Worth A Punt (part three)?

THE main line of the Englund Gambit remains 1.d4 e5!? 2.dxe5 Nc6 3.Nf3 Qe7.
Black threatens to regain the sacrificed pawn, so White traditionally develops a piece while preventing a capture on e5, viz 4.Bf4.
This seems to have been first played by Alexander Alekhine in 1933. Two previous games in ChessBase's 2020 Mega database feature Swede Erik Lundin playing 4.Qd5, which is less convincing but is still reasonably popular today.
A much-less popular alternative to 4.Bf4 is 4.Bg5, although it tends to transpose to the main line after 4...Qb4+ 5.Nc3 Qxb2 6.Bd2.
But the most-popular move in Mega20 is 4.Nc3!?, which does not attempt to hold on to the e5 pawn. The main idea is that after 4...Nxe5, White can get on with developing his game, starting with 5.e4. This was tried in last year's high-level encounter Alexander Riazantsev (2645) - Saveliy Golubov (2469), 73rd Moscow Blitz, which continued 5...Nf6 6.Be2 d6 7.0-0 h6 8.Nd4 g6 9.f4 Nc6 10.Nd5, when White was better but the game was drawn in 71 moves.
Black usually meets 4.Bf4 with 4...Qb4, which can easily throw White into a panic. The main line continues 5.Bd2 Qxb2, but  Stockfish10 and Komodo10 prefer 5...Qe7!?, which was played by Golubov to draw against a 2699 in the same 73rd Moscow Blitz.
But while 5...Qe7!? might be a good practical choice, most Blacks opt for 5...Qxb2 in the hope White will fall for 6.Bc3?? Bb4!, after which White is lost, eg 7.Bxb4 (not 7.Qd2? Bxc3 8.Qxc3 Qc1#) Nxb4.
White should instead meet 5...Qxb2 with 6.Nc3, when 6...Bb4 7.Rb1 Qa3 8.Rb3 Qa5 9.a3 Bxc3 10.Bxc3 Qc5 11.e3 leaves White with the bishop-pair and a lead in development. But material is level and Black has the better pawn-structure.

How should Black proceed?

*****
*****
*****
*****
*****
The normal continuation is 11...Nge7, but the engines prefer two moves that do not appear in Mega20, ie ...Nh6!? and ...b6!?
Looking at 11...Nh6!?, the engines continue 12.Be2 0-0 13.0-0 b6 14.Ng5 d5!? 15.g4!? (not 15.exd6?? Qxg5), when they much prefer White, but there is a lot of play left.
There is no doubt many Whites will be ready for 4...Qb4, but there is a virtually unknown, but highly promising, alternative, which will be looked at in part four of this series.

Worth A Punt (part two)?

AFTER 1.d4 e5!? 2.dxe5, the main move is 2...Nc6, which White usually meets by developing a knight of his own, while simultaneously holding on to the e5 pawn, viz 3.Nf3.
Black normally replies with the tricky 3...Qe7, but Emil Diemer of Blackmar-Diemer Gambit fame liked to sac his f pawn.
Edgar Walther - Diemer
Swiss Championship (Thun) 1956
3...f6!? 4.e4!?
The analysis engines Stockfish10 and Komodo10 strongly prefer this over the obvious 4.exf6. After the latter, one of Diemer's games continued 4...Nxf6 5.Nc3 d5 6.Bg5 d4 7.Ne4? (White has an edge after 7.Nb5 Bb4+ 8.c3 dxc3 9.Qxd8+ Nxd8 10.Nxc3) Nxe4!, when Black wins a piece. G Brachtl - Diemer, Baden Seniors 1978, saw 8.Bxd8 Bb4+ 9.Nd2 (9.c3 dxc3 does not help White) Bxd2+ 10.Qxd2 Nxd2 11.Bxc7 Ne4 (0-1, 37 moves).
4...fxe5 5.Bc4 Nf6 6.Nc3
6.Ng5!? d5!? 7.exd5 Na5 has a Two Knights' theme about it. White can hold on to his extra pawn with 8.Bb5+ c6 9.dxc6, but 9...Qxd1+ 10.Kxd1 Nxc6 gives Black some compensation, although the engines prefer White.
6...Bc5?!
The engines much prefer the pinning 6...Bb4.
7.0-0 Rf8?!
Almost certainly better is the natural 7...d6.
8.a3?!
White may have a small edge after this, but the move is slow. Better is the engines' 8.Ng5 or 8.Nd5.
8...d6 9.Qd3 Nh5?
Black more-or-less equalises with 9...Bg4.
10.b4 Bb6 11.Nd5 Bg4

How should White proceed?

*****
*****
*****
*****
*****
12.Nxb6
If 12.Bg5, Diemer may well have planned to sac the exchange with 12...Rxf3!? After 13.gxf3 Qxg5 14.fxg4 Qxg4+ 15.Kh1 0-0-0 the engines prefer White but the position is not clear.
Also unclear is 12.Ng5 Nf4 13.Nxf4 exf4 14.Nxh7 Ne5.
12...axb6 13.Ng5 h6 14.Ne6 Bxe6 15.Bxe6 Nxb4?!
Diemer consistently chooses the sharpest option, but the calmer 15...Nf4 16.Bxf4 exf4 may well have served Black better.
16.Qe2 g6!? 17.Bxh6 Rh8 18.Qf3 Qe7 19.axb4 Rb8!? 20.Bf5!? Rxh6 21.Bxg6+ Rxg6 22.Qxh5
The smoke has cleared, and White is a pawn up. But the mass of heavy pieces means it will not be easy to convert his advantage.
22...Qf7 23.Qh3 Ke7 24.Qh4+ Kd7 25.f4 exf4?
This helps White. Much better is mobilising his passive rook by 25...Rbg8.
26.Rxf4 Qg7 27.Qh3+ Re6 28.Raf1 Qd4+ 29.Kh1 Rbe8 30.Rf7 Kc8(??)
Did Diemer really play this? According to the score in ChessBase's 2020 Mega database he did, but much more sensible is breaking the pin on the e6 rook by 30...Kd8.
31.Qa3(?)
Much stronger, if …Kc8 really had been played, is 31.Rf8.
31...R6e7 32.Qa8+ Kd7 33.Qxb7 Qxb4 34.Qd5 Kd8 35.Rxe7 Rxe7 36.c4 Kd7 37.h3 Re5 38.Qd3
Stronger is the engines' 38.Qg8!, eg 38...Rxe4 39.Rf7+ Re7 40.Qg4+ Ke8 41.Rf1 Kd8 42.Rf8+ Re8 43.Rxe8+ Kxe8 44.Qc8+ Kf7 45.Qxc7+, although queen-and-pawn endings are rarely simple.
38...Rc5 39.Rf7+ Kc6 40.e5?!
Stockfish10 likes 40.Kh2, but the rook-and-pawn ending after 40...Qxc4 41.Qxc4 Rxc4 is not clear.
40...Qe1+ 41.Rf1 Qxe5 42.g4
This gets a double-question mark in Mega20, but the move is fine (albeit the position is equal).
42...Kb7 43.h4?!
Still trying to win, but it was safer to take the draw, for example 43.Rf5 Qe1+ 44.Rf1 Qe6 45.Rf4.
43...d5! 44.cxd5??
This is the losing move. White had to play something like 44.Rf5, although Black is better after 44...Qe1+ 45.Kg2 (best) dxc4.
44...Rc3 45.Qe2? 0-1
45.Qe2? loses instantly, but 45.Qd2 Rh3+ 46.Kg1 Rg3+ 47.Kf2 Rxg4 is also hopeless.
This is an imperfect game, to say the least, but it illustrates some of the attacking chances a tactically-minded Black can generate.

Worth A Punt?

IMAGINE  a defence to 1.d4 that receives virtually no serious attention from anyone with the white pieces.
Further, imagine it is a defence played by the likes of Paul Keres, Rudolf Spielmann, Kurt Richter and Emil Diemer.
Add to this the fact that Alexander Alekhine was among those Whites who on occasion failed to beat the defence (he was once significantly worse against an amateur after just seven moves), and you might think it is indeed worth giving it an outing.
Because it is a gambit, it can be especially effective at relatively fast time limits, and as a surprise weapon (which it almost always will be).
The defence is the Englund Gambit, which arises when Black meets 1.d4 with 1...e5!?
Here is the game where a future world champion struggled to show his superior talent.
Alekhine - V Lifschitz
Bern (Switzerland) Simul 1922
1.d4 e5!? 2.dxe5 d5!?
Overwhelmingly most popular is 2...Nc6, which makes 2...d5!? a surprise move in what is already a surprise defence. But probably a better way of implementing this idea is 2...d6!?
3.exd6!?
This is White's usual response, at least in ChessBase's 2020 Mega database, but the analysis engines Stockfish10 and Komodo10 reckon White is much better off with a move such as 3.e4, which is less good after 2...d6!?
3...Bxd6
For his pawn, Black has opened diagonals for both bishops and has a developed piece, while White has merely opened a diagonal for one of his bishops.
4.Nc3 Nf6 5.e4 0-0 6.Bg5?!
This natural-looking move may be a mistake. The engines give 6.Bd3 with advantage to White.
6...Re8 7.Bd3?!
Better is the engines' 7.Qf3.

How should Black proceed?

*****
*****
*****
*****
*****
7...Nc6
Strong is 7...Nxe4!, one point being that 8.Bxd8?? loses to 8...Nxc3+, eg 9.Ne2 Nxd1 10.Bxc7 Nxf2 11.Bxd6 Nxd3+ 12.cxd3 Bg4 etc. Better is 8.Nxe4, but 8...Qxg5 9.Nge2 Nd7 is very good for Black.
8.Nge2 h6 9.Bh4 Bg4 10.h3 Bxe2 11.Nxe2 g5 12.Bg3 Nxe4 13.Bxe4 Rxe4 14.0-0
White gets his king to safety, but meanwhile Black has won back his pawn and has fully equalised (½–½, 27 moves).
(to be continued)

Count The Weaknesses

HERE is an instructive rook-and-pawn ending from Tartakower and du Mont's 500 Master Games Of Chess.

