I've always had a problem with what to respond to 1.d4. I hate all the positions that arise after ... d5, but was not introduced to ... Nf6 when I was learning to play chess, so I still don't know which Indian to play. I tried the KID for a while, but didn't like it. I've had the Nimzo recommended, but haven't learnt it yet. So mostly I've muddled along in the KID or some boring 1 ... d5 opening, making up lines as I go along.

Today, when an opponent pulled out 1. d4 on me, I decided that as long as I'm making up lines, I may as well make up an opening instead, so the positions are unfamiliar to my opponent as well. So after ... Nf6 2. c4, I tried ... d6, with the idea of Nbd7 and an immediate c5 or e5 to strike back in the center. I planned to defer the development of my bishops until I had more of an idea of where they would be well placed.

Well, there's nothing new under the sun, and it turns out this goes by the name of the Old Indian Defense. Wiki notes that it is "considered sound" but since I've never heard of it before, I assume the problem is that it is inferior to the KID because of the inactive dark-squared bishop (although no doubt it could transpose to some lines of the KID)? Is this opening still played at Master level chess at all?

Here's the game I played in the opening. Comments welcome I guess, not sure how much there is to comment on. My patzerish analysis (without aid of an engine) follows: all the moves look reasonable to me up until 7. f3?!, which looks a bit weakening when I am going to have more space on the K-side. The developing Nf3 or Nge2 looks better. The pin on the knight won't be a problem in this position. Then 10. h4? deserves the full question mark. Obviously the plan is to trap my bishop, but even if I had been forced to play h7 and then retreat my bishop there, it looks to me like White is over-reaching, with two minor pieces still at home, pawns thrust forward all over the board, and an exposed king with nowhere to go. As it was, I was able to get three pawns and the bishop pair for my piece, after which the position looks like at least =+ to me. My opponent then blundered with 15. Nf3 which loses on the spot. I'm not sure I wrapped it up as quickly as I could have, but I was just making sure I didn't blunder back.

[Event "ICS Rated standard match"]
[Site "freechess.org"]
[Date "2009.12.07"]
[Round "-"]
[White "appleberry"]
[Black "ChrisV"]
[Result "*"]
[WhiteElo "1701"]
[BlackElo "1681"]
[TimeControl "900+15"]

1. d4 Nf6 2. c4 d6 3. Nc3 Nbd7 4. e4 e5 5. d5 Nc5 6. Bd3 Bg4 7. f3 Bh5 8.
b4 Nxd3+ 9. Qxd3 Be7 10. h4 O-O 11. g4 Nxg4 12. fxg4 Bxh4+ 13. Kd2 Bxg4 14.
Kc2 f5 15. Nf3 fxe4 16. Nxe4 Rxf3 17. Be3 Bf5 18. Rh2 Rh3 19. Rah1 Rxh2+
20. Rxh2 Bg3 21. Rg2 Qh4 22. Nf6+ Qxf6
*

we need the pgn noob

nice ninja edit

it is playable but very passive.

5. d5 seems premature, because there's no bishop on g7 that pressurises d4 like in the KID white can (and should) keep the central tension longer. Playing d5 sometimes allows black even queenside counterplay because he's basically a tempo up on the KID (...e5 and Be7 compared to ...e5, g6 and Bg7). The bishop can also be activated with a Bd8-b6 maneuvre.

Yeah I think you're right, being basically a tempo up on the KID was part of my plan in the opening, since I found fairly frequently in the KID I would end up with pawns on d6 and e5, with my g7 bishop useless.

6...bg4 was bad,
on that f3 is good move, instad of b4, he'd should go with nge2, be3, qd2, 0-0-0 followed by g4 and your bishop is aghful.

I think sometimes the plan in this variation involves not gaining a tempo, but losing some by playing g7-g6 and Be7-f8-g7

actually usually they play here ne8-g6-ng7 and f7-f5

The Old Indian is definitely playable, and not _that_ passive in my opinion. You can also transpose to the KI early on (Nbd7 variations have been making something of a comeback lately).
Ways to make something out of the dark squared bishops also include exchanging it via Be7-g5, and sometimes in the Old Indian it also is manouvred to c7/b6/a5 by the way of the manouvre Qe8-Bd8.
By the way, 3.-Bf5 is also a good way to get into fairly uncharted and playable waters.

It's eminently playable - what do you play vs. e4? Might fit into a broader repertoire involving the Pirc/Philidor.

a lot of people play the move order 1.d4 d6 2.c4 e5 which gives Black a few more opportunities compared to the 1...Nf6 lines. 1.d4 d6 2.Nf3 Bg4 is considered playable for Black or you can play Nf6 and transpose into "normal" positions. The disadvantage of this move order is 2.e4!

The old-indian is perfectly sound, but passive.
Too many lines that looks like an inferior Kings-indian for my taste though.

I can't say I'd recommend the Old Indian to students. Here's my reasoning:

(1) If you are under about 1600-1800, you should probably choose a classical response. The Tarrasch is fine and more or less sound.

(2) If you are over 1800, you need to begin developing a real response to d4, so either the King's Indian or the Nimzo-Indian.

Basically, I don't have much place for it in the repertoire. That's not to say it's bad, I just don't recommend it.

Specific thoughts:

(a) 5. d5 is fine. The main line is 5. Nf3, but 5. d5 actually performs better. Incidentally, it's fine to play an early d5 in the King's Indian too.

(b) 5...Nc5 isn't right. It's the main line, but it's not good and hasn't been for a while. I'd definitely recommend 5...Be7. He should answer 5...Nc5 with 6.f3 followed by 7.Be3. The knight still might go out to c5, but better to retain flexibility.

So you're saying, "I can't recommend it because I don't recommend it for any rating group." Obviously true, but sort of lacking substance. What is it that you don't like in the opening?

There is a player group I can recommend it to.
Veteran players who have played some similar openings in the past, like KI, benoni and no longer wish to spend oceans of time preparing these openings.

I dont think its a bad opening for players starting out either, but that group will probably want to switch to the Kings indian later on.

I'm more curious about this part of his post. Why do you restrict real responses to 1. d4 to just the KID/NID?? What about the slav, semi-slav, various QGD's, Gruenfeld, etc, etc??

I think that the idea is that the KID/NID have more easily understood plans suited to players of that level and less positional manoeuvering.

Like I said: too hard / modern for weak players, not good enough for strong players.

Fair enough. I always recommend that students fully learn their opening. I think it takes much less time to learn a narrow repertoire than people think, and I don't really find much of an excuse for being unprepared.

If you've played the King's Indian / Benoni and are looking for a lower maintenance option, you could do worse than the Old Indian. Still, I would just recommend the King's Indian. There is a whole lot of theory in the King's Indian to learn in order to stay right at the cutting edge, but you can learn enough to play the line pretty easily (and you needn't really refresh it).

You only really need to be state of the art above master strength (and you need to spend time preparing your openings then).

I always recommend the King's Indian or the Nimzo-Indian to serious players developing their first real response to d4. These are the two most common high level responses to d4, and if you are going to be a strong chess player you should understand them.

I recommend attacking players learn the King's Indian and positional players learn the Nimzo. I tend to pair the King's Indian with the Sveshnikov and the Nimzo with the Najdorf.

Over time, and based on the students results, I recommend changes. I recommend QGD's below 1800 more or less exclusively, and occasionally return to it if the student struggles with Indian defenses. The Gruenfeld I usually recommend to players who are struggling with certain parts of the King's Indian.

I play the Slav in tournament chess but I have never had a student strong enough to recommend it to. The Slav is either too boring (the quiet lines) or too theoretical (the main lines) for my students.