Ok I’’m going to nit-pick Davies “the Dynamic Reti” but really it’s a great book with a lot of neat ideas.

As a long time flanker I like Davies book, but most of Chapter five I didn’t care for    3. e4   What’s this?  Davies states, “if white wants to keep the game along Reti lines then the little tried 3.e4 seems the best bet.”  HUH?   Looks to me that the positions resulting from 3.e4 has more in common with a  King’s pawn opening than a Reti.  I don’t recommend this at all.  It’s hardly a hypermodern approach and in my opinion runs counter to the flank positions someone who plays the Reti is going to be adept in handling.  3. e4 abandons the flank approach from the get go.   



  Davies obviously uses 1.Nf3 for its transpositional qualities  something to keep in mind. and I have to admit I use the same strategy when playing the Larsen. 


I  don’t agree with his assessment that “ 3.Na3 and Qa4+ does not give white anything and the roads to equality are well mapped out.  The positions resulting from 3 Na3 are pretty complex.  (even at the GM level)    


When playing  the Reti as well as the Larsen, transposing into a Sicilian or a Kings Indian is usually what 50% of my opponents are trying to force me to do.  Then the other half want to transpose into a Queens Gambit of some sort!  So playing 3.e3 (ala the Accepted isn’t going to stun an opponent either ) 

After 1. Nf3 ..1.g3   1. b3  and 1. c4 playing something with a Flank  theme is usually what causes my opponent the most consternation.   Hence,  I believe the best idea in the whole chapter   is the game 31 Alburt-Dlugy  with 1.Nf3 d5 2. c4 dxc 3. g3!?  A totally hypermodern approach…. 
anyway just my IMO
 

Back to the orginal question, it seems best after 3 .. b5 to play 4 b3!? since holding the pawn via 4 .. cb 5 Bxb5+ c6 6 Bc4 ba looks very dangerous for black. 4 .. Nf6 5 e5 Ng4 6 Bb2 etc is fine too.
The potential problem with throwing in 4 a4 c6 is that after both 5 b3 Nf6 6 e5 Ng4 7 Bb2 Qb6!? (the difference) 8 Bd4 c5 9 Bb2 etc black seems to get more play than he perhaps deserves. Certainly it's a lot messier.
(7 bc Nxe5 8 NxN Qd4 9 Nxf7 Qxa1 10 Nxh8 Qxb1 seems fine for black.).

(btw my computer likes 4 d3 which looks anti positional although 4 .. cd (cooperative!) 5 Bxd3 a6 6 o-o etc does give white a large development edge.).

After e5 It's perhaps best to hold off playing 5 Qb3 and just play Nc3.
Then if black moves his bishop on c8 Qb3 *might* work and Nf6 Ng5 might be annoying too. It's a bit reminicent of some kind of mutant open game this.
Still this kind of pawn snatch is dangerous (need to work which ones work and which don't) and the main plan should be to achieve d4 somehow.

I certainly can't see how this can really be better than the 3 e4 QGD -> after 3 .. e5 white's main plan is to play d4 somehow after all.

Hello,

I am not really a supporter of white's cause in this opening, but doubt anything better than tranposing to QGA with 3.e3 objectively speaking. 
            Looking at the old book "secrets I learned from the masters", Edward Lasker suggests that after 3e4!? black equalises most easily with 3...Nc6! 4.B*c4 e5 5. Qb3 Qd7, rather than 3...c5 4.B*c4 Nc6 5b4 when however black takes pawn, white gets advantage, i.e if N*b4, Ne5 followed by Bb5+. (In a later Fine game, black defended succesfully with 5...e6). Not exactly cutting edge theory, but some ideas.

Bye John S

Alumbrado,

not commiting in the center until it's needed, controlling it with pieces, and avoiding some hard worked lines in other openings, is the Reti's point. My, little, experience with this line shows me that sometimes white will be forced to play d3, temporally, to support the e4 pawn. (not d4 in just one move to open the game with great piece activity).  I mean 3.e4! can work but it depends more on black's than white's play.

But..., I'm just a newbie in chess


Thx

But, what's the point to hold the d pawn back? 


Thx

Hello,

Having different opinions is what makes chess fun. For me, since taking up the QGA, this line looks pretty decent for black. For delaying d4, a Keres v Fine game from 193x is a good model. Whilst against 1.Nf3 d5 
2.c4 dc 3Na3, found a game of Edward Lasker with e5, which looks quite fun. (This recently resurfaced in a SOS article). Whilst 3...c5 is the most solid, from Botvinnik v Fine for example. 
          It is my understanding that Davies is recommending 3.e4!? because the other lines are so solid.

Bye John S

Holding back on the d-pawn may be another good option for White but I still don't think it changes the status of 2...dxc4 as a fairly solid option for the second player.  Certainly I don't think it deserves the "?!" label.  But it is good to know White has a variety of means to attempt to cause Black problems.