White has just played 49.Kg1-g2 in José Capablanca - Boris Kostić, game one, Havana match 1919  

How many weaknesses does each side have?
*****
*****
*****
*****
*****
*****
*****
*****
[Notes in italics are algebraicised from 500 Master Games Of Chess.]
Now that the battlefield is cleared, manoeuvring will be superseded by active advances.
Black's disadvantage consists in the fact that he has two weaknesses at c5 and g7, whereas White is vulnerable only at b3.
I guess congratulations are in order if you reached the same conclusion as Tartakower/du Mont regarding the number of weaknesses. I am fairly certain my count, if I had not read their notes first, would have been rather different.
For what it is worth, Stockfish10 reckons the position is dead-equal, while Komodo10 gives White an advantage of 0.14 pawns, which is as good as equal.
49...Ra6 50.Kf3 Ra2 51.Bg3 Ke8
A more-stubborn resistance is offered by 51...Ra7, guarding his second rank, and ...Kf7, to be followed by ...g6, relieving the blockade.
The engines are happy with Kostić's choice.
52.Bf4 Ra6 53.Be3 Ra1 54.Kg4 Ra7?!
In chess, as in life, the right perception often comes too late.
The engines strongly dislike the text, preferring 54...Kf7 or 54...Rg1+.
55.Kh5 Kf7 56.Rd5 Ra3
Seeking salvation in counterplay, for if  56...Rc7 [then] 7.h4 (relying on zugzwang) Bf8 58.Rd8 Be7 59.Ra8, and Black must, one way or another, let go.
After the above variation, a sample line given by the engines runs 59...Bd6 60.Ra6 Be7 61.Rb6 Bf8 62.Bf4 Rd7 63.Rc6 Rd3 64.Rc7+ Ke8 65.Kg6 Be7 (if 65...Rxb3?, then 66.Rc8+ Ke7 67.Be3 etc) 66.Be3 Kd8 67.Rb7 Rxb3 68.Kxg7 Ke8 69.Rb5 Rc3 70.Bxc5 Rxc4 71.Bxe7 Kxe7 72.h5, when the game will be decided on the kingside. This is a long line, but the engines have White winning, or at least massively better, throughout.
57.Rd7?!
Clearer, according to the engines, is 57.Bxc5 Bxc5 58.Rxc5 Rxb3 59.Rc7+.
57...Ke8 58.Rd3 Kf7 59.h4 Ra7 60.Rd5 Ra5?
This seems the decisive mistake. The engines give 60...Rc7, protecting c5 and Black's second rank.
61.Rd7 Ke8 62.Rd3
The engines' 62.Rc7!? may be even stronger.
62...Kf7?!
The engines give 62...Ra3 63.Kg6 Kf8 64.Rd5 Rxb3 65.Bxc5 Bxc5 66.Rxc5 Rh3 67.h5 with a large advantage to White.
63.Rd5
Señor Capablanca has composed a Zugzwang Symphony! If 63...Ke8, [then] 64.Kg6, and if 63...Bf8, [then] 64.Rd7+ Be7 65.Rc7, and still the black king has to give up the control of  g6!
63...Ra3 64.Bxc5 Bxc5 65.Rxc5 Rxb3
Numerically even, but the active position of White's king and rook decides the day.
66.Rc7+ Kf8 67.Kg6 Rf3 68.Rf7+ Ke8 69.Rxg7 Rf4
Or 69...Rxf2 70.Rb7 Rf4 71.c5 Rxh4 72.Kxf6 with a huge advantage.
70.h5 Rxc4 71.Kxh6 Kf8
If 71...b3, then 72.Rb7, eg 72...Rc3 73.Kg6 etc.
72.Rb7 Rg4 73.f3
Challenging a decision.
73...Rg5 74.Rxb4 Kf7
Clearly not 74...Rxf5 75.Kg6 Rg5+ 76.Kxf6, winning at once.
75.Rg4
In a subtle manner White frees his king.
75...Rxf5 76.f4 Ra5 77.Rg7+ Kf8 78.Rb7 f5 79.Kg6 Ra6+
If 79...Rc6, then 80.Rb8+ Ke7 81.h6 etc.
80.Kxf5
Many positions with rook, bishop's pawn and rook's pawn versus rook are drawn, but this is not one of them.
80...Ra5+ 81.Kg4 Ra6 82.Kg5 Rc6 83.f5 Kg8 84.f6 Rc8 85.Rg7+ Kf8 86.h6 1-0
A victory of logic and nerves.

Ideas Behind The Jobava-Prié (conclusions)

THE move 3.Bf4, after 1.d4 d5 2.Nc3 Nf6, lags well behind 3.Bg5 in ChessBase's 2020 Mega database - 4,153 examples of the former compared to 18,144 of the latter.
But the figures for last year alone show a complete reversal - 556 for 3.Bf4 and 294 for 3.Bg5.
Clearly the Jobava-Prié is the Queen's Pawn Game of the future (or at least the present).
But as yet the literature is limited, and even more so from Black's view.
Here are what I believe this series has shown the main ideas of the Jobava-Prié to be.
1) White has a crude threat in Nb5, targeting the c7 square regardless of whether there is a Black pawn there. Even very strong players have occasionally forgotten that the move ...c5 does not stop the c7 square being vulnerable.
2) Black usually needs to take prophylaxis against Nb5. A favourite at the highest levels is an early ...a6 to allow a quick ...c5. The main alternatives are ...c6 and the pinning ...Bb4.
3) White's dark-square bishop on f4 can easily be challenged by Black's dark-square bishop from d6. White's three main responses - Bxd6, Bg3 and ignoring the threat (assuming the f4 bishop is not just hanging) - are quite different and have long-term consequences, but may often be a matter of taste. It is best to decide how you will (normally) react before you sit down at the board.
4) Black has a major decision to make on whether to develop his light-square bishop early and actively outside the pawn-chain on f5, or leave it 'trapped' behind the lines by playing ...e6. Expressed like this, it might be thought the former option is obviously better, but things are not so clear. The bishop is certainly more active on f5, but it can be hit by white expansion, eg f3, g4 and h4 etc, or sometimes f3 and e4. It can also be easily challenged by Bd3, when Black has similar choices to White under 3). When the bishop is left passively on c8, it need not be 'trapped' for long - swopping it for White's light-square bishop via ...b6 and ...Ba6 may be possible.
5) Black usually pressurises White's centre with ...c5 (except in lines where ...c6 is played). In general, Black's chances lie on the queenside, as indicated by the fact that Black usually forms the queenside-pointing pawn-chain f7-e6-d5.
6) White usually hopes to play in the centre and on the kingside. Often this comes via a slow build-up after castling short, but sometimes White can castle long, especially if Black has played ...Bf5 and allowed White to push kingside pawns with tempo.
7) White has a very good chance of getting the basic starting position of the Jobava-Prié, either by 1.d4 d5 2.Nc3 Nf6 3.Bf4 or 1.d4 Nf6 2.Nc3 d5 3.Bf4.
After 1.d4 d5 2.Nc3, a French Defence player might try 2...e6, while a Caro-Kann player could choose 2...c6. In each case, White can transpose into these e-pawn openings, or stay in Jobava-Prié territory with 3.Bf4.
After 1.d4 Nf6 2.Nc3, Black can head for a Pirc with 2...d6. Grandmasters have played 3.Bf4!? even here but, theoretically at least, it is better to be ready for the consequences of  3.e4.
Similarly, strong players have been happy to play Bf4 after 1.d4 e6 2.Nc3 f5, and even after 1.d4 f5.
8) The Jobava-Prié is not some sensational wunderkind guaranteed to blow blacks off the board. True, there will be games where an inexperienced Black, at least inexperienced in terms of the Jobava-Prié, will go fundamentally wrong at an early stage. But generally White should get on with development and look forward to a middlegame where White's chances lie in the centre and against Black's king, while Black's chances, although not to be underestimated, lie on the less-critical queenside. It is comforting to know that even if White loses the plot at some stage and gets into trouble on the queenside, there is always the possibility Black will allow a coup against the black king.

Ideas Behind The Jobava-Prié (part 11)

AFTER 1.d4 d5 2.Nc3 Nf6 3.Bf4 c6, White usually gets on with development, starting with the flexible 4.e3.
Much less popular is 4.Nf3, which rules out f3 as a response to ...Bf5.
Black does now normally play 4...Bf5, one point being that 5.f3, intending a later e4, will be less effective when White has already spent a tempo on e3.
Christophe Philippe (IM 2380) - Frederik Fries (2297)
Oberliga South West 2017-18
5.Bd3
First essayed, it seems, by Alekhine in a 1933 simul.
As when Black plays ...Bd6 to confront White's dark-square bishop, so here Black has three main choices, which are, in increasing popularity in ChessBase's 2020 Mega database: 5...e6, 5...Bg6 and the game's …
5...Bxd3
Now White has a choice. 6.Qxd3, which gives White a lead in development, is most popular, but Black is very solid after 6...e6.
6.cxd3!?
This scores much better than 6.Qxd3.
6...e6 7.Nf3 Nbd7 8.0-0 Be7
8...Nh5!? was seen in Yusup Atabayef (IM 2473) - Alexey Dreef (GM 2652), World Rapid Championship (Doha, Qatar) 2016, which continued 9.Bg3 Nxg3 10.hxg3 Bb4 11.Qb3 Bxc3 12.bxc3 Rb8 13.c4 with a slight edge for White (½–½, 37 moves).
Philippe - Fries reached the position after 8...Be7 by the move-order 1.d4 d5 2.Nf3 c6 3.Bf4 Bf5 4.e3 e6 5.Bd3 Bxd3 6.cxd3 Nf6 7.0-0 Be7 8.Nc3 Nbd7.
9.Ne5 0-0

How should White proceed?

