Wednesday, March 27, 2013


"Bullet and Blitz have nothing to do with chess". That's a sentiment that's shared among many strong chess players, and I'm inclined to agree. Nevertheless, once in a while my bullet games produce a gem, typically with the - involuntary - help of my opponent. In this game, for example:

My opponent had just played Kg2-h3?? because I was threatening lethal discovered checks with my knight on e4. However the problem with the King on h3 is that I can now force it forward towards my pieces with a series of checks, eventually leading to this rather pretty finish:

Not a particularly high-level game, but despite the many mistakes and inaccuracies I do enjoy these kinds of games

Monday, March 25, 2013

Episode VI - How to Analyze a Position and Formulate a Plan

A critical skill for any chess player to have is the ability to accurately and systematically analyze positions. During a chess game it is very tempting to just look at "interesting" moves or randomly start calculating variations. This approach however, will almost always neglect the deeper positional characteristics of a position. And unless you're playing blitz, this will also make it hard to make good use of the time allotted to you.

Therefore, now that we have covered some of the basic tactical elements in chess, in our last lesson I taught Pyre a more systematic way to analyze a position, and form a plan based on the results of the analysis.

There are many good books about this topic. I think eventually I'll post a list of useful resources for beginners and novices. The approach I was taught many years ago by my first chess coach is based on Karpov and Mazukevich's recommendations in their book "Find the Right Plan". By the way I think this is a book that does not get the attention it deserves.

The basic idea is that in order to choose the right course of action, a chess player must first have a good understanding of the current position. This will not only make it easier to select a good plan, it will also help identifying candidate moves.

According to Karpov and Mazukevich, to get a good understanding of any given position, a player must look at each of the following 7 evaluation criteria:

1. Material Balance
2. Immediate Threats
3. King Safety
4. Open Files
5. Pawn Structure, Strong and Weak Squares
6. Center and Space
7. Development and Coordination among minor and major pieces

This analysis does not need to be performed after every move, but it has to be done regularly. For example, a player could form the habit of doing this after move 10, 20, 30, 40 etc. This analysis should also be done whenever the position on the board changes dramatically, for example after a player sacrifices material.

After discussing these principles in general terms, Pyre and I looked at an example to practice this approach. We analyzed the game between Boris Gelfand and Vassily Ivanchuk from the 2013 Canadidates Tournament in London, which had been played earlier the same day. The players reached the following position after Gelfand's 17th move Ng5:

Gelfand-Ivanchuk at the 2013 FIDE Candidates Tournament in London 2013; position after 17. Ng5
We tried to to come up with a good plan for Black. As the first step, we analyzed the position through the lens of the 7 evaluation criteria:

1. Material Balance: the material balance is even. The only difference is that White has the bishop pair and a knight while Black has two knights an a bishop.
2. Immediate Threats: neither side has any immediate tactical or positional threats.
3. King Safety: Both kings are fairly safe. Black and White have castled queenside which means that neither side can easily launch a pawn storm on the opponent. Due to the missing c-pawn, White's king is a little bit more exposed, but this will probably not have a significant impact on the game.
4. Open Files: There are no open files in this position. White has the half-open c- and g-files available to him and could potentially double up rooks on either of them. Black currently controls the half-open d-file.
5. Pawn Structure, Strong and Weak Squares: Despite the double pawn on the f-file, White has the better pawn structure. He doesn't have any potentially weak pawns. At the same time his pawns control many important central squares.
White has strong squares on e5 and g5, and possibly c5 even though Black can control that square with a pawn if necessary. F3 is a weak square though it is currently not easy for Black to place a piece there. White's pawn structure also does not have any real weaknesses.
Black on the other hand has (somewhat) strong squares on e4 and d5, but also several weak squares: e5, g5 and g6 are permanently weak, and c5 can only be controlled if Black is willing to compromise his queenside pawn structure. The backward pawn on g7 is weak and in an endgame the pawns on e6 and h4 are potential weaknesses as well.
6. Center and Space: The center is closed. White has a slightly more solid pawn mass in the center while Black's minor pieces better control the central squares. Neither side has an obvious space advantage. Both Black and White have more mobility on the queen side than the king side.
7. Development and Coordination among minor and major pieces: Both sides have completed their development. The white knight on the strong square g5 looks impressive, but it is unclear what role it serves there. This in fact may be a good example that a "strong" square isn't necessarily also a "useful" square. White also has the bishop pair.  The bishop on c4 is very strong indeed, indirectly targeting the Black's weak pawn on e6. The bishop could also pin Black's knight on c6, which could be very unpleasant for Black indeed, especially if White doubles up in the half-open c-file. However, White's main problem and in fact the defining characteristic of this position is the bad bishop on h2. The bishop doesn't contribute anything to White's game, and activating or exchanging it will require a lot of time.

