Stefan Buecker wrote on 12/24/15 at 12:31:11:
I wonder whether I ever looked at 3...f6 again. Thought the analysis was in the same file as 3...Bb4, but there was only 3...Rb8!. [...] Basically my idea was: "static defences" like Qxf4 followed by d6 and f6 in the King's Gambit work, as what can White do? A few years later I understood that the answer to this rhetorical question was: White can do a lot. Even if the first move were OK, after 14...Nef4 followed by h5-h4, White has a ton of possibilities. I still can't find a clear refutation, but the whole concept is clearly very optimistic.
A few years later, after working on the Englund Gambit, I might have recommended 14...exd4! instead. The arrangement of pieces and pawns in the center is really similar to the Englund.
Yes that is the point of Steinitz' artistic, irritating, sometimes ridiculous approaches on his strongpoint e5 concept. He had quite some tries on that very subject, the one or the other even ending in complete reocupation of his back rank with his pieces...
But in fact f7-f6 takes at the earliest possibility the sting out of the spanish concept - the weakness e5 is covered.
This idea reurns in the Modern Steinitz Defence, Alapin Variation: 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 d6 5.Bxc6 bxc6 6.d4 f6 (C73). Is it to bad?!
I'm a bit confused why to first save e5 with f7-f6 and than to surrender that strengthened point with e5xd4 (your suggestion 14...exd4). Well yes, Nimzovich's "Sägestellung" against pawn e4...
Stefan Buecker wrote on 12/24/15 at 12:31:11:
Whether all this is relevant at all for the assessment of 3...f6, I have my doubts... Several moves of Tarrasch were "pragmatic" (a3, Bc4, h3, Kh2), not necessarily the most challenging in a theoretical sense. So if the opponent is not a very strong player, maybe it can work.
Pragmatic, yes. After f7-f6, the Bb5 has played his mayor role, or am i wrong? Yes, due to the pin of the Nc6 he further put's a question to the ideal d7-d6 (but more than once that never the less may work as there are a7-a6 and b7-b5 to break the pin). Returning to c4 is natural, but a tempo down to the Italian (but a weakness up on a2-g8 - but is this too relevant?!). a2-a3 saves the italian like bishop (against Nc6-a5) and h2-h3 is... normal Lopez, isn't it.
The pragmatism bases on the work of the strongpoint e5. I don't see how this can be overruned. What are those challenging moves?!
Browsing through Steintz' games with the strongpoint e5 concept he more than once ended in a sort of king's indian structure.
He had not too much luck with it, which may have its reasons in the exhaustive backward tended play - as for example Tarrasch - Steinitz shows.
At the end he surrendered the exaggerated attempts on it. May be a sign...
But as you Stefan showed with your 14...Nef4 idea there are more active concepts - which in return give white some prospects for play. "White has a ton of possibilities", well that is a bit much of it...
Stefan Buecker wrote on 12/24/15 at 12:31:11:
For a "theoretical" discussion, the Marshall Gambit should be a priority, but final conclusions are not easy to see. Presently I think Black is fine.
No, no that is not you Stefan.That is perhaps Max Euwe of the 1950s. Not only some selected main variations count... Not today.
Shall we really confine ourselfs to search for novelties in move 38 or later??
Remeber after all: Even Magnus Carlsen once played the North Sea Defence (1.e4 g6 2.d4 Nf6 - okay that's another topic but quite offside as well) against Aronian if I am right. He got his share of played, but lost, but not due to the opening.