*****
*****
*****
*****
*****
This position has been reached 10 times in Mega20, with nine different moves being tried.
White has a central pawn-majority, and may hope to advance there to create prospects of a kingside attack. Alternatively he can think about a Minority Attack on the queenside.
Black is very solid, but must create counterplay - he cannot just sit tight and give White time to optimally organise his forces.
This game features the strongest players to reach this position in Mega20, and Philippe's continuation is particularly interesting.
10.Bg5!?
Perhaps thinking of gaining kingside space with f4, which would also open a way for the white rooks to join a kingside attack.
10...Ne8 11.Bxe7 Qxe7 12.Qb3
The exchange of bishops has lessened the chances of a successful outcome on the kingside, so White switches to the queenside.
12...Nd6 13.a4 a5 14.f4?!
White plays f4 after all, but it is doubtful if his position can stand such expansion. The analysis
engines Stockfish10 and Komodo10 prefer a more prosaic approach, such as 14.Nxd7 Qxd7, followed by manoeuvring, for example 15.Nb1!? Rb8 16.Nd2 with equal chances.
14...f6
The engines reckon White is overstretched after 14...Nxe5!?, eg 15.fxe5 Nf5 or 15.dxe5 Nf5, in both cases with annoying pressure against e3, and the prospect of the knight jumping into d4 if the e3 pawn moves.
15.Nxd7 Qxd7 16.e4
This is the difference - White is able to mobilise his central pawns.
16...dxe4 17.dxe4 f5 18.d5!? exd5 19.exd5
19.e5? is comfortably met by 19...Ne4.
19...c5?
Better is a move such as 19...Kh8, rather than allowing the d pawn to become a base for actions on the e file.
20.Rae1
Black's weak squares on the e file afford White a way into his position.
20...Rae8 21.Re5 Nf7
The knight was on a good blockading square, but it is difficult to come up with a better suggestion for Black.
22.Rfe1!?
Simple and good is 22.Re6, but the text grows on the engines the longer they are given to consider it..
22...Nxe5 23.fxe5 f4
The only good move, according to the engines.
24.e6 Qd8
24...Qd6? 25.Nb5 helps White.
25.Ne4 Qh4
Equal, according to Komdo10, but Stockfish10 reckons White has a won game. Give them long enough, and Stockfish10 sticks to its guns, but Komodo10 comes to favour White.
26.Rf1 Qg4
Komodo10 suggests sacrificing the f pawn by 26...f3!? 27.gxf3 Qh3 with a position hard to assess. The engines come up with various lines - Stockfish10 much preferring White, but Komodo10 giving Black excellent equalising chances.
27.Qf3 Qg6 28.h3 b6 29.Kh1
White calmly consolidates, apparently confident his central passed pawns at least fully compensate for the exchange, so there is no need to rush.
29...h6 30.b3 Kh7 31.Re1 Re7 32.Nd2 Rf5 33.Qe4 Rf6 34.Nc4 Qxe4 35.Rxe4 Kg6 36.Kg1
Black has been unable to organise restraining pressure on the black passers, let alone blockade them, and White has a choice of ways to win, eg 36.d6 Rexe6 37.d7 Rd6 38.Nxd6 Rxd6 39.Re6+! Rxe6 40.d8=Q, but the text is even more sure.
36...Kf5 37.Re5+ Kg6 38.Kf2 Rb7
Black at last succumbs to the urge to protect the b6 pawn, but the game ends immediately.
39.e7 1-0

Ideas Behind The Jobava-Prié (part 10)

IN most lines after 1.d4 d5 2.Nc3 Nf6 3.Bf4, Black puts pressure on the white centre with a well-timed, and sometimes not-so-well-timed, ...c5.
But Black has a major alternative in 3...c6, which covers the b5 square and opens the d8-a5 diagonal for the black queen to later pressurise the white queenside from b6 or a5.
The downside is White has fewer concerns with his centre.
Jens Ove Fries-Nielsen (IM 2365) - Erik Blomqvist (IM 2451)
Swedish Championship 2012-13
4.f3!?
As in the Veresov, trying to build a classical centre with pawns on d4 and e4 becomes a serious option when Black does not play ...c5.
4...Bf5
Stopping 5.e4, but now White gets to expand on the kingside.
Stockfish10 and Komodo10 like the little-played 4...Nh5!? Jonny Hector (GM 2509) - Kassa Korley (FM 2343), Copenhagen 2013, continued 5.Bg5 Qc7!? (the engines much-prefer 5...Qb6) 6.Qd2 h6 7.Be3 Nd7 8.Bf2, when the active 8...e5!? looks promising. The game saw 8...e6 9.e4 Nhf6 10.e5 Ng8 11.f4 with a position that somewhat resembles a French in which Black has played the passive ...c6 (1-0, 25 moves).
5.g4 Bg6 6.h4 h5!?
Top-level players have favoured this over 6...h6.
7.g5 Nfd7 8.e4!?
8.e3 is more popular, but the text is consistent with White's strategy and is preferred by the engines.
8...dxe4 9.fxe4
But here the engines prefer 9.Nxe4!?
9...Qa5 10.e5 e6 11.a3 Qb6 12.Rb1 Be7 13.Bd3!? Qxd4!?

Who stands better?

*****
*****
*****
*****
*****
Both IMs are playing enterprisingly in a difficult position where one slip could easily be fatal. The difficulty of the position can be gauged from the fact that Stockfish10 reckons White is winning, while Komodo10 reckons the position is equal.
14.Nge2 Qb6 15.Bxg6 fxg6 16.Qd3 Nf8
Stockfish10 loves the dynamism of White's set-up. Komodo10, usually the more positional of the two engines in my experience, has been generally happy with Black's play, but now concedes White a slight edge.
17.b4 Nbd7 18.Ne4
The engines prefer 18.Bg3!? or 18.Nd4, with their usual difference of opinion.
18...Rd8!?
The engines prefer 18...0-0-0, or 18...Qb5 to get the queens off.
19.Qb3 c5 20.Rf1
20.0-0!? is possible as 20...c4+ can be met by 21.Qe3.
20...Qc6 21.Qf3 Nh7 22.b5 Qb6 23.Nd6+?!
White is better after 23.Bg3 Rf8 24.Qc3, according to both engines.
23...Bxd6 24.exd6 0-0 25.Qe4 e5!
Now Black gets an initiative.
26.Qd5+
26.Bxe5? Nxe5 27.Qxe5 Rxd6 leaves White's king suddenly looking very vulnerable stuck in the centre.
26...Kh8 27.Be3?
27.Bd2 keeps White in the game.
27...Qa5+ 28.c3
Or 28.Bd2 Rxf1+ 29.Kxf1 Qa4, when White's position is collapsing.
Similarly bad for White is 28.Qd2 Rxf1+ 29.Kxf1 Qa4.
28...Qxa3?!
Again 28...Rxf1+ 29.Kxf1 Qa4 is strong, eg 30.Bf2 Rf8 31.c4 Nb6 32.Qxc5 Nxc4, when Black remains a pawn up and attacking.
29.Rxf8+?!
The engines prefer 29.Qb3, now trying to get the queens off for White.
29...Nhxf8 30.Kf2 Qa4 31.c4 Nb6 32.Qxc5 Nfd7
Black has managed to mobilise his forces, and there is no doubt who has the more-endangered king.
33.Qb4 Qc2 34.Rc1 Qe4 35.Qb3 Qxh4+ 36.Kg2 Qg4+ 37.Ng3 Rf8 38.Bf2 Rxf2+! 0-1
39.Kxf2 Qf4+ picks up the white rook. I imagine the last few moves, at least, were played in a time scramble.

Ideas Behind The Jobava-Prié (part nine)

AFTER 1.d4 d5 2.Nc3 Nf6 3.Bf4 e6 4.e3, Black can pin the c3 knight with 4...Bb4.
Such a pin is common in the Veresov, but here it can be even more effective as there is no white bishop pinning the f6 knight, which might be able to drop into e4.
One point of the line is that Black can play a quick ...c5 without worrying about a white knight landing on b5 and attacking the c7 square.
Baadur Jobava (GM 2708) - Oleg Korneev (GM 2501)
Tseshkovsy Memorial (Sochi, Russia) 2014
5.Bd3 c5 6.dxc5
White should be careful. 6.a3? lost a pawn in Chongsheng Zeng (GM 2536) - Jianchao Zhou (GM 2613),  Chinese Championship (Xinghua) 2016, after 6...Bxc3+ 7.bxc3 Qa5, as 8.Qd2? runs into 8...c4 and 9...Ne4.
6...Nbd7
Black wishes to capture on c5 with this knight and so prolong the pin on the white c3 knight.
7.Ne2
This scores much better in ChessBase's 2020 Mega database than the equally popular 7.Nf3.
7...0-0!? 8.Bd6!?
White makes it as difficult as possible for Black to regain the pawn.
8...Re8 9.Bb5 a6
Stockfish10 and Komodo10 suggest sacrificing the exchange with 9...b6!? 10.Bc6 Ba6 11.Bxa8 Qxa8. After the further moves 12.0-0 bxc5 13.Bg3 e5, Black has more space and the bishop-pair, but I suspect most humans would rather have the white pieces.
10.Bxd7 Nxd7 11.Qd4 Qa5 12.0-0-0!?
This gives the impression of castling into it, but may be OK.
Lars Oskar Hauge (IM 2468) - Dmitrij Kollars (GM 2564), Pro League (rapid) 2019, saw 12.0-0 Bxc5 13.Bxc5 Qxc5 (½–½, 79 moves).
12...Bxc5!?
Black is at least equal after 12...b6, eg 13.Bc7 Bxc5 14.Qa4 Qxa4 15.Nxa4 Ra7 16.Nxc5 Rxc7.
The consistent 12...Nxc5 also looks fine as 13.a3? is met by 13...Nb3+ 14.cxb3 Bxd6.
13.Bxc5 Qxc5

How should White proceed?

*****
*****
*****
*****
*****
14.e4!
White gets an initiative.
14...Qxd4 15.Nxd4 dxe4 16.Nxe4
The position is simplified, but White has a handy lead in development.
16...Nf6 17.Nxf6+ gxf6 18.Ne2
Taking control of the d file.
18...e5 19.Nc3 Be6 20.Nd5 Bxd5 21.Rxd5 Kg7 22.Rhd1 Re6 23.Rd7 b5 24.R1d6 Rae8 25.Kd2 f5 26.b3 f4!?
This looks strange. Normal, but not ecessarily better, is 26...Kf6.
27.g3!?
Allowing Black to undouble his pawns also seems odd, but the idea may be to make it more difficult for Black to create a passed e pawn..
White could have created a passed pawn of his own with 27.c4.
27...fxg3 28.hxg3 b4 29.c3 bxc3+ 30.Kxc3
White remains better. He has only two pawn-islands and will find it much easier to create a passer.
30...h5 31.b4 e4 32.a4 e3 33.fxe3 Rxe3+ 34.Rd3 Rxd3+?!
The engines strongly prefer 34...R3e4, eg 35.Kb3 Rb8 36.R7d4 Re1 37.Kb2 f5, when it is hard to see how White makes progress.
35.Rxd3 Re6 36.b5 axb5 37.axb5 Re4?
The engines' 37...Kg6 is better.
38.b6
The pawn cannot be stopped.
38...Re6 39.b7 Rb6 40.Rd7 Rb1
Or 40...Kf6 41.Kc4 Ke6 42.Kc5 Rb1 43.Rd6+ Ke7 44.Rb6 Rc1+ 45.Kd5 Rd1+ 46.Ke4 etc.
41.Kc4 Kg6 42.Kc5 Kg5 43.Rxf7 Kg4 44.Rf4+ Kg3 45.Rb4 1-0

Ideas Behind The Jobava-Prié (part eight)

AFTER 1.d4 d5 2.Nc3 Nf6 3.Bf4 e6, White usually continues 4.e3.
The black reply 4...c5?! scores well, but it is doubtful if White plays the thematic 5.Nb5.
Then the forced 5...Na6 covers the c7 square, but leaves Black in a quandary over how to get rid of the pesky knight on b5.
Much more common is for Black to develop his dark-square bishop, with 4...Bd6 being narrowly more popular than 4...Bb4 in ChessBase's 2020 Mega database.
The latter scores a better percentage, and I will look at it in the next part of this series, but here I will cover ...Bd6..
Emilio Córdova (GM 2602) - Anton Kovalyov (GM 2602)
American Continental (San Salvador) 2016
5.Nf3
The main line. White saves a tempo on moving the light-square bishop and is unconcerned about contracting doubled f pawns - the continuation will show why.
5...Bxf4 6.exf4 0-0
6...Qd6 can be met by 7.Ne5, which is where the white king's knight likely wants to go at some point anyway.
7.Bd3 b6!?
Slightly more popular is 7...c5, but playing to swop off Black's bad bishop seems logical.