Formulating a Plan:
After this in-depth analysis of the position, we started formulating a plan for Black. In my opinion assessing the 7 evaluation criteria is the easier part. It is more difficult to come to the right conclusion what the analysis of these criteria means, and which of the criteria are more important than others in any given position. For example, is it more important that White has the better pawn structure in this position, or that Black has two knights, which in a closed position such as this one should be favorable? It takes a lot of skill and experience to draw the right conclusions from the analysis. However, for Pyre's purposes it isn't important to get it right 100% all the time. In fact, even grandmasters regularly get this kind of analysis wrong. The point is that if Pyre learns to do this analysis regularly and systematically, he'll be able to take his chess to the next level.
In this example the critical point is to realize that White's bad bishop on h2 is the single most important characteristic of the position.
Basic chess strategy recommends that when one of your opponent's pieces is - temporarily or permanently - locked out of the game, the right course of action is typically opening up the game on the other side of the board because that is where in essence your opponent will be a piece down.
In this position Black can achieve that by playing c7-c5, ideally after bringing the rooks over to the c-file. The thrust c7-c5 opens the position without exposing Black's king too much.

And this is pretty much what Ivanchuk did in the game. The entire game can be found here:

The game ended in a draw because as so often Ivanchuk ended up in time trouble and couldn't convert the very promising position he achieved after opening the game on the queenside.
However, for the purpose of this exercise it isn't necessary to analyze the remainder of the game. The important point was to do an in-depth analysis of a given position with Pyre, which is something we'll keep doing going forward.

Vassily Ivanchuk and MarineKingPrime
A Word about Vassily Ivanchuk:

To me, Vassily Ivanchuk is the MarineKingPrime of the chess world. "Chucky" as he's called is one of the most brilliant players on the chess circuit, and has been a dominant force at the top level for many years. However, his inability to keep his nerves under control has prevented him from ever being a serious contender for the world championship title, or win more top level tournaments. In the 2013 Candidates tournament alone he's already lost 3 games on time.

Despite his volatile performance, Ivanchuk is one of the very few players who, on a good day, can beat anybody, including Carlsen, Anand, and the rock-solid Kramnik.

All of this is very reminiscent of MarineKingPrime who clearly has the skills to win MLGs but rarely does so because - among other things - it seems he is a little too emotional and nervous.

The following clip is from a chess olympiad match between Ivanchuk and Kramnik:

I'd be nervous too if I was playing Vladimir Kramnik, though 2700 ELO Super GMs typically play their moves a little more confidently than Ivanchuk does here.

Nevertheless, Ivanchuk is a fan favorite, and I have great admiration and respect for his creative chess, too!

Saturday, March 23, 2013

The 2013 Candidates Tournament in London

Magnus Carlsen during one of the early rounds of the 2013 Candidates tournament
We are almost at the half-way mark of the 2013 Candidates tournament in London. Most people seem to expect Carlsen to win the tournament.
Well... time is on Carlsen's side, therefore I would like to see Aronian or maybe even Kramnik win the chance to play a match with Anand. Carlsen will become world champion sooner or later anyway. In the meantime Aronian and Kramnik deserve a shot at the title. I like Aronian for his creative and tactical play style, and Kramnik is simply one of the most solid players around. His win rate isn't quite as impressive as it used to be, but he almost never loses a game, and considering the high caliber tournaments he's playing in that really is quite an achievement. However, with almost half the games played in London it is already pretty clear that Kramnik lost his chances to take first place. Right now it's a neck to neck race between Carlsen and Aronian. I hope Aronian comes out ahead, but I fear it'll be Carlsen. I have nothing against Carlsen, in fact I admire his seemingly effortless style of winning drawish positions. I just think success is coming a little too quick for him. If he has to work for the title just a few years more, he'll most certainly become a better player overall.
Vladimir Kramnik and Levon Aronian at the opening ceremony of the FIDE Candidates
Tournament in London
Garry Kasparov is on record saying that winning the candidates tournament is more difficult than beating Anand in the subsequent match. I'm inclined to agree.

Monday, March 18, 2013

Episode V - How much money does a chess grandmaster make?

Before I give you guys an update on Pyre’s progress, I want to take a look at salary levels in chess compared to SC2.

First a quick comparison of my chess earnings and Pyre’s SC2 income. Pyre was at some point ranked #1 on the North American grandmaster ladder. I used to be a 2200 ELO chess player.  This means that relatively speaking Pyre is much better at SC2 than I am at chess. 

I used to play chess semi-competitively for about 10 years until I stopped when I went to university. In those 10 years, I won about $2,500 in prizes. Mostly cash prizes at open tournaments, rapid, blitz or bughouse tournaments. But also many (worthless) chess books, and a digital chess clock.  Not bad for a kid in high school, but of course not nearly enough to consider myself a chess “professional”. In fact I am glad that it was always clear to me that my chess wasn’t nearly good enough to pursue a “professional” career.  

Pyre has made about $1,000 in SC2 so far. Approximately $800 in prizes (local LANs and WCS) and $200 from coaching.  Considering that SC2 only came out in mid-2010, he’s made his money a lot quicker.

Now lets take a look at what’s happening at the very top. The following two screenshots are from I’m sure the numbers are inaccurate but they should still serve as a rough indication of where these people stand financially. According to the site, the figures are the combined earnings of the players from 2010-2013. Click on the images to enlarge.

