The scheme d4/Nf3/e3?bd3 is completely ineffective against a KID setup, and the Stonewall scheme is even worse. White needs something more sophisticated.
It is harder to argue with you, as you are a IM, but sophistication isn't all that important in chess. The colle is just a system of development which uses pieces to push both d4 and e4. Playing systems with a lot of complex tactics and positional play can lead to absolute disaster, thus sometimes simple is better.
The scheme d4/Nf3/e3?bd3 is completely ineffective against a KID setup, and the Stonewall scheme is even worse. White needs something more sophisticated.
It is harder to argue with you, as you are a IM, but sophistication isn't all that important in chess. The colle is just a system of development which uses pieces to push both d4 and e4. Playing systems with a lot of complex tactics and positional play can lead to absolute disaster, thus sometimes simple is better.
Nah, what the IM is saying is that your setup provides no resistance to Black's ideas, which is more likely to lead to 'absolute disaster' than playing smart/sophisticated moves is.
Besides, @pfren suggested that simply placing the bishop on e2 is an improvement, which is hardly a complicated change to make.
KID has been analysed for decades!
We dont need to reinvent the wheel.
Here are T40 nets Leela choices!
The idea for white is to play on Queen side as fast as possible and here is one of moves against main line.
9. b4!! is Leela's choice, it immediately threaden to push c5! And after a couple of moves, white has plan to push c5! The plan is to push a5 , exchange pawns and then c5!
The main stratategy is simple,
If white can damage black queenside by c5 as fast as possible, white win. If black can play f5 as fast as possible black win. Generally white is a little bit faster in accurate play!
Another possible variation of classical King Indian, if black play Nh5!( or Nbd7,Ne8) for f5 play!!
There are not much variation for black to play against white' Nf3, Be2 classical set up. These are common variation for black
1. e5, then f5 break
2. c5 break
3. c6, Qc7 or Qe7
4. a6,c6, b5 break.
5. a6, Rb8, b5 break
But as long as white can play faster in queenside, white is ok!
If black totally neglect queen side,
In human games, inaccurate play in either sides cause the loss!
(Human will always do frequent inaccurate play) 
Well... if only so easy?!
Off the top of my head...
KID variations
Classical, Saemich, Orthodox, Petrosian, Gligoric, Fianchetto, Exchange, Smyslov, Makogonov, 4 pawns attack, Mor De Plata, Kazakh, Uzbek, Panno, Byrne, Benoni style, Averbakh, Yugoslav, Bayonet attack, London style, Wojo version... and am sure i have missed some glaringly obvious variation?!
And... is brilliant. For Lion is the most beautiful Chess opening.
From white player point of view, white already chose classical. So no need to discuss other variations.
check post #28, major variations are all mentioned.
From white player point of view, white already chose classical. So no need to discuss other variations.
check post #28, major variations are all mentioned.
Sorry to say that you have barely scratched the surface.
The whole line is quite impractical to play OTB, because 1. white does not get much, 2. He has no "easy" moves at his disposal, 3. Black's "inferior" deviations from the mainline are all but easy to "refute".
Apparently somebody let the engine run for 24 hours in the starting position and it decided the Blackmar Diemer Gambit was the best opening. I'm slightly curious as to what engine, but it's still more anecdotal evidence that following the engine in non-forcing opening lines is silly.
Yes, engines don't like the KID much. People keep writing of this opening, convinced White should win... love the anecdote about Mr KID Slayer... who then lose five straight?!
I know an engine crazy guy that analysed King's Indian defense with an engine and decided it's an almost forced loss for Black. He tried his engine lines and scored the "amazing" 0/5 in his last 2 tournaments as white against King's Indian. This is what is happening with people that hardly understand what they are talking about and only rely in useless engine evaluations.
The funny is, he insists that King's Indian is a forced loss for Black!
Or probably Kramnik's collection of games in the KID as white... Very concrete strategic planning, paired with CORRECT engine usage/opening preparation, and not just staring numbers running on the screen, like most people do.
I know an engine crazy guy that analysed King's Indian defense with an engine and decided it's an almost forced loss for Black. He tried his engine lines and scored the "amazing" 0/5 in his last 2 tournaments as white against King's Indian. This is what is happening with people that hardly understand what they are talking about and only rely in useless engine evaluations.