How should White proceed?

*****
*****
*****
*****
*****
8.Ne2!?
Komodo10's choice. I presume the idea is that White hopes to attack on the kingside, and may castle long if the right conditions arise.
8.Qe2 prevents an immediate ...Ba6, but Komodo10 and Stockfish10 reckon Black is fine after a line such as 8...c5 9.b3 Qc7 10.Qe5 Qe7. It may well be that 8...a5!? is also reasonable, although the engines slightly prefer White after 9.0-0 Ba6 10.Bxa6 Nxa6 11.Ne5.
8...Ba6 9.Ne5 c5
White's light-square bishop cannot avoid a swop.
10.c3 Bxd3 11.Qxd3 Qd6 12.0-0 Nc6 13.Rfe1
White's game is a little freer, but Black can engineer queenside counterplay before the situation gets serious on the kingside, where the f4 pawn gives White more space.
13...Rac8 14.Rad1 Rfe8 15.h3 cxd4 16.Nxd4!?
This comes to be the engines' top choice, but not by much.
16...Qxd4 17.Qxd4
White avoids 17.cxd4?!, which gives him an isolated d pawn and opens the c file, where Black already has a rook.
17...Qc5 18.g4 Re7 19.a4 ½–½
This game suggests that if Black knows what he is doing, White cannot expect much from allowing a swop on f4. Perhaps 5.Bg3 is the way to go as 5...Bxg3!? 6.hxg3 gives White a half-open h file and promotes his h pawn, that covers one square (g3), to a g pawn that covers two squares (h4 and f4).

Ideas Behind The Jobava-Prié (part seven)

BLACK'S most-popular response after 1.d4 d5 2.Nc3 Nf6 3.Bf4 is to open a diagonal for his dark-square bishop with 3...e6.
White usually continues 4.e3, but his most-successful move percentage-wise in ChessBase's 2020 Mega database is 4.Nb5, trying to take immediate advantage of Black not covering the b5 square.
Baadur Jobava (GM 2690) - Viswanathan Anand (GM 2782)
World Blitz Championship (Riyadh, Saudi Arabia) 2017
4...Na6
Overwhelmingly most popular, but Komodo10 very marginally prefers 4...Bd6!? 5.Nxd6 cxd6, when Black argues his central pawn-majority and half-open c file balance White's bishop-pair. Semetey Tologontegin (IM 2393) - Julio Catalino Sadorra (GM 2574), Asian Indoor Games 2017, continued 5.e3 Nc6 6.Nf3 0-0 7.Be2 Qe7 8.0-0 e5 9.Bg5 Bf5 10.Bg5!?, when Komodo10 and Stockfish10 suggest 10...Rd8 (Sadorra played 10...Bf5) with a reasonable game for Black.
Also possible is 4...Bb4 5.c3 Ba5, when Baadur Jobava (GM 2669) - Boris Gelfand (GM 2734), Eurasian Blitz Cup 2016, saw 6.a4 a6 7.b4 axb5 8.axb5 b6?! (8...Bd7!? worked out well in Michael Duggan (186 ECF) - Spanton (168 ECF), Bournemouth 2019, which continued 9.bxa5?! - 9.e3 seems better - Bxb5 10.Qb3 c6, when White's a pawn proved a major weakness and ultimately cost him the game) 9.bxa5 bxa5 10.e3, when Black's a pawn proved a major weakness (1-0, 48 moves).
5.e3 c6 6.Nc3 Nc7 7.Nf3 Bd6 8.Ne5 0-0 9.Qf3!?
Belarusian GM Vladislav Kovalev preferred the normal 9.Bd3 in wins against strong opposition  in 2015 and 2016.

How should Black proceed?

*****
*****
*****
*****
9...Bb4!?
Previous high-level games had seen 9...c5 and 9...Nb5?! Anand's novelty seeks to take advantage of White's somewhat denuded queenside.
10.Bd3
Trying to save the b pawn by attacking the black dark-square bishop only makes matters worse, eg 10.a3? Bxc3+ 11.bxc3 Ne4, or 10.Nd3? Bxc3+ 11.bxc3 Nb5.
Possible is 10.Bg5, but the engines prefer the text.
10... Nb5
This move works now that the c3 knight is pinned.
11.0-0!?
White can save the pawn with 11.Bxb5 cxb5 12.Bg5, with an unclear position after 12...Be7.
11...Nxc3 12.bxc3 Bxc3 13.Rab1
White's lead in development gives him full compensation for the pawn, according to the engines. That makes it a difficult position for Black to play at a blitz time-control, but Anand is a speed specialist.
13...Nd7 14.Qg3
There is no mate (or perpetual) after 14.Bxh7+?? Kxh7 15.Qh5+ Kg8, eg 16.Rb3 Nxe5 17.Bxe5 f6 18.Rxc3 fxe5 19.e4 exd4 20.Rh3 dxe4 etc.
14...Nxe5 15.Bxe5 f6 16.Bd6 Rf7 17.Rb3 Ba5 18.e4?!
The engines reckon White still has decent compensation for his material deficit after 18.c4 Bc7 19.Bxc7 Qxc7 20.Qf3, although they are starting to favour Black.
18...Bc7 19.Bxc7 Qxc7 20.e5 c5 21.Bxh7+??
Seeing his compensation ebbing away, Jobava gets desperate (or simply miscalculates). 21.Qh4 keeps the game going.
21...Kxh7 22.Qh4+ Kg8 23.Rh3 Kf8 24.Qh8+ Ke7 (0-1, 33 moves).

Ideas Behind The Jobava-Prié (part six)

ONE of Black's most-popular responses to 1.d4 d5 2.Nc3 Nf6 3.Bf4 is to, somewhat provocatively, ignore what White is up to.
Instead Black gets on with his own development with 3...Bf5.
As in the Veresov, ...Bf5 cries out for the response 4.f3, and that has been the choice of Baadur Jobava, Hikaru Nakamura, Arkadij Naiditsch and many other grandmasters.
This move scores a much-higher percentage in ChessBase's 2020 Mega database than the marginally more-popular 4.e3.
The main line after 4.f3 continues 4...e6 5.g4 Bg6 6.h4, when opinion sharply divides.
Baadur Jobava (GM 2716) - Robert Hovhannisyan (GM 2611)
EU Championship (Yerevan, Armenia) 2014
6...h6
Almost as popular is 6...h5 - grandmasters have played both moves. For what it is worth, the analysis engines Stockfish10 and Komodo10 prefer the text.
7.e3 a6
Sooner or later, in many lines of the Jobava-Prié, Black feels the need to spend a tempo on ...a6.
8.h5 Bh7 9.Bd3 c5!?
The capture 9...Bxd3 is more-or-less automatic in a somewhat similar position in the Caro-Kann Classical, and was played by Ian Nepomniachtchi in a 2014 World Rapid Championship loss to Baadur Jobava. The text is preferred by Stockfish10, but Komodo10 rates the two moves equally.
10.Nge2

How should Black proceed?

*****
*****
*****
*****
*****
10...Bxd3
The engines prefer 10...Nc6. Spanton (1908) - Koby Kalavannan (2100), Southend 2016, continued 11.Qd2 b5 12.a3 Qa5 13.Kf2!? Qb6 14.Bxh7 Nxh7 15.Kg2 cxd4 16.exd4 Be7, when the engines slightly prefer Black (½–½, 38 moves).
11.cxd3!?
We have seen this idea before (part three of the series).
11...cxd4 12.Nxd4 Nbd7
12...Nc6!? 13.Nxc6 bxc6 14.d4 is probably nothing for Black to worry about.
13.Nde2 e5 14.Bg3 Bd6 15.d4 0-0 16.Qb3!?
Jobava ups the stakes. More solid was 16.0-0.
16...Qe7 17.Bh4 exd4?
Better is 17...Bb4. The text loses a pawn
18.Nxd5 Qe5 19.Nxf6+ Nxf6 20.Nxd4
White is a pawn up. Black tries to win it back with …
20...Nd5 21.Nf5 Nxe3
… but 22.Nxe3 Rfe8 23.0-0-0 is winning for White.
Instead the game saw ...
22.Qxe3?? Qxb2
… when Black is slightly better, although the game was drawn.

Ideas Behind The Jobava-Prié (part five)

WHITE has a radically sharper, but little-known way, of meeting 3...c5 (after 1.d4 d5 2.Nc3 Nf6 3.Bf4) that is reminiscent of the Albin Countergambit, but with colours reversed.
There are just 11 examples of 4.e4!? in ChessBase's 2020 Mega database, compared with 368 for 4.e3, but five of the 11 outings are by players rated well over 2500, and the move could easily pack a surprise punch at club level.
Kirill Alekseenko (GM 2563) - Sergei Rublevsky (GM 2683)
Russian Rapid Championship (Sochi) 2017

Position after 4.e4!?