It is clear that a handful of Korean pro gamers is doing very well. At the same time though, the difference between Korea and North America is quite significant. I think there are a lot of people in the community who will disagree with me here, but I think that in order to consider yourself a “professional” SC2 (or chess for that matter) player, you need to a) play SC2 full-time, and b) make enough money doing it to support yourself financially.  So if for example someone makes $12,000/year playing SC2 but still lives at home with his parents, I wouldn’t consider that person a “professional” SC2 player, even though $12,000 is certainly quite a bit of money.

Chess professionals also complain that their sport is underfunded, and FIDE, the world chess federation, is certainly not doing a great job at attracting sponsors.

In her excellent article “Making Money in Chess”, Russian Grandmaster Natalia Pogonina gives the following ballpark figures for chess grandmasters’ earnings:

Global Top      3:  >$1 million / year
Global Top   10:  > $200,000 / year
Global Top   50:  > $100,000 / year
Global Top 100:  ~ $50,000 - $70,000 / year
(The article is a good read, and many observations apply to professional SC2 as well)

This of course also varies by year because the prize funds available to top players very much depend on the tournament schedule. For example, the prize fund at the 2012 world championship match between Viswanathan Anand and Boris Gelfand (one of the weakest matches in recent history) was $2.55 million, shared something like 65%-35% between winner and loser.
Earlier this week, the FIDE candidates tournament kicked off in London to determine a challenger for Viswanathan Anand for the 2013 World Championship later this year. The candidates tournament is an 8 player double round robin tournament with a total prize fund of 510,000 Euro, with 115,000 Euro for first place (and much more at the coming world championship match) and 21,000 Euro for last place.

For winning the 2012 World Blitz Championship (a 2-day event), Alexander Grischuk received $40,000
So at the highest level, chess players make good money. It’s not so easy when you aren’t part of the global top 50 or so. Grandmasters who have to travel the world to play in open tournaments – because they don’t get invited to the prestigious round robin tourneys – have a hard time supporting a family.

The Gibraltar Chess Open, one of the largest chess opens in the world, awards the following prizes:

This, however, is an exception and most open tournaments offer far less money to the winner. Therefore it isn’t surprising that many chess GMs are forced out of their careers sooner or later.
There have been many (mostly second-rate) chess grandmasters over the years who quit their professional chess careers or abandoned the game altogether in favor of other more financially secure activities. Most recently, poker has attracted the attention of quite a few chess GMs, but there are other examples as well:

After several years as one of Western Europe’s strongest chess players, German GM Dr. Helmut Pfleger stopped playing competitive chess in the 1980s to pursue his medical career.  The funny thing about Dr. Pfleger (whom I lost to in a simultan exhibition many years ago) is that he looks very much like my dad, dresses like my dad, shares almost the same birthday as my dad, and has the same medical degree.
Currently, British GM “Lucky” Luke McShane (just over 2700 ELO) works as a foreign exchange trader in London and plays chess tournaments only when time allows. They call him “Lucky” Luke because in his career he’s had many important games where his opponents overlooked simple wins.

Current State of Pyre’s Game:

Earlier this week Pyre managed to get a 1500 rating on for the first time! He started playing chess seriously only in January, so this is clearly a pretty impressive result indeed. Congrats, Pyre! And again I must say I’m very satisfied that as his coach I seem to be doing a few things right, too. He’s now set his sights on 1700, and as I argued in one of my previous articles, this is where the real work begins. Since the beginning of the year, Pyre has learned the basics of chess strategy, and he started ridding his game of some fundamental tactical errors. Now we can actually start playing some “real” chess, which is something I am very much looking forward to. In particular, we will be looking at some positional games by Capablanca, Karpov, Steinitz and Lasker.

We’ve also started looking at pawn endings. Knowing just a few principles about basic pawn endings is a good way to noticeably improve one's results. In his seminal work “Beyond Good and Evil” Friedrich Nietzsche warned us that “when you gaze into an abyss, the abyss also gazes into you”. I’m always reminded of this when I study pawn endings. Complex pawn endings are incredibly difficult and very often when I sit down to analyze them I realize that the more I look into them, the less certain I become about my judgment until at the very end I seem to doubt even my most basic conclusions. I’ll most certainly devote a future article to pawn endings.

The 2008 World Championship match between Viswanathan Anand and Vladimir Kramnik

Here are a few positions from Pyre’s recent games that allow some interesting observations:
The avid reader of this blog will remember that Pyre's account on is 3hitu. I noticed that sometimes Pyre is shell-shocked and throws away games when things don’t go as planned and the opponent suddenly plays some tactical shocker. For example:

In position 1, Black has just captured a knight on g3. Pyre probably thought that the knight was protected, but realized too late that due to the pin and Black’s bishop on c5 the pawn on f2 can’t recapture on g3. Therefore, Pyre resigned, probably frustrated over this “blunder”. However, he overlooked that White can simply play d3-d4. This breaks the pin and attacks the bishop on c5 so that White is going to get the piece back.
In position 2, Black has just played Nc2, and at first glance it appears as if White’s going to lose an exchange. Pyre seemed to think so too because he lost his head and took on a7 in desperate search for some sort of compensation. However, he could have easily salvaged the situation by playing Nd4 instead, forking Black’s queen and knight, thereby forcing the exchange of the two knights. This would have allowed Pyre to continue the game. Up to this point he’d held his ground against a much stronger opponent pretty well.
The lesson from these two examples is that one shouldn’t give up too easily when something unexpected happens, for example overlooking some sort of tactical threat. It’s tempting to just give in to shock and frustration and simply resign the game, but with a cool head it is sometimes possible to save seemingly lost positions.

