The theory of the Winawer Poisoned Pawn has built up over decades into such a dense and complex thicket that it requires something out of the ordinary for any variation to be called by the impressive title above. But this is exactly how Emanuel Berg labels the variation starting with 12 h4 (‘!’ according to Berg), instead of 12 Qd3, in his recent book on the 7 Qg4 Winawer. Certainly White has so far recorded excellent results with it.
John Watson’s Play the French, 4th edition (Everyman, 2012) considers the same line, though much more briefly, and he recommends the response 12 … b6, with analysis leading to equality. Berg devotes one full chapter to this line and concludes that it’s insufficient for Black.
The latest issue of The New Winawer Report, now posted on the Winawer page, considers this line, via the very recent game (from last month) J. Geller-Bellahcene, Meurthe-et-Moselle IM 2014. In the diagrammed position Black has just played 16 … Ne3, leaving Berg’s analysis. His GM opponent went wrong immediately and was crushed in a mere five more moves.
A curious aspect of this new cutting edge, the last word in computer-aided theory (at least for the moment), is that it has much in common with Pickett’s line 11 h4 considered in the last issue, introduced in the 1970’s and practically ignored since then. In particular some lines, notably those where Black plays 12 … b6, feature the same Rh3 theme, covering the third rank.