*****
*****
*****
*****
*****
4...Nxe4
This variation is in its infancy and there is no consensus as to how Black should react.
Nils Grandelius (GM 2643) - Pia Cramling (GM 2453), EU Blitz Championship (Tallinn, Estonia) 2016, saw 4...dxe4 5.dxc5 Qa5 6.Bb5+ Bd7 7.Qe2 a6 8.Bc4 Nc6 9.0-0-0 Qxc5 10.Nxe4 Nxe4 11.Qxe4 with an unclear position (1-0, 41 moves).
Black went wrong immediately in Boris Grachev (GM 2626) - Dmity Kryakvin (GM 2595), Russian Blitz Championship (Sochi) 2018, with 4...Nc6?, when the thematic 5.Nb5 would have been embarrassing. White whiffed with 5.Bb5+?, and the game was eventually drawn.
The only other known move, apart from the text, is 4...cxd4, which was played by an 1871 in 2019. After 5.Qxd4 Nc6 6.Bb5 Bd7 7.Bxc6 Bxc6 8.e5 Ne4!? 9.e6!? the position was unclear.
Clearly there is a lot to be discovered in these lines.
5.Nxe4 dxe4 6.dxc5 Qa5+ 7.c3 Qxc5 8.Qa4+ Nc6 9.Qxe4 g6
Olexandr Bortnyk (GM 2579) - Alexander Grischuk (GM 2747), ICC Blitz Final 2016, continued 9...Bf5 10.Qe3?! (Stockfish10 and Komodo10 prefer 10.Qa4, so 10...e5 can be met by 11.Be3) e5 11.Qxc5 Bxc5 12.Bg3 0-0-0, when Black had a large lead in development (0-1, 44 moves).
10.Qc4 Qa5 11.Qb5 Bg7 12.Qxa5 Nxa5 13.Bb5+ Nc6 14.0-0-0 Be6 15.Kb1 a6 16.Bxc6?!
Giving up the bishop-pair on a fairly open board with rival pawn-majorities and the possibility of opposite-side castling must be questionable. The engines reckon White should retreat the bishop to a4, d3 or even e2.
16...bxc6 17.Nf3 0-0 18.Be5 Bh6 19.Nd4 Bd5 20.f3 f6 21.Bg3 e5 22.Nc2 a5 23.Rhe1 Rfd8 24.b3 a4 25.Kb2 c5?!
Black is much better after 25...Be6, or 25...axb3 26.axb3 Be6, according to the engines.
26.c4 axb3 27.axb3 Be6 28.Kc3?
Black's advantage is smaller after 28.Rxd8+ Rxd8 29.Re2, according to the engines.
28...Bf5 29.Bh4
Or 29.Rxd8+ Rxd8 30.Re2 Bc1 (threatens 31...Rd3#) 31.Ne1 Ba3 32.Bf2 g5 (creating a retreat square for the f5 bishop) 33.g4 Bg6 34.h4 gxh4 35.Bxh4 Rd1 36.Re3 (White is almost in zugzwang) Bb4+ 37.Kb2 Kf7, and Black improves his position at his leisure (the engines have Black the equivalent of well over a rook ahead).
Perhaps best is the engines' 29.b4 Rxd1 30.Rxd1, but Black invades and wins a pawn with 30...Ra2 31.Ne1 Ra3+ 32.Kb2 cxb4 33.Nc2 Rc3 34.Nxb4 Rxc4.
29...Rxd1 30.Rxd1 Ra2 31.Ne1 Be3 32.b4
Best, according to the engines, but Black can now wrap matters up with 32...Bd4+. Rublevsky played 32...g5 33.Kb3 Re2 34.Bg3 cxb4 35.Kxb4 Bd4 and won anyway in 56 moves - the bishops were just too strong.

Ideas Behind The Jobava-Prié (part four)

BLACK can immediately attack White's centre after 1.d4 d5 2.Nc3 Nf6 3.Bf4 by playing 3...c5.
This is possible because 4.Nb5? is simply met by 4...Qa5+, forcing 5.Nc3, after which 5...cxd4 gives Black a large advantage.
Instead White normally plays 4.e3 (4.dxc5?! d4), supporting the d4 pawn and opening a diagonal for White's light-square bishop.
Semetey Tologontegin (IM 2369) - Oleg Korneev (GM 2511)
Cheliabinsk Panchenko Memorial 2019
4...cxd4 5.exd4 a6
5...e6?! is a natural-looking move that has ensnared some very strong players, for example Baadur Jobava (GM 2669) - Igor Kovalenko (GM 2668), Eurasian Cup Blitz (Almaty, Kazakhstan) 2016, continued 6.Nb5 Na6 7.c3!? Be7 8.Nf3 0-0 9.Bd3 Bd7 10.a4, when the Latvian GM lost patience with the annoying knight: 10...Bxb5?! (Stockfish10 and Komodo10's 10...Nh5 11.Be3 Nb8, planning ...a6, seems a better try) 11.axb5 Nc7 12.0-0 Bd6 13.Ne5. White's bishop-pair and queenside pressure gave him a large advantage (1-0, 61 moves).
However, the fact that Black plays ...a6 anyway is one reason why 3...a6 is popular at the highest levels (see parts two and three).
6.Bd3!?
Overwhelmingly more popular is 6.Nf3, but the text has been Jobava's choice and has also been played by Magnus Carlsen.
6...Nc6 7.Nge2
Carlsen played the eccentric-looking 7.Nce2 in a 2016 rapid game.
The text avoids a pin from Black's light-square bishop, which suddenly looks short of decent squares.
7...Bg4!?
Ruslan Ponomariov shut his light-square bishop in with 7...e6 against Jobava in a 2016 Olympiad loss.
The problem with the text is that the bishop becomes a target for white kingside expansion.
8.f3 Bh5 9.h4 Bg6 10.g4 e6 11.h5 Bxd3 12.Qxd3
Taking with the c pawn, as seen in Carlsen - Nakamura in part three of this series, is not appropriate here as White would not have a mobile centre, and the isolated d pawns would be weaknesses.
12...h6 13.0-0-0
Better than the Kf2?! seen in a similar position in part two.
One of the attractions of the Jobava-Prié for an attacking player is that White often gets to castle long in relative safety, while Black is unsure whether to castle short, where he faces an imminent pawn-storm, or to keep his king in the centre, when his pieces may lack coordination.
13...Bd6 14.Qe3 Na5 15.b3

How should Black proceed?

*****
*****
*****
*****
*****
15...Rc8!?
Korneev is the first to vary from Jobava (GM 2646) - Liviu-Dieter Nisipeanu (GM 2657), Spilimbergo (Italy) 2018, which saw 15...Bxf4 16.Nxf4 Rc8 17.Nce2 Qd6 18.Nd3 0-0 19.Kb1, when the engines prefer White but in the game Black's queenside attack triumphed.
16.Kb1 Nc6 17.Bxd6 Qxd6 18.Nf4 b5 19.Nd3 Nd7 20.Ne2 Ne7 21.c3 a5 22.Rc1 a4!?
Komodo10's choice, but it allows White to lock the queenside.
23.b4 Nb6 24.Nf4 Nc4 25.Qe2 Na3+ 26.Kb2 Nc4+ 27.Ka1
The IM declines to see if his GM opponent is happy with a draw - Tologontegin believes he is better, or at least that he has the safer king.
27...Nc6 28.Rhg1 Qe7 29.Rce1 Rd8 30.Qf2 0-0
Black at last castles, and so gets his king's rook into play. But unless he can engineer a sacrificial breakthrough on the queenside, which seems unlikely, the best he can apparently hope for, barring a blunder from White, is a draw.
31.h3 e5!?
Seeking central counterplay before White organises a kingside attack.
The engines suggest 31...Qf6, but after 32.f4 the storm clouds are gathering.
32.g5 hxg5 33.Rxg5 e4?
Komodo10 reckons 33...Qe6 holds, but Stockfish10 continues 34.dxe5!, when the engines agree White is much better, eg 34...Qxh3 35.Nf4 Qc8 36.Qg3 Kh8 37.Ng6+! Kg8 (forced), and now the simple 38.Nxf8 is good enough for a large advantage, although Stockfish10 reckons 38.Nh4!? is even stronger.
Perhaps best is Stockfish10's suggestion of 33...Kh8, although White is clearly better after, for example, 34.Qh4.
34.fxe4 dxe4 35.Qg2?
Very strong is the engines' 35.h6 g6 36.Rxe4! as Black gets mated after 36...Qxe4? 37.Qf6 Qh1+ 38.Rg1 etc. Also winning is 35.Reg1.
35...exd3?
The engines' 35...Qf6 seems to hold as 36.Rf1? runs into 36...Qh6, when Black threatens a fork on e3 (as well as capturing the d3 knight), and 37.Qxe4 is met by another fork, 37...Nd2.
36.Rxe7 Nxe7 37.Rxg7+ Kh8 38.Qg5 Nf5?
This lets White mate, but the engines' 38...Rd6 offers only limited hope.
39.Qxf5! Kxg7 40.h6+ Kxh6 41.Qf6+ Kh7 42.Ng5+ Kg8 43.Qh6 1-0
Quite a demolition job by the IM, and a fine advert for the Jobava-Prié, with the only obvious blemish being the error at move 35.

Ideas Behind The Jobava-Prié (part three)

AFTER 1.d4 d5 2.Nc3 Nf6 3.Bf4 a6!? 4.e3 Bf5, White's main alternative to 5.f3 is 5.Bd3, which has been played by Magnus Carlsen and Baadur Jobava, the Georgian grandmaster after whom the opening is partly named.
The idea is firstly to swop off Black's well-placed light-square bishop, but at the same time, at least as played by Carlsen and Jobava, to strengthen White's centre.
Magnus Carlsen (GM 2832) - Hikaru Nakamura (GM 2785)
Norway Masters Blitz (Stavanger) 2017
5...Bxd3
5...Bg6 and 5...e6!? are playable alternatives.
6.cxd3
Quick development with 6.Qxd3 has been played by a 2557, but presumably Carlsen and Jobava feel it is a position in which strengthening the centre is more important.
6...e6 7.Nf3 Bd6 8.Bg5!?
Stockfish10's choice.
Jobava last year chose 8.0-0 against a player rated more than 400 points below him, but could only draw.
8...Nbd7 9.0-0
Stockfish10 much prefers 9.e4!?, but Komodo10 is unsure.
9...h6 10.Bh4 0-0 11.e4
White starts to mobilise his central pawn-majority, which seems his main trump in the position.
11.Be7 12.Qb3 Nb6 13.Rac1 c6 14.Rfe1 Rc8 15.h3 Nfd7
Seeking an exchange of bishops to relieve the cramp in Black's position.
16.Bg3 Re8 17.a3 Bf8 18.Qa2 Nf6 19.b4
Starting a Minority Attack with the aim of giving Black a queenside pawn weakness.
19...Ra8
Black cannot really prevent a4 followed by b5, but if they come he will meet the latter with ...axb5, opening the a file for his repositioned queen's rook.
20.Qb2 Nh5 21.Bh2 Nf6 22.Red1 Qe7
The engines reckon 22...a5 gives Black a good game.
23.Nd2 Rec8 24.Nb3 Nfd7 25.Re1 Qd8 26.Nc5 Ra7 27.a4
After some cat-and-mouse manoeuvring, Carlsen proceeds with his Minority Attack.
27...Nxc5 28.dxc5!?
The engines prefer leaving Black with a backward b pawn on a half-open file after 28.bxc5.
28...Nd7 29.exd5 exd5 30.d4 b6!?
Nakamura immediately seeks to undermine White's queenside space advantage.
31.b5!?
The engines prefer a less-sharp move, such as 31.a5.

