Sveshnikov.

I think this give us a realistic list of candidate moves to work with: h3, d5, dxe5, Bb1 and Bxf6. That’s actually a lot of playable moves for any middlegame position in chess and it shows why the Ruy Lopez is such a rich opening.

Let’s look at them one after the other:

  1. h3 threatens to win the two bishops or at least threatens to get rid of the pin. Furthermore it creates luft for the king and a retreat square for the knight. It looks like a decent move.
  2. Closing the center with d5 removes the tension from the center and in this position it is difficult to follow it up with the break b3. The other problem is that this move is very committal.
  3. Taking on e5 also removes the tension from the center and petty much the same considerations apply.
  4. Bb1 is a double purpose move: It uncovers the Ra1 and it overprotects e4 in order to follow up with h3, g4 and dxe5. It does allow tactics in the center though.
  5. Taking on f6 and playing Nd5 can be answered with Bd8 and it is not clear if the powerful knight is enough to compensate for the two bishops. This move is very committal also.

If Fabi doesn’t want to committ yet, he is down to two choices: h3 and Bb1. I think h3 would have been the natural move in this position and it is also the top candidate of the engines, but he decided on the rather artificial Bb1 with all the consequences. For me it was a strange decision, but who am I?

About the predictions, I lied! Here are they for round 3:
So-Aronjan 1/2
Anand-Nakamura 1/2
Topalov-MVL 0-1
Svidler-Caruana 0-1
Giri-Ding 1/2

 

* An argument can be made for not looking any further since in quiet positions improving the pawn structure is more important than improving the pieces.