The funny is, he insists that King's Indian is a forced loss for Black!
I know about your stage of level. You are just in a stage of knowing how many lines/ variations of this or that opening.
I know ECO 500, know all variations of KID. My preparation is for precise play as much as I can. Bookworm like you knows dozen or 100 of variations only, but you dont know which line to choose for your opening repertoire.
You have big mouth, but little achievement here. If you think you are better than engine, I suggest to see a doctor for your mania.
I challenge you in 3 min blitz. Show me how good your preparation/knowledge/skill is.
If you are chicken, then run away ok!!!
Engines are not perfect, but ways better than human. I am not just regular engine user, I know most of evaluations, how does those engines works. Most people dont understand the evaluation of engines ,but I understand every +0.01 where and why those come from.
For example,
Here is Stockfish evaluation
https://hxim.github.io/Stockfish-Evaluation-Guide/
If I use Leela, I know how she works.
Go OTB club and talk those stuff.
This is online chess. Most of online games are played in 3 min blitz. I am not challenging you in 1 min bullet where time factors significantly involve in outcome of the game.
2500 GM can comfortably play 2500 level blitz.(average blitz rating)
2000 OTB players can play 2000 blitz.( average blitz rating)
If you think you are 2000+ OTB, proove it.
There is difference in stage of chess, your level is sticked to theory only. My posts are about precise play and constructing a good opening repertoire.
Yes, engines don't like the KID much. People keep writing of this opening, convinced White should win... love the anecdote about Mr KID Slayer... who then lose five straight?!
Losing 5 straight would be fine. More likely is the guy was playing people hundreds of rating points over him and he lost. If he had won all 5, that wouldnt mean a thing either. So much focus on winning and losing, but not learning.
And regarding chess knowledge and move choices of Leela, this is how GM Larry Kaufman ( Author of Komodo engine) see Leela in today discussion.( A/B engines means Stockfish, Komodo and Houdini)
All my previous posts are from Leela , not from stockfish. ( It has already proven in many tests, Leela is superb in opening and positional understanding but bad in tactics and endgame , stockfish is superb in endgame and tactics but weak in opening and positional understanding).
My main reason for playing e4 in KID is that the large pawn center Black gives can either stifle the Bishops or if Black plays c5 before castling the dc dc Qxd8+ Kxd8 Nge2 gives White a Maroczy Bind setup. I will play the classical with f3 and work to create a battery on the c1 h6 diagonal to try to weaken the dark squares around the King. The main ideas that I have to contend with are the Ne8, f5 lines. Playing the Colle System is safe but in a double edged opening, why give Black any breathing room. Also, if White intends to play for the queen-side and move his attack through to the King, it would seem that Bb2 hampers this. I don't have a lot of theory of engine work to draw off of, but it seems that the Colle while safe is not going to be totally sufficient against a week prepared King's Indian player.
My main reason for playing e4 in KID is that the large pawn center Black gives can either stifle the Bishops or if Black plays c5 before castling the dc dc Qxd8+ Kxd8 Nge2 gives White a Maroczy Bind setup. I will play the classical with f3 and work to create a battery on the c1 h6 diagonal to try to weaken the dark squares around the King. The main ideas that I have to contend with are the Ne8, f5 lines. Playing the Colle System is safe but in a double edged opening, why give Black any breathing room. Also, if White intends to play for the queen-side and move his attack through to the King, it would seem that Bb2 hampers this. I don't have a lot of theory of engine work to draw off of, but it seems that the Colle while safe is not going to be totally sufficient against a week prepared King's Indian player.
If you're committed to this idea, consider putting your LSB on e2 instead of d3. On d3, it does very little apart from playing a minor support role in a potential e4 break that you're not really aiming for in this line, anyway.
On e2, it at least defends g4 in a lot of lines, helping to deflect black counterattacking chances while you try to orchestrate the exchange of DSB's.
You might find Yaacov Norowitz's material on the subject instructive, since this was his anti-KID setup back when he was mostly playing stonewalls.
This is a very good anti-Grunfeld and anti-KID system (not just for Yaakov's beloved blitz and bullet, but for any time control), but Black may pick a Benoni setup, where the pawn at e3 does not make much sense.