How should Black proceed?

*****
*****
*****
*****
*****
31...axb5
Black is slightly better after 31...cxb5!?, according to the engines, eg 32.axb5 (32.c6!?) bxc5 33.bxa6 Nf6 34.Qb5 cxd4 35.Ne2 Qe8 36.Nxd4 Qxb5 37.Nxb5 Rxc1 38.Rxc1 Rxa6, when Black is a pawn up but his pieces are a little uncoordinated.
32.axb5 bxc5 33.bxc6 Rxc6 34.Nxd5 (½–½, 67 moves)

Ideas Behind The Jobava-Prié (part two)

HISTORICALLY, Black's first response to 1.d4 d5 2.Nc3 Nf6 3.Bf4 was 3...a6!?
The move occurred in James Mason - Max Weiss, Vienna 1882, and might cause a few sniggers.
But, as I mentioned in part one of this series, it has since been played by Carlsen, Caruana, Svidler and Nakamura, so should be taken seriously.
Black's obvious point is to prevent a white piece landing on b5. Black argues that a tempo spent on a passive move in this position is not important as White is not in a position to open/put-pressure-on the black centre with c4 or e4.
White usually replies 4.e3 rather than 4.Nf3. Both moves are useful and absolutely normal, but the former gives White the option of meeting 4...Bf5 with 5.f3!?.

Position after 5.f3

This plan occurred in the following game from the 2018 Korchnoi Memorial in St Petersburg.
Denis Pershin (2403) - Hannes Stefánsson (GM 2511)
5...h6!?
More popular is 5...e6, but the text is the choice of the analysis engine Komodo10, and the game soon transposes to better-known lines.
6.g4 Bh7 7.h4 e6 8.Bd3 Bxd3 9.Qxd3
The play is reminiscent of the main line in the Classical Caro-Kann, but with a less-open centre.
9...c5 10.Nge2 Nc6 11.Kf2?!
Most players have preferred 11.0-0-0.
11...Bd6 12.Na4?!
This decentralisation does not help White's cause.
Stockfish10 gives 12.Kg2 Rc8 13.Bxd6 Qxd6 14. Ng3, but prefers Black.
Komodo10 gives 12.dxc5 Bxc5 13.Nd4 0-0 14.Nxc6 bxc6 15.Bg3, again with an edge to Black.
12...cxd4 13.exd4 Bxf4 14.Nxf4 Qd6 15.Ne2 e5!?
Provoking an immediate crisis. The calmer 15...0-0 is not necessarily better after White replies 16.g5.
16.dxe5 Nxe5 17.Qd4
The engines reckon White had to find 17.Qa3!?, although Black must be better after 17...Qxa3 18.bxa3 Rc8.
17...Rc8
This simple move proves very difficult to meet.
18.Nec3
The engines reckon best is 18.g5 Nfg4+! 19.fxg4 Rc4 20.Qb6 Nxg4+ 21.Ke1 Qe5 22.Nac3 0-0 23.Rd1 Ne3, but with a winning attack for Black.
18...Nfxg4+! 19.Kg2
Or 19.fxg4 Qf6+ 20.Kg2 Qf3+ 21.Kg1 Qg3+ 22.Kf1 Rc4 23.Qg1 Qf3+ 24.Ke1 0-0, when White will not survive long.
19...Rc4 20.Qb6 Rc6 21.Qd4 0-0 22.Rae1
Or 22.fxg4 Rc4 23.Qf2 Rxg4+ with a winning attack.
22...Rc4 23.Qb6 Qd7 24.b3 Rf4 25.Rh3 d4 26.Nd1 Rc8 27.Re2 Nxf3!? 28.Rxf3 Qd5 0-1
In the final position, White is getting mated. Note how the black knight at g4 was able to stay at its precarious-looking post for more than a third of the game.
White has a major alternative, after 4...Bf5, in 5.Bd3, seeking to swop off the well-placed black bishop. I will look at this plan in part three.

Ideas Behind The Jobava-Prié

THE most popular move at the highest levels after 1.d4 d5 or 1.d4 Nf6 is 2.c4.
This is far from being the case in club chess.
In my last 100 games with the black pieces that opened 1.d4 d5, a list that stretches back to late 2017, 2.c4 was played just 41% of the time.
It used to be that the most serious alternative to 2.c4 was a Colle with 2.Nf3, 3.e3 and 4.Bd3.
The Trompowsky (1.d4 Nf6 2.Bg5) and Pseudo-Trompowsky (1.d4 d5 2.Bg5!?) became popular in the 1990s, until it was largely replaced by the London (1.d4, 2.Nf3 and 3.Bf4) and more recently the New London (1.d4 and 2.Bf4).
At the same time the Blackmar-Diemer Gambit (1.d4 d5 2.e4!?) had its dedicated followers, and the Veresov , aka the Richter-Veresov, (1.d4, 2.Nc3 and 3.Bg5) enjoyed a burst of popularity in the early years of this century.
Today there is a (relatively) new kid on the block: 1.d4, 2.Nc3 and 3.Bf4.
I have been calling it the Barry Attack, and in some quarters it is known as the New Veresov, but I think it is fair to say that the name Jobava-Prié  is becoming generally accepted.
This honours Georgian grandmaster Baadur Jobava and his French counterpart Éric Prié.

The main starting position of the Jobava-Prié

Note that White can be sure of getting this position in the vast majority of games, whether Black meets 1.d4 with 1...d5 or 1...Nf6, as in each case White's follow-up of 2.Nc3 threatens to establish a classical centre with 3.e4.
The oldest game in ChessBase's 2020 Mega database featuring the basic Jobava-Prié tabiya after 3.Bf4, is James Mason - Max Weiss, Vienna 1882.
The Austrian Weiss, presumably caught by surprise, replied 3...a6, which looks like a duffers' move, except it has subsequently been played by those well-known non-duffers Carlsen, Caruana, Svidler and Nakamura.
Mason - Weiss continued 4.e3 e6 (Svidler and Nakamura preferred 4...Bf5) 5.Bd3!? c5 6.dxc5 Bxc5 7.Nge2?! Nc6 with a position that was at least equal for Black (but 1-0, 35 moves).
More than 20 years pass before a second 'Jobava-Prié' appears in Mega20, and a further 25 years before the third game.
Alexander Alekhine seems to have been the first superstar to use the opening, but today's adherents include Carlsen, Nakamura, So and Rapport.
In some ways the Jobava-Prié can be thought of as a queenside Italian Game (1.e4 e5 2.Nf3 Nc6 3.Bc4).
One of the main ideas of the Italian Game is to attack the square f7 if Black gets careless. Similarly, one of the main ideas of the Jobava-Prié is to attack c7 if given the chance.
Naturally there are major differences. A white capture on f7 usually involves either a check or a fork of rook and queen, whereas a capture on c7 will often involve a check and a fork, but this time of rook and king.
A white attack in the Italian Game can often be countered by a well-timed ...d5, with the d pawn being supported by the queen and king's knight.
The counter ...e5 in the Jobava-Prié is less likely to be successful, not least because it will probably be supported by a lone knight at best.
In this series I plan to look at ideas in the Jobava-Prié for White and Black.

Bad Wörishofen Prizes

MY score of 6.5 in the nine-round Bad Wörishofen U2000 saw me finish equal-second with  four others (fifth on tiebreak), winning 130 euros.
Vic Rumsey won our 40-euro prize for the shortest win, 50 euros for topping his rating group in the seniors, and a surveillance camera in the closing-ceremony tombola.
Ray Kearsley and Paul Stokes won nothing.

Bad Wörishofen Concluded

MY round-nine game from Bad Wörishofen U2000.
Maik Kopischke (1799) - Spanton (1837)
Réti Opening
1.Nf3 d5 2.g3 Nc6 3.c4
This has been played by grandmasters, but much more popular are 3.d4 and 3.Bg2.
3...dxc4 4.Na3?!
A better way to recapture the pawn seems to be via 4.Qa4. Also popular is the gambit move 4.Bg2.
4...e5 5.Nxc4 e4 6.Nh4N Bc5 7.d3
This is the choice of Stockfish10 and Komodo10, but it highlights how dubious White's opening has been.
7...Nf6 8.Be3 Bxe3?!
The engines much prefer 8...Bb4+ 9.Bd2 Qe7.
9.Nxe3 exd3 10.exd3 0-0 11.Bg2 Nd4
This is almost certainly better than 11...Nb4?! 12.d4 Nbd5.
12.0-0 Re8 13.Rc1 c6 14.Qd2 Be6 15.b3

How should Black proceed?

*****
*****
*****
*****
*****
15...a5
I could have won rook and two pawns for bishop and knight by 15...Bxb3?! 16.axb3 Nxb3 17.Qb2 Nxc1, but after 18.Rxc1 the engines prefer White.
16.Nf3 a4 17.Nxd4 Qxd4 18.Rc3?!
The engines prefer 18.b4.
18...axb3 19.axb3 Ra3
Even stronger, according to the engines, is 19...h5 20.h4 Ra2.
20.Nc2
White offered a draw.
20...Ra2 21.Nxd4 Rxd2 22.Ra1 Nd5 23.Rcc1
23.Bxd5 Bxd5 24.b4 g6 also leaves Black well on top.
23...Rxd3 24.Nxe6 fxe6
White has lost a pawn, and still has a weak b pawn, but has bishop-versus-knight and has targets at b7 and especially e6.
25.b4
The engines prefer the passive-looking 25.Rab1!?, when one plan is to advance the b pawn with the aim of exchanging it to give Black a second isolani.
25...Rd2?!
25...Rb3 26.Rcb1 Rxb1+ simplifies Black's task.
26.Bh3?
The engines reckon White has almost equalised after 26.b5 cxb5 27.Rab1.
26...Rb2 27.Re1 Kf7 28.Re4 Rxb4 29.Re2 Re7 30.Rae1 Nc7 31.f4 Rb5
Not so convincing is 31...g6 32.f5! gxf5 33.Bxf5.
32.Bf1 Rd5 33.Bg2 Rd6 34.f5 Red7
Clearer is the engines' 34...Nb5.
35.Bh3 e5
The best way to give up the pawn.
36.Rxe5
36.f6!? gxf6 37.Bxd7 Rxd7 leaves Black with three pawns for the exchange.
36...Rd1?!
Almost certainly better is a move such as 36...Rd5, eg 37.R5e2 Rd2 38.Re5 Kf6 39.g4 R2d5 40.R5e2 Rd1.
37.Kf1?!
White seems able to put up a better fight with 37.f6 R7d5 38.fxg7 Rxe5 39.Rxd1 Re7 40.Bg2, although Black's extra pawn after 40...Kxg7 should probably tell in the long run.
37...Rxe1+ 38.Rxe1 Kf6 39.Rb1 Nd5 40.Re1
Not 40.Rd1?? Ne3+.
40...Re7 41.Rxe7?
The minor-piece ending is hopeless. White had to try something like 41.Rc1, eg 41...Ne3+ 42.Kg1 when 42...Nxf5?? 43.Rf1 Re5 44.g4 is drawn, according to the engines. But Black remains much better by not taking the f pawn.
41...Nxe7 42.g4 h5
The game finished:
43.gxh5 Nxf5 44.Ke2 Ke5 45.Kd3 Nd6 46.Kc3 c5 47.Bg2 b5 48.Bf1 b4+ 49.Kb3 Kd4 50.Ba6 Ne4 51.Bb7 Nd2+ 52.Kc2 Nf1 53.h3 c4 54.Ba6 Ne3+ 55.Kd2 b3 56.Kc1 c3 57.Bc8 Nd5 58.Be6 b2+ 59.Kc2 Ne3+ 60.Kb1 Kd3 61.Bf7 Kd2 62.Bg6 Nc4 63.Bf5 Na3+ 64.Ka2 b1Q+ 0-1
My final score of +6=1-2 saw me gain 17.6 Fide elo.