In this position we have a pawn race on our hands. Pyre’s problem is that White’s pawn is one tempo ahead. In this position Pyre played the very dangerous move Kd5 to support the advance of his pawn. It’s the right idea, however, unless it is absolutely necessary (and here it is not) one should NEVER put one’s king on a square that allows the opponent to promote a pawn with check. It didn’t matter in this particular case, but it is advisable to avoid this risk altogether. In this position for example by playing Kc5-d4 etc.

This is one of my favorite games of Pyre so far. In position 1, he’s clearly lost. There’s nothing he can do to stop Black’s distant passed pawns on the queenside. However, in this position Pyre tried a final trick and played Nf6!, inviting Black to take the knight. If Black simply ignores the knight on f6 and plays a5-a4-a3 etc., Pyre’s position is hopeless. However, Black took the knight and in doing so gave Pyre two dangerous passed pawns. A few moves later they reached Position 2 which was Black’s last chance to draw the game. After … Ng4+! Pyre played Kh8, threatening g7+. Black found the only move … Nf7+!, but after Pyre’s response Kh7 he played … d2?? and resigned after g7+. If Black plays the knight back to g5 with check instead, he’d draw by perpetual check because White is forced to move his King between h7 and h7. If White takes the knight on f7, his pawns are blocked and Black can promote his d-pawn. If White plays Kh6 in response to Nf7+, Black wins a tempo to play d2 because g7+ is no longer a threat.
Obviously both players didn’t fully understand this position, but nevertheless Pyre’s move Nf6! was very clever and for that move alone I think he deserved to win the game. I was very impressed when I saw this game.

And one example from my own games:

This probably won’t come as a surprise to a StarCraft 2 audience, but in chess it is typically favorable to have the initiative. In many cases, it is even recommended to sacrifice material in order to (re)-gain it.  As an illustration look at position 1 below: My position is clearly worse.  I’m a pawn down, my queenside is falling apart, and I’m not sure how much longer I can hold on to the d2-pawn. On the other hand, the position of Black’s king has been compromised, and my knight would be very strong on f5 indeed.
Therefore, instead of defending a hopeless position, I decided to launch a counter-attack and much to my surprise after moves like Nh4 and Qg4+ we reached Position 2 rather quickly, with Black’s king spectacularly mated in the middle of the board.

In all seriousness though, teaching chess to Pyre has clearly shown me how lacking my chess has become over the years. I know I'm still a pretty good player, but not being as good as I once was is pretty frustrating. I’m not sure if I’ll ever achieve it, but I would like to take a shot at 2300 ELO at some point. At the very least though, teaching chess to Pyre has rekindled my interest in the game, and I haven't felt this excited about chess in a long time.

RevTiberius featured on "Inside the Game"

Recently this blog was featured on djWHEAT’s “Inside the Game”. The segment starts at roughly 37:10.

I certainly didn’t expect that so many people would take an interest in these articles, but of course I’m delighted that the SC2 community seems to find this interesting. 

I’d like to take this opportunity and comment on some of the things that were said on the show. I should clarify that I am blogging about teaching chess to an SC2 Grandmaster, but I do not necessarily believe that the two games can or should be compared in each and every aspect. On the contrary, the games are quite different, and only in a few areas are comparisons permissible.

Thanks for calling my articles “fascinating” and “fucking cool”. I appreciate the compliment. And you are right. SC2 is frequently compared to chess, and these – often erroneous – comparisons are one of the reasons why I decided to start this series.
I think you hit the nail on its head when you said that SC2 skills transfer to chess only to a very limited extent. However, the “meta-skills” (not sure if that’s a word) are most certainly transferable. To take Pyre’s example. His knowledge of SC2 build orders and strategies, his APM etc. are obviously useless in chess, but the same kind of skills he needed to become an SC2 GM in the first place (dedication, hard work, aptitude for strategy games etc.) will most certainly go a long way in helping him become a better chess player. The way I explain chess to him is also a bit different from how I'd explain it to a non-SC2 player (e.g. when I compare the typical Bishop sacrifice on h7 to a baneling bust)

I agree very much with Idra’s comments that the two main differences between chess and SC2 are real time vs. turn basedand perfect vision vs. imperfect vision. However, I think the second point is often misunderstood and needs clarification:
In SC2 the fog of war prevents you from seeing what your opponent is doing. But once you scout your opponent, it is usually easy to draw conclusions. For example, if you scan your opponent, and see a robotics bay that is being chrono-boosted, you don’t have to be Sherlock Holmes to figure out what your opponent is going to throw at you.
In chess, on the other hand, you have perfect vision of what your opponent is doing, but that does not mean you have perfect understanding of what he is up to. It is quite common to misinterpret what you are seeing. Misreading my opponent’s intentions is the reason behind many of my losses in chess.

