There are a few recent games (Godena-Short and Godena- Bacrot are
two examples) where the games started as in the Miles (c6) and after a few moves become a normal Kengis (g6)
5...c6 6.Nd2 g6 7. Ndf3 Bg7 8. Bc4
Here one wonders whether Bc4 was really necessary since this move
justifies c6. Perhaps with the "slow" 8. Be2, White would be able to
show that 5...c6 was passive and that Black lost one time with it.

About the Kengis, I think that a long time ago there was a discussion
here about the merits of 6. c4 (after 5....g6) and that it might be the reason (as Markovich said) that the Kengis is not played anymore. Since I was not involved in the discussion I do know who was but I remember a comparison of some lines involving either Be3 or Bf4
from White.

Anyway, I do not think that the other move, 6. Qf3 "might conceivably refute the Kengis" as John Cox says in his book.

[quote author=Patzer link=1239356947/0#5 date=1239478821]

Perhaps because of 6.c4. 
If Black wants to play this structure I think he is better advised to play 5...c6. This avoids the c4 line (6.c4 Nb4!), and black keeps the option of transposing to the Kengis-structure.[/quote]

Except that 5...c6 is not really all that necessary in the Kengis.  Black often wants to play ...c5 directly.  5...c6 is a passive move; 5...g6 isn't.  The big mystery is why the former is more popular than the latter.  I do think it has to do with 6.c4, but I'm not sure if this is justified.

On

"I remember analysing this line a little while ago with 6.Qf3 Be6 7.c4 Nb4 8.Qxb7 Qxd4 9.Qxa8 Nc2+! 10.Ke2 Qxe5+ 11.Kd1 Nxa1! 12.Qxb8+ Kd7

the main problem I had with it was that it's impossible to avoid a draw after 13.c5, as White is threatening Qb5+ and a draw by repitition"

AGREED!

Michael

An update on my previous post

On 6 Qf3 Be6 7 c4 Nb4 8 Qb7 Qd4 9 Qa8 Qxe5+! (instead of 9... Nc2+) Cox recommends 10. Be2 ! (my evaluation) and in fact after seeing a few lines, I would even say that 9.... Qxe5+ is ?!.

Going back to your line  6. Qf3 Be6 7. c4 Nb4 8. Qb7 Qd4 9 Qa8 Nc2+
10 Ke2 Qe5 11 Kd1 Nxa1 as suggested by you seems much better
than the old 11. ... Qd4+ 12. Bd2!

The point is that after 6. Qf3 Be6 7. c4 Nb4 8. Qb7 Qd4 9 Qa8 Nc2+
10 Ke2 Qe5 11 Kd1 Qd4+ 12. Bd2! Nxa1 13. Qxb8+ Kd7 14. c5 is
quite strong sinc eBlack's Queen does not control the e5-b8 diagonal

After 6. Qf3 Be6 7. c4 Nb4 8. Qb7 Qd4 9 Qa8 Nc2+
10 Ke2 Qe5 11 Kd1 Nxa1 12. Qxb8+ Kd7 13. c5 Black appears to be Ok with an even game.

So, I guess your suggestion is good contribution to the line from Black's point of view.

Thinking about this line yesterday and why it's not played so often these days, I took a quick look at 6 Qf3. After 6 ...Be6 7 c4 Nb4 8 Qb7, one line Burgess (1996) gives, as being good for White, is 8 ...Qd4 9 Qa8 Nc2!? 10 Ke2 Qe5 11 Kd1 Qd4 12 Bd2!. Maybe I'm missing something elementary but what, I wondered, is wrong with the 'obvious' 11 ...Na1 instead? Anyone got any info on this -- or any other thoughts about the Kengis's relative scarcity at the moment?