Back On Top

I WAS back on top board, having been upfloated to play the sole leader, in round eight of Bad Wörishofen U2000.
Spanton (1837) - Wolfgang Müller (1823)
French Rubinstein
1.d4 d5 2.Nc3 e6 3.e4 dxe4 4.Nxe4 Nf6!?
More popular are 4...Nd7 and 4...Bd7, but there are well over 4,000 examples of the text in ChessBase's 2020 Mega database.
5.Nxf6+
This is generally considered best. The idea is Black has to either damage his kingside pawn-formation or prematurely commit his queen.
5...gxf6 6.Nf3 b6 7.Bd3
I rejected the most-popular move, 7.Bb5+, because I thought 7...c6 8.Bd3 Bb7 did not especially help White, and anyway Black could play 8...Ba6!?
7...Bb7 8.Qe2 Qd6
WM at first placed his queen on d5, but did not let go, eventually releasing it on d6.
9.Be3 a5?!
Presumably designed to dissuade long castling by White, but reducing the chances of the black king finding safety.
10.0-0
The engines do not like 10.0-0-0!? but it seems playable.
10...h5?!
The engines prefer 10...Rg8 or 10...Nc6, but with advantage to White.
11.c4
The engines at first prefer 11.d5!?, and if 11...Bxd5?, then 12.Rad1, eg 12...Qe7 13.Bb5+ c6 14.c4 Bxf3 15.Qxf3 with a strong attack.
11...Nd7 12.Rfd1 Bh6 13.Rac1
I had been looking at the stronger move 13.c5, but for some reason I cannot recall, I did not play it.
13...Bxe3 14.fxe3?!
This restricts the activity of the white queen and removes the possibility of the protective pawn-move f3, so probably better was 14.Qxe3.
14...c5 15.a3!?
Planning to blast open the queenside with an immediate b4 if Black castles long.
15...Ke7?!
The engines prefer 15...h4, but still like White.
16.d5 Rag8 17.dxe6?
Good is the engines' 17.Be4.
17...Qxe6 18.Nh4 Rg5 19.e4 Rhg8 20.Rf1 Ne5 21.Rf5??
21.Bb1 is level according to Komodo10, but Stockfish10 much prefers Black.
21...Nxd3 22.Qxd3 Bxe4 23.Rxg5 Bxd3 0-1

Unusual Closed Sicilian

Boris Litfin (1948) - Spanton (1837)
Bad Wörishofen U2000 Round 7
Sicilian Closed
1.e4 c5 2.Nc3 g6 3.Bc4!?
Only fifth-most popular in ChessBase's 2020 Mega database, but it has been played by grandmasters and has some interesting points.
3...Bg7 4.Qf3 e6 5.Nb5 d6 6.Qg3 e5
Grandmaster Alexander Ivanov has twice played 6...Ke7!?
7.Qb3 Nh6 8.Qd3 Bf8 9.Qf3 Bg7 10.Qd3 Bf8 11.Qg3N Bg7 12.d3 a6 13.Nc3 Nc6 14.Bg5 f6 15.Bxh6!?
Stockfish10 comes to quite like this, but prefers 15.Bd2.
15...Bxh6 16.Nge2 Nd4 17.Nxd4 cxd4 18.Ne2!?
Stockfish10 comes to narrowly prefer this over the more-active 18.Nd5, but Komodo10 rates both moves roughly the same.
18...f5
I felt 18...Qe7 was too slow, but the text is somewhat loosening. Stockfish10 and Komodo10 suggest 18...Kf8!?
19.f4?!
Stockfish10 gives 19.exf5 Bxf5 20.0-0 with a slight edge for White. Komodo10 suggests 19.c3 f4 20.Qf3, but reckons Black is fine after 20...Bd7 21.Rc1 Rc8.
19...Qa5+
Judging by his body language, I suspect BL missed this.
20.c3 dxc3 21.Nxc3 Bxf4 22.Qf2
Black has won a pawn and has the bishop-pair, but where will the black king find safety?
22...Rf8
The engines suggest 22...Bg5!? 23.h4!? Bh6 24.h5 Bd7, slightly preferring Black, but the position is unclear.
23.0-0 fxe4 24.Qh4 Qd8??
I considered 24...h5?, but did not trust the move. The engines continue 25.g3, eg 25...exd3 26.Bxd3 (not 26.gxf4?? Qc5+) Be3+ 27.Kg2 Bf5 28.Bxf5 gxf5 29.Qxh5+ Kd7 30.Rxf5, when White has restored material equality and still has the safer king.
For some reason, 24...h6!? never occurred to me, but best seems to be the engines' 24...Bf5!?, one point being that 25.Qxh7?? loses to 25...0-0-0 as suddenly the black king is much safer than White's.
25.Qxh7 b5!?
Desperation, but now ...Bf5 is easily met by dxe4 as Black no longer has ...Qc5+.
26.Qxg6+ Kd7 27.Be6+ Kc7 28.Nd5+ Kb8 29.Bxc8 Kxc8 30.Qg7
Even stronger is the engines' 30.a4.
30...Kb8 31.Ne7 Qe8
Or 31...Qb6+ 32.Kh1 Re8, when Black is very passive.
32.Nc6+ Qxc6 33.Qxf8+ Ka7 34.Qe7+ Kb6 35.dxe4
Black's exchange sacrifice means the king survives, but White's only real problem is making the move-40 time-control (all games at Bad Wörishofen are 40 moves in 100 minutes, then 30 extra minutes to finish, with a 30-second increment from the start).
35...Qc5+ 36.Kh1 Ra7
Not 36...Bxh2? 37.Rfc1.
37.Qf6 Rh7 38.Rxf4!?
A practical decision, but 38.h3 was perfectly safe.
38...exf4 39.Qxf4 Rc7 40.h3

By giving back the exchange, White has reached the time control with a two-pawn edge. However, advancing the g and h pawns will inevitably expose the white king, so the win is not completely straightforward.

*****
*****
*****
*****
*****
40...a5
Hoping to somehow create counterplay on the queenside.
41.Rd1 Rc6 42.Rf1 Qd4 43.Qf2!?
This is the engines' choice, but it costs White a pawn.
43...Qxf2 44.Rxf2 Rc1+ 45.Kh2 Re1 46.Kg3?!
White is still winning after this, according to the engines, but it is simpler to start pushing the passed pawns with 46.g4.
46...Rxe4 47.h4 d5 48.h5 Kc5 49.Kh3?!
The engines prefer 49.Kf3, which clears a lane for the g pawn while being well-placed to intercept the black passer.
49...Kd4 50.g4 Ke3 51.Rf5 d4 52.Kg3?!
52.h6 d3 53.h7 Re8 54.Re5+! Rxe5 55.h8Q is one win. Another is 54.g5 d2 55.Rf1 Ke2 56.Ra1 d1Q 57.Rxd1 Kxd1 58.g6 etc. White should also win after 52.Rxb5 d3 53.Rd5 d2 54.h6 Rd4 55.Rxd4 Kxd4 56.h7 d1Q 57.h8Q+.
Even the text is probably good enough, but the win becomes trickier.
52...d3 53.Rf3+ Kd4 54.Rf1?
By now White is down to having only one winning line, namely 54.h6 Re1 55.h7 Rh1 56.Rf7, eg 56...d2 57.Rd7+ Ke3 58.g5 d1Q 59.Rxd1 Rxh7 60.Rd5.
54...d2 55.Rd1 Kd3 56.Kf3??
And now White is lost. Clearly the rook is going to have to be sacrificed, so it makes sense to do it immediately. After 56.Rxd2+ Kxd2, both players have to find only-moves, but the engines reckon the game should end in a draw by 57.g5! Re1! 58.g6! Ke3! 59.Kg2 Re2+ 60.Kg3 Re1 61.Kg2 Re2+.
56...Re1 57.Rxd2+ Kxd2
The tempo lost at move 56 costs White the game. It finished:
58.Kf4 Rf1+ 59.Kg5 Ke3 60.h6 Rh1 61.Kg6 Kf4 62.g5 Rg1 63.h7 Rxg5+ 64.Kh6 Rg2 65.h8Q Rh2+ 66.Kg7 Rxh8 67.Kxh8 Ke4 68.Kg7 Kd3 69.Kf6 Kc2 70.b4 axb4 71.Ke5 Kb2 0-1

It's That Petrov Again!