This is also the reason why I - despite only reaching Top 8 diamond league in SC2 - can perfectly appreciate Code S SC2 games. However, despite being a very strong chess player, I do not understand many grandmaster chess games simply because perfect vision of what’s going on is not enough.

I very much agree with EG.Incontrol’s statement that many of the strategic concepts in chess and SC2 are best suited for comparison
For example when I teach Pyre that in chess when you are far ahead of your opponent, you can start trading inefficiently, he immediately understands because the same concept holds true in SC2 as well. 
Or the reason why many double-pronged attacks work so well in SC2 is closely related to the ideas behind the “Principle of Two Weaknesses” in chess.

@Mainstream Media
I also agree that the mainstream media doesn’t do a good job in comparing SC2 to chess. But that’s just because most journalists simply don’t know anything about SC2, and only very little about chess.
I regularly read The Economist, and once in a while they have an article about e-Sports. To me it always seems as if these articles were written by people who regurgitate second and third hand knowledge of SC2, and not by people who really know what they are talking about. The mainstream media also does a poor job at covering chess. The number of simple factual errors in their reporting never stops to amaze me.

Episode IV - Pyre Wins His First Trophy

It was only 2 months ago that I started teaching chess to SC2 Grandmaster Pyre. Therefore I was very surprised when he emailed me the following picture:

Pyre won his first trophy at a local high school chess event. Though I realized his talent very quickly, I did not expect him to score so well at over-the-board tournaments so soon. Of course most of the credit belongs to Pyre, but it is very gratifying to see that as his coach I seem to be doing something right, too.
I'd like to take this opportunity and take a closer look at the current state of Pyre's game:

Current State of Pyre's Game:

I regularly go through the games Pyre plays on to look for things that spark my interest. The following six positions I think exemplify Pyre’s recent progress very well, and also indicate where more work needs to be done. Pyre’s account on is [b][red]3hitu[/b][/red]. What I find most significant about the following examples is that they show that [b]Pyre has begun transitioning from simply making moves to formulating and executing plans[/b]. The very fact that he is already making short-term and long-term plans is significant progress indeed even if some of his plans are ill-advised or tactically flawed. . This kind of progress is far more important than fluctuations in his rating. I really don't care much whether Pyre's rating is 1100, 1200, 1400 or something like that. What I care about is improvements in his game.

In this position Pyre “saw” the hanging rook on d6 and took it. He won the game soon after, but 34. Qg8 checkmate would have been better instead. This is a good – though extreme – illustration of Emanuel Lasker’s recommendation “when you see a good move, look for a better one”. However, this is not just a problem for beginners. I’m very familiar with this kind of mistake, too.  I frequently overlook excellent moves after finding a good one.
In this example, Pyre’s move didn’t change the outcome of the game. It’s really quite frustrating when it does.

This position is a good example of how Pyre’s ability to formulate and execute plans has increased since the beginning of the year. In this position, he is a pawn down, but has a very strong attack against Black’s king. Pyre played 20.Ng4 which is a move that I am certain he could not have made 2 months ago. The idea behind it is to distract the knight on f6, which is the only defender of h7, where Pyre is threatening mate. I was very pleased indeed when I saw him play 20.Ng4.

In this position, Pyre’s bishop is under attack, and I’m sure that until recently he would have simply retreated it to maybe d4 or f6. In the game however, Pyre played 23.Qh6, threatening mate on c1. An interesting choice I find. I’m not sure if he simply overlooked 23. … Qxh2 or whether he didn’t like the response 24.Qh3. In any case, I was impressed by the fact the he put some thinking into the position and came up with something other than an obvious move with the bishop.

In this position, Pyre got really lucky. Black had sacrificed a knight on g4, and Pyre was imprudent enough to take it. Note to Pyre: In these kinds of positions taking a knight on g4 is almost never a good idea if Black gets an open h-file in return. Mate is usually inevitable. In this case, too. However, Black was too impatient and went for 10. … Qh2+, a useless check that allowed Pyre’s king to slip away via f2. Had Black played 10. … g3! first to block White’s escape route, there’s nothing Pyre could have done to prevent mate on h1. As the great Bobby Fischer said: “Patzer sees a check, patzer plays a check”.

In this position, Pyre has a nice bishop/queen battery on the diagonal b1-h7, and he eventually managed to win the game through an attack on the kingside. This position, however, is significant because both Pyre and his opponent overlooked a great defensive resource for Black. Pyre now played 11.e5, which is the right idea at the wrong time, because it allows Black to play 11. … Nb4!, forking Pyre’s queen and bishop and thereby trading Pyre’s important bishop on d3. So when you line up your pieces on that diagonal, it is important to make a prophylactic move like a3 first to make sure no knight shows up on b4.

