A mildly funny observation in the King’s Indian

So, I find myself in the middle of a homework assignment and my brain is desperately trying to find a way to procrastinate. Writing 500 words about didactic learning theories currently feels like undergoing a medieval torture method, whilst writing 2000 words about some random chess opening no issue at all. And I can safely guarantee that the latter has much higher quality than the former.

Why do I encounter the King's Indian?

Alright, so I wanted to talk about openings, the King’s Indian in particular. I am relatively bad at remembering deep opening lines and principles (I feel). I have never been greatly interested in reading large amounts of classical literature, and the chess opening book is no exception. I rather read a dictionary. So, naturally, I developed some skill in avoiding main line theory and steer towards playable (and sometimes unplayable) side lines.

As black, I have some experience with the King’s Indian defense. Against 1.d4 I usually go 1. ...Nf6, and if I consider the stars to be properly aligned, I follow white in the main line 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 0-0. And to avoid the full main line I typically throw in a well-timed Na6!!, and life goes on.

Now, something new happened ‘recently’. I started playing the Modern opening as black against 1.e4, which looks something like 1. ...g6, 2. ...d6, 3. ...Bg7. White has many options, but most of them consist of randomly pushing forwards n of their central pawns, with n being a random integer between 2 and 4. The attentive reader might realize here that for n=3 and with the chosen pawns e, d and c, the resulting position becomes dangerously similar to the typical Kings Indian setup.

In fact, the only difference is the move order. The first example started with 1.d4, the second example 1.e4. Now, I hate dividing people up into groups (just kidding, I love doing that to be honest), but there is some truth in the statement that some people are ‘1.e4 players’ and some are ‘1.d4 players’. Not applicable for everybody, but definitely for some. And even though the resulting position is identical, I start to notice a difference in how both types of players treat it. (Spoiler: the d4-players turn out to be superior, and as such a player myself, this aligns nicely with my own arrogance.)

Where to put the light squared bishop? On e2 or d3?

Sooner or later, white finds themselves in the mood to castle, and upon realizing that castling short is most convenient, they find themselves with some stupid bishop on f1 ruining all the fun. With minor frustrations they spend a move getting it out of the way, some call this behavior ‘developing’, and they have to choose between Be2 and Bd3.

Here it starts. The King’s Indian is most commonly reached via 1.d4 and and 1.d4 players tend to be trained to put it on e2. I think this just originates from opening knowledge. However, 1.e4 players seem to play Bd3 more frequently. Surely, there is of course a huge variation, but I rarely encountered Bd3 previously. Since I started playing the Modern as black, it seems to happen somewhat frequently. I thought about it, and it does make some sense. Perhaps it’s just my subjective observation, but many lines in 1.e4 openings consist of throwing out the bishops as far a possible. Further is better. And Bd3 is furthest in this case. Fair enough, right? Or am I too barbaric now?

Unfortunately, it turns out the bishop on d3 is not very good. There are a few reasons. Black usually plays ...e5, after which the central e-pawns are locked and the poor bishop ends up being stuck, looking at its own pawns on c4 and e4. If white decides to lock the whole center by playing d5, black typically responds by launching a pawn storm on the king side. See the analysis board below. The bishop on d3 is not really effective in stopping blacks ...f5 push. On e2 however, it tends to be quite useful in stopping/delaying the ...g4 push. Furthermore, when white goes for the main line plan Ne1 and f3, the knight would love to go to d3 next, and it would be a shame if there were to be a stupid, dumb, pointless bishop sitting there like an unproductive co-worker.

The bishop is not much better on d3 when the center explodes. Both white and black can exchange in the center, I focus on whites dxe5 first. If this trade ever happens, and black takes back dxe5, the central e-pawns are still locked, but the d-file has opened up. Play starts to revolve around this open file. Black would love to put a horsie on the weakened d4-square, for example. Pinning the defending knight on f3 with a well-timed Bg4 goes along well with this plan. The light squared bishop on d3 is just really annoying for white. It is sensitive for tactics (like, a black knight on c5 and a black queen/rook on d8, for example), it prevents white from playing Rd1 with tempo on the black queen. On e2, it would at least help defending against a possible pin.

When black takes exd4 and white takes back Nxd4, similar issues arise. Admittedly, the bishop defends to e4-pawn that will soon be attacked along the half-open e-file, but usually f3 is preferred as a defender. The bishop on d3 is somewhat blocking whites activity on their half-open d-file, and once again it is really sensitive to tactics, especially by jumpy knights. Also, the knight on d4 lacks its natural defender, because the bishop is blocking the file.

I think there are many games where white ends up castling, moving the f1-rook to a better square, and tuck the not-very-loved bishop away on f1, where it can observe all the action from a far.

Considerations

I am not a chess expert. Others are much better. I think I raised some fair points about the difference between d3 and e2, but I might have missed stuff or talked non-sense. My point is however strengthened by statistics. Master games within the Lichess openings tree state that 6. Be2 is played 45.038 times, whereas 6. Bd3 is played only 111 times. Thus, the former is preferred 405 times more often than the latter.

In a more surprising turn of events, the strongest and fishiest engine on the planet is not too harsh about 6. Bd3. In my highly scientific experiment, I limited the Lichess engine to very low computational resources and time, giving me the result that 6. Bd3 is considered +0.3, whereas 6. Be2 is slightly better evaluated as +0.5. Some difference, but nothing spectacular.

Also, I don’t know whether my observation that 1.e4 players and 1.d4 players play the position differently, is statistically sound. I noticed the Bd3-move like once or twice in a game, and recently it happened again, so I remembered it. Let’s call it a hypothesis... (Good scientific texts should always have the central hypothesis right at the end.)

Implications

This blog post is probably not the most shocking ever. In fact, I wonder whether it will have significant impact on other peoples lives. However, I feel more confident steering chess games towards a King’s Indian defense when people started with 1.e4. Even if this confidence is false, please keep me in this delusional state of mind, I strongly need it.

Conclusions

The main conclusion I would like to share, is that writing this text took a little over one hour. Albeit, without diagrams yet. Writing equally long school assignments takes a more week... Life is hard.