This position is from the same game. Pyre now chose to play 19.Kh1 in order to play Ng1 to open the d1-h5 diagonal for the queen, a plan that ultimately resulted in Black’s resignation. Of course this was not a forced sequence. Black could have done several things to prevent this, and Pyre could have found a better way to activate his queen – one without moving the knight in the wrong direction. Nevertheless, I was very pleased when I saw this maneuver because it shows that Pyre is really developing the skills to formulate sensible plans. This is because while one should always look for the best move, from a practical point of view it is equally if not more important to have a realizable plan even if that includes moves that are - objectively speaking - not the best.

Bad Manner

Thankfully there is not nearly as much BM in online chess as there is in SC2. However, I noticed there’s a special kind of BM that I frequently encounter when I play someone who's much weaker than me.
Here are a few examples:

I'd be curious to know how much BM other chess players have to put up with. In top level chess tournaments rules around BM are very strict and players forfeit their games if they even just refuse to shake hands. This is a clip of the famous handshake incident at the Corus 2008 grandmaster tournament between Nigel Short and Ivan Cheparinov:

In my next article in this series I'll take a look at typical salaries of SC2 professionals compared to chess grandmasters. Any SC2 GMs or chess professionals willing to offer their thoughts on this are welcome to contact me.
I will also talk a little bit more about the theory behind formulating plans in any given position.

Episode III - Keeping up with the Kardashians

When former World Chess Champion Boris Spassky was asked in an interview whether he preferred chess or sex, he famously replied “depends on the position”.

A few days ago I was checking some chess news websites, just minding my own business, when suddenly I saw something that almost made me choke on my coffee:

Kim Kardashian wants to learn chess. Have we really sunk this low? I guess she got interested in chess when she heard that mating is the objective of the game…

So now the question is: Can Pyre keep up with Kim Kardashian? Of course he can. It may or may not seem that way to him, but over the past month or so I have already seen him make significant progress. Some of his recent thoughts and analyses are evidence of a much better understanding of the game compared to, say, 4 weeks ago.

[b]The ugly truth is that improvement in chess comes slower than improvement in SC2[/b]. I think Pyre once mentioned that he started in Gold League and became a Grandmaster within a few months.  Unless you are a prodigy, making comparable improvements in chess takes much longer.  One of the main reasons is I think mechanics. Even if you can’t play SC2 any “better”, you can always play “faster”. In chess, if you want to play BETTER, you need to play BETTER.

I think one of the reasons why chess is so fascinating to Pyre is that it must be so much easier for him to improve his chess compared to SC2. He already plays SC2 at a very high level, and for him it must be much harder to get the joy of getting better in SC2 compared to chess.
In my experience though I enjoy being good at a game very much, I enjoy getting better at a game even more.

Pyre recently already started showing initial success at offline/over-the-board tournaments. In my next article I’ll probably showcase one of this recent games to highlight some of his improvements.
And I’m gonna make sure he’ll always be better than Kim Kardashian…

The ability to recognize recurring patterns is an important skill of strong chess players. A significant part of my decision making in a chess game is based on intuition and experience rather than calculation and analysis.

For example, in one of my recent games I reached the following position:

I had just played Rook f8-h8 to counter White’s lethal threat Queen e3-h6. Generally speaking my position is a mess, White has a very strong attack and his Knights have two excellent squares on f6 and g5. Whether White’s position is objectively winning is unclear, but from a practical point of view his position is much easier to play than mine.
Now White committed a terrible blunder and played King e1-e2??, undoubtedly to activate his rook on a1. However, after my devastating response Bishop b7-a6 check the game is quickly over because to counter the check White has to either sacrifice his Queen on d3, or move his back on to the 1st rank, which disconnects the rooks again, and after I take on h1 it’s checkmate.

A few days later, I reached the following position in another game:

My opponent’s position is clearly better than mine: he has launched a very strong attack against my King. His Bishops and his Queen are targeting my king side, I was already forced to weaken my pawn shield by playing Pawn g2-g3. Black has just played Pawn h7-h5 with the intention to play h5-h4 to undermine my pawn shield even further and to open the h-file for his Rook.
To counter these very serious threats I had just played Rook a1-c1 to open the c-file, maybe in conjunction with Knight c3-b5 in order to force Black’s Queen off the critical diagonal b8-h2.
My opponent completely mishandled the position and a few moves later we reached the following position:

I don’t need to analyze this position in great detail for you. Suffice it to say that my next very strong move Bishop b2-a3 prompted Black to erroneously sacrifice his Queen to counter the check. Of course he lost the game soon after.

The lesson from these two examples is that I immediately recognized an important pattern in both positions: I had a fianchettoed queen side Bishop, and my opponent’s King was forced to come forward to e2 (or e7).  In these kinds of position I know that the Bishop check on a3 or a6 is usually lethal or at least very unpleasant.

I know this pattern, and when it occurs, I don’t have to calculate much because I’m already familiar with the implications. This also means that I never “overlook” a move like Bishop a3: while beginners have to “find” moves like Bishop a3, I already “know” it’s there and can immediately analyze its consequences when this pattern occurs in a game.

This also distinguishes a chess Grandmaster from a strong amateur player like me. A Grandmaster’s knowledge of typical patters greatly exceeds mine, and in many positions a Grandmaster simply “knows” what to do while I have to think about it.


I want to finish this article with a little retroanalytical problem. Typical chess problems ask you to solve questions like “Find Mate in 3” or “How does White win a piece?” Retroanalysis seeks to answer questions about previous events of a chess game. To solve such problems, you don’t have to be a strong player. Basic logic is all you need. For example, in the diagram below the question is

Has at any point a pawn been promoted in this game? The answer to this seemingly impossible question is surprisingly simple. I’ll post it here in my next article. Enjoy~

Sunday, March 17, 2013

Episode II - Initial Observations and Insights

This is the second in a series of articles about my efforts to teach chess to Pyre, a high-ranking North-American SC2 Grandmaster

I’m now about 3 weeks into my project to teach chess to SC2 Grandmaster Pyre.

As I described in the first article in this series, my initial task was to figure out how much he already knows about the game. To this end we’ve played a number of games and I’ve also observed some of his games against players at his level.
My initial assessment of his skills is this:
For someone who picked up chess fairly recently, he has a pretty good understanding of beginner level strategies and ideas. At the same time his knowledge of tactics is limited and as a result most of his losses are due to losing material to standard tactics such as forks and pins. This is by no means unusual. In fact it’s quite normal and the reason why in our sessions I mainly focus on tactical themes. I noticed that Pyre has quite a good understanding of what needs to be done once I've explained a position to him, but he's sometimes lost when he tries to make sense of a position on his own. That, too, is perfectly normal, and I have no doubt that over time he'll develop his "game sense" to know what needs to be done. It's a combination of experience and intuition, and it takes a while to acquire.

One of the very common tactical themes we discussed so far is “deflection”. The basic idea is to deflect one of the opponent’s key pieces – usually through sacrifice or check – so that it can no longer fulfill the task assigned to it. It so happened that in one of my own recent games on a very interesting example of this came up (you may have to click on the diagram and enlarge it):

I was White, and I had sacrificed an exchange in order to reach this position. My thoughts were kind of like this:
- Black’s King is still in the center and kind of naked
- I have a rook on the open f-file
- My Queen is threatening to take on e6
- My knight can easily join the attack via f3 and then either to e5 or f5
- Other than being an exchange up on me, Black has no immediate threats

I wanted to take on e6 with my Queen, but realized that after Black plays Queen to e7 in response, I have no further checks and no way to continue the attack.  But still Queen takes on e6 is obviously the way to go, so I had to find a way to distract Black’s Queen temporarily. And here I brilliantly (though I say so myself…) found the move Pawn d4 takes c5. This move activates my previously useless Bishop on b2 and pins the Black Queen to the rook on h8, thereby inviting the Queen to take my Bishop. And that’s what Black did, but after Queen takes Bishop, I take on e6, followed by Rook to f7, and Black has no defense against checkmate. A nice little example of how powerful the deflection of a defending piece can be.

It has to be said though that Black, instead of taking my Bishop, could have and should have played Pawn to e5 to close the dangerous diagonal. However, this move is not so easy to see as evidenced by the fact that both I and my opponent (both of us pretty strong players) overlooked this move during the game. I only discovered it after I spent some time analyzing the position after the game was over.

In any case, this game was another important example that tactics really are the “mechanics” of chess, and that one must have a good understanding of chess tactics in order to become a good player.  Also, “Deflection” is closely related to the idea of “Decoy”, which is a maneuver I’ll introduce to Pyre in our next session.

I can’t emphasize enough how important it is as a beginner to study chess tactics. Doing just one or two chess puzzles a day can make a really big difference.
A 1500 player doesn’t really play “better” chess than a 1000-1100 player like Pyre, he just makes fewer tactical mistakes.
The same holds true for me, too. I’m a 2100 player, and most of my games are won and lost on tactics. Bottom line: practicing tactics is like taking the Elovator…

But of course we don’t just do tactics. We are also looking at famous chess games. So far I’ve shown Pyre 2 classics:

Nigel Short – Jan Timman, Tilburg 1991
Anatoly Karpov – Victor Korchnoi, FIDE Candidates Final 1974

I picked these games not only because they are brilliant achievements by the players and really fun to go through, but more importantly because these games are based on straightforward strategic plans and feature many of the tactical patterns we have already discussed.

One of the next games I’d like to show him is one of Capablanca’s positional masterpieces as an illustration that one doesn’t always have to launch an all-out attack in order to win the game.

Tactics in Chess = Mechanics in SC2?
I’m not trying to say tactics in chess and mechanics in SC2 are the same thing, but I think it’s an interesting comparison. Without solid tactics, one will never be a good chess player because one will keep losing too much material needlessly. Tactics are the foundations upon which a good player executes his strategies and plans.
Similarly, in SC2, strategies and build orders are useless if one doesn’t have the mechanics to execute them efficiently.

Transferable Skills?
Pyre is undoubtedly a very strong SC2 player.  But does that make it easier for him to learn chess? And are chess players predestined to become strong SC2 players? I don’t think so. Though I think it is fair to say that if one likes strategy games in general, it wouldn’t be unusual to enjoy both games, as different as they may be.

Chess Software and Videos
So far I’m not introducing chess software in our sessions because I think this would be a great disservice to Pyre. That’s like giving kids in first grade math a calculator.
I believe it is essential – especially for beginners – that one reaches one’s own conclusions as erroneous as many of them may be in the beginning. If you use chess engines too early in your development you’ll distrust your own judgment and never build up the confidence necessary to make it through complex calculations.
I also have mixed feelings about watching chess videos. Of course they are entertaining and do help to some extend. But at the end of the day it’s like watching Day9’s SC2 commentary. It’s fun, it’s instructive, but unless one actually plays SC2 and practices what he preaches, watching his videos alone won’t make one a better player.

Having said that, I think among all the material that is available on YouTube, Daniel King’s channel clearly stands out:

As much as I recommend Daniel King’s videos, it’s just no substitute for practice. Chess is fun, but there’s no denying that it also requires some work.

Feedback and suggestions to my articles are highly welcome. Especially from people with experience in online coaching. I have a lot of experience in both coaching people at chess as well as being coached, but I've never done it online. I'd like to hear from people who have experience in online chess lessons.

I’ve been asked why I’m teaching Pyre for free when an experienced chess coach of my skill level typically charges about $40/hour.
The answer is quite simple. In my chess career I’ve already played hundreds of serious tournament games and literally thousands of blitz and bullet games. While I keep enjoying this, my marginal utility (yes, I have an econ degree among others) of PLAYING chess is somewhat decreasing. On the other hand, TEACHING a friend to get better at chess gives me more satisfaction than simply winning a few more blitz games. And when I pursue my hobbies, I have no financial interests. It kind of defeats the purpose.


My friend Pyre is a Top 16 Starcraft 2 Grandmaster and arguably one of the strongest Terrans on the North American ladder.

Recently he asked me whether I would help him become a better chess player, and naturally I accepted the challenge. The avid reader of this blog will recall that I never was a great Starcraft 2 player. I retired at the top of my diamond division. However, I used to be a 2200 ELO chess player, and though I’m not quite that good anymore, I’m still a pretty strong player.

Pyre’s current online rating is around 1050. His goal is to reach 1500.  I know that this seems a daunting task to many beginners, but I’ve played a great deal of chess over the years and in my opinion the difference between 1000 and 1500 isn’t all that great. Going from 1000 to 1500 is certainly easier than going from 1500 to 2000, or, as I myself had to learn the hard way many years ago, going from 2000 to 2500.

In my opinion, the only real difference between a 1000 player and a 1500 player is that the 1500 player has a better understanding of basic chess strategy and tactics. However, the upside is that basic chess strategy and tactics are fairly easy to learn, and in my experience the real litmus test of a chess player is not whether he or she reaches 1500, but whether they continue to develop from there because once you have mastered the fundamental concepts of chess, it gets a little harder to improve further.

I also should mention that I firmly believe that anybody who is serious about becoming a better chess player MUST also play the game offline over-the-board. Online games and chess lessons are a lot of fun and certainly help, but the best way to get better at chess is to play and analyze games at your local chess club.

So now the question is how do we actually get Pyre’s skills from 1000 to 1500? My coaching style combines elements of Gandalf, Obi-Wan Kenobi, and Gunnery Sergeant Hartmann.

Currently we’re still getting started and regularly play longer games so that I can get a better understanding of what Pyre already knows and where his weaknesses are. For the same purpose I have also flipped through some of the games he’s played against people at his level. At the same time, our games provide a good opportunity to discuss some of the tactical and strategic themes that arise. For example, the following game provided an opportunity to discuss the so-called “smothered mate”, arguably one of the most dramatic maneuvers in chess.

These games are also important because I need to know whether Pyre is a more tactical/aggressive player, or prefers more strategic/positional games. Based on his macro-heavy/no-nonsense approach to Starcraft 2, I kind of expected him to be more on the strategic/positional side, but in chess he appears to prefer more aggressive, all-out attack on the king kind of games. I found that pretty interesting.

Once I've concluded my assessment of his style, we'll start working on more technical aspects of the game, such as the fundamentals of

a) Strategy: e.g. developing minor and major pieces, pawn structures, typical plans, activity and initiative etc.

b) Tactics: e.g. how to calculate variations efficiently, how to choose among several moves, and of course typical beginner level tactics such as various check mates, forks, pins and skewer, zugzwang, double attacks etc.

c) Opening Theory: I firmly believe that beginners and players below maybe 1600-1700 really do not need to know all that much about the different chess openings. At that level a player is much better off dedicating his or her time to the fundamentals of strategy or tactics. However, Pyre and I will spend some time on some of the basic chess openings to make sure he plays those openings that result in the kind of positions he's comfortable with and enjoys playing. To reiterate: It is NOT necessary for him to memorize all the different lines of the French defense, for example, but he does need to form an opinion on whether the kind of position that typically comes out of a French opening is something he wants to go for.

Over the coming weeks and months I will chronicle our progress here including lessons learned, successes, failures, and interesting observations. In the meantime I encourage everybody to check out Pyre's stream (link above).

And no, I am not getting paid for these lessons, my awesomeness comes free of charge :-)