FOR the third time in my last four games with White, I faced the Petrov.
Spanton (1837) - Jo Steinschuld (1552)
Bad Wörishofen U2000 Round 6
1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.c4!?
I explained some of the pros and cons of this move in https://beauchess.blogspot.com/2020/03/an-unusual-line-against-petrov.html
5...Be7 6.d4 0-0 7.Bd3 Nf6
After 7...Ng5, Stockfish10 gives 8.Nc3 Nxf3+ 9.Qxf3 with a slight edge to White. Komodo10 gives 8.Bxg5 Bxg5 9.Nxg5 Qxg5 10.0-0, also claiming a slight edge for White. In both cases I guess the advantage is mainly based on White's extra central space.
8.0-0 d5
The engines prefer this to the more-aggressive 8...Bg4.
9.Nc3
In my previous two Petrovs, albeit not in exactly the same position, I met ...d5 by pushing past with c5, but here 9.c5?! can be well-met by 9...b6.
9...dxc4
Marginally more popular in ChessBase's 2020 Mega database is 9...Nc6, although the text has been played by strong GMs.
10.Bxc4 Nbd7!?
This may look slow but it has been tried by attacking players including Ivanchuk.
11.Re1!?
I felt it was not clear where the dark-square bishop should go, but the king's rook almost certainly belongs on e1.
11...Nb6 12.Bd3
Strong players have overwhelmingly preferred 12.Bb3.
12...h6!?
The engines like the text and 12...c6.
13.Ne5!?
A 2205 played Bc2!? in a 1998 game.
13...Be6
Not 13...Qxd4?? 14.Bh7+ etc.
14.Be3
14.Ng6 Re8 15.Nxe7+ Rxe7 is equal, according to the engines. However, Stockfish10 suggests 15.Rxe6!? fxe6 16.Qe2, when Black is the exchange up but his king's position is seriously weakened on the light squares.
14...Nbd5 15.Rc1 c6
Not 15...Nxc3?! 16.bxc3 Bxa2? as the bishop is trapped after 17.c4. Stockfish10 also points out 17.Bxh6!?, eg 17...gxh6 18.Qd2.
16.Bb1 Nxc3?!
This stabilises White's centre without even winning the bishop-pair. Probably better is 16...Nxe3. I intended 17.fxe3 with an unclear position. The engines reckon 17.Rxe3!? Re8 18.Ne2 is also approximately equal.
17.bxc3 Qc7?!
The engines prefer 17...Ba3, eg 18.Rc2 Re8 19.Rce2 Bd6 20.Qc2, when Stockfish10 reckons White has the upper hand, but Komodo10 rates the position as level.
18.Qc2?!
18.Qd2 is difficult to meet.
Stockfish10 gives 18...Ng4 19.Nxg4 Bxg4 20.Bxh6!, which wins a pawn as 20...gxh6? 21.Qxh6 f5 22.Bd3 b5 (to stop Bc4+) 23.Bc2 Rf7 24.Bb3 is a massacre after 24...Bf8 25.Qg6+ Bg7 26.h3!
Komodo10 suggests 18...Ba3, but then 19.Bxh6! Bxc1 20.Qxc1 strongly favours White,
18...Rad8?!
Komodo10's 18...Bd6 is met by 19.g4 with a strong attack.
Stockfish10 gives 18...Rfe8 19.Bf4 Rad8 20.Bg3 Qb6 21.Bh4 g5!? 22.Qd2! Nh7 23.Bg3 with a slight edge for White, although in practice the position would be very tricky for Black to play.
19.Bf4 Qa5?
This loses. I expected 19...Bd6, when I intended 20.c4 with promising play.

White to play and win

*****
*****
*****
*****
*****
20.Nxf7!
Also strong is 20.Ng6, but the text is best.
20...Bxf7
20...Rxf7 21.Rxe6 is hardly an improvement.
Also very bad is 20...Kxf7 21.Qg6+ Kg8 22.Bxh6 Rf7 23.Rxe6.
21.Rxe7 1-0
Black's resignation may seem premature, but White's threats include 22.Bc7. If the fork is stopped by 21...b6, then 22.Be5 is devastating.

Kitty News

VIC Rumsey won in 15 moves today in the Bad Wörishofen seniors when his higher-rated opponent resigned after blundering at least a bishop.
That beats Ray Kearsley's 17-move checkmate in round two, and means Vic is on track to win our 40-euro shortest-win kitty.

Blunder!

I HAVE been fighting a heavy cold, but today the cold fought back.

White has just played 9.e2-e4 in Bernd Kievelitz (1800) - Spanton (1837), Bad Wörishofen U2000 Round 5

*****
*****
*****
*****
*****
*****
*****
*****
*****
*****
8...Be7?? (1-0, 24 moves)

Revisiting An Unusual Petrov Line

I WAS on board one in today's fourth-round of Bad Wörishofen U2000
Spanton (1837) - Swidbert Dohmes (1735)
Petrov Defence
1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.c4!?
I explained the basic thinking behind this move, and the suggested antidotes, in a post earlier this month from Fareham: https://beauchess.blogspot.com/2020/03/an-unusual-line-against-petrov.html
5...Be7
The main alternative to the text is 5...Nc6, when the main line in ChessBase's 2020 Mega database goes 6.Nc3 Nxc3 7.dxc3 Be7 8.Bd3 Bg4 9.Be4 with a position Stockfish10 and Komodo10 reckon is more-or-less equal.
6.d4 0-0 7.Bd3 Nf6
If 7...Ng5, the most popular line runs 8.Nxg5 Bxg5 9.Qh5!? Re8+ 10.Kd1 h6 11.Bxg5 Qxg5 12.Qxg5 hxg5 13.Nc3 with a position favouring White, according to the engines.
8.0-0 Bg4 9.h3 Bh5 10.Nc3 Nc6 11.Be3 d5 12.c5!?
Via a slight transposition, we have reached the same position as in my game earlier this month at Fareham. Black then played 12...a6, but I pointed out the engines recommend:
12...Re8

How should White proceed?

*****
*****
*****
*****
*****
13.g4!?
I had forgotten that the engines give 13.a3, but it is possible the text is better.
13...Bg6 14.Bb5 a6?!
It looks strange to spend a tempo to oblige White to make a capture he planned to make anyway, but it is difficult to come up with a fully satisfactory move. The engines suggest 14...Ne4 15.Ne5 Nxe5!?, but after 16.Bxe8 it is hard to believe Black has full compensation for an exchange.
If 14...Rf8?!, then 15.Bxc6 bxc6 16.Qa4 or 16.Ne5.
15.Bxc6 bxc6 16.Ne5 Qb8 17.g5?
Over-elaborate. Simple and good is 17.Nxc6, and if 17...Qxb2, then 18.Qb3 Qxb3 19.axb3, eg 19...Bf8 20.g5 Nh5 21.Nxd5.
17...Bh5?
This hits the queen but leaves the black knight short of usable squares, so very much the lesser evil is 17...Ne4 when, after 18.Nxc6 Qb7 19.Nxe7+ Rxe7 20.Bf4, White is only slightly better, according to the engines.
18.Qc2 Ne4 19.Nxe4 dxe4 20.Nxc6?!
There is no need to defend the b pawn, so the engines prefer 20.Qxe4, and if 20...Qxb2, then 21.Rfb1, eg 21...Qc3 22.Qxc6 with a large advantage.
20...Qc8 21.Nxe7+ Rxe7 22.Kh2
White is a pawn up, but Black has decent compensation thanks to the white king's exposure and the attacking potential of opposite-coloured bishops.
22...Bf3 23.Rg1 c6
Hoping to use the b8-h2 diagonal, and effectively preventing White playing d4.
24.Bf4 Qf5 25.Bd6 Re6 26.Qd2 Rae8 27.Qf4 Qd5 28.Qe3 Rg6 29.Rg3
Apart from anything else, this threatens Bxf3.
29...Ree6 30.Qb3?!
White keeps a good edge with 30.Bf4, according to the engines.
30...h5
I thought my 30th move forced queens off as White is threatening a back-rank mate. It was only after moving that I saw 30...Rxd6! 31.cxd6 Qxd6, when the engines reckon Black has full compensation for the exchange.
31.Qxd5 cxd5 32.h4
With queens off the board, the white king is no longer in serious danger, and the main feature of the position is White's 3-1 queenside majority.
The game finished:
32...f6 33.gxf6 Rgxf6 34.Be5 Rf7 35.Rc1 Rc6 36.Kg1 Rb7 37.b3 g6 38.Rc2 a5 39.Rb2 Kf7 40.a3 Ra6 41.Kh2 Raa7? 42.c6 Rb5 43.c7 Ra8 44.Rg1 Rc8 45.Rc1 Bg4 46.Rc6 Bd7 47.Rf6+ 1-0

Eventful Game

Round three at Bad Wörishofen U2000.
Cornelius Renk (1758) - Spanton (1837)
London System
1.d4 Nf6 2.Nf3 d5 3.Bf4 e6 4.e3 Bd6 5.Bd3
This move has attracted strong grandmasters and scores an excellent 62% in ChessBase's 2020 Mega database.
5...Bxf4 6.exf4 0-0
Castling early can be a mistake in closed games. That does not come into play here, but 6...Qd6 was 'natural'.
7.0-0
CR said he spent a lot of time considering 7.g4!? It is not liked by Stockfish10 or Komodo10, but play would be tricky.
7...b6 8.Re1
If White tries to prevent ...Ba6 by playing 8.Qe2, Black can reply 8...a5.
8...Ba6 9.Nbd2!?
Komodo10 quite likes this. CR said his idea is to develop quickly, and he liked the way a pawn on d3 would cover the e4 square.
9...Bxd3 10.cxd3 Qd6 11.Ne5 c5 12.Ndf3 cxd4?! 13.Qa4
My idea was to meet 13.Nxd4 with 13...Qb4, hitting the pawn on b2, the knight on d4 and, latently, the pawn on f4.
Best, according to the engines, is 13.Rc1 Nfd7 14.a3 Nxe5 15.fxe5 Qd7 16.Nxd4 with a small edge for White.
13...Nbd7?
Good is 13...Nh5 as 14.g3 runs into 14...f6 15.Ng4 Nxf4!
14.Rac1
White is temporarily a pawn down but has a big lead in development.
14...Nc5 15.Qa3 a5
The engines prefer 15...Nfd7 16.b4 f6!? 17.bxc5 Nxc5 18.g3 (doubtful is 18.Ng4?! Qxf4 19.h3 e5) fxe5 19.Rxe5 Qd7 20.Nxd4, but give White the upper hand.
16.Nxd4 Rfc8 17.Nb5 Qf8
I rejected 17...Qd8? because of 18.d4 Na6 19.Nc6 with Ne7+ to follow. Even stronger, according to the engines, is 19.Nd6 Rxc1 20.Rxc1 and Ndxf7.
18.Rc3?!
White has a large advantage, according to the engines, after, for example, 18.d4 Ncd7 19.Nc6 Qxa3 20.bxa3 Kf8 21.Nd6 Rc7 22.f5 Nb8 23.Ne5 Rxc1 24.Rxc1.

How should Black proceed?

*****
*****
*****
*****
*****
18...Ncd7?
I missed that after 18...d4! 19.Nxd4, Black has 19...Nd5 20.Rc2 Nxf4 with much the better game.  So White should play 19.Rc4, when 19...Nd5 20.Rxd4 Nb4 21.Rc4 f6 22.d4 fxe5 23.dxc5 Rxc5 24.Rxc5 Qxc5 25.Nc3 is equal, according to the engines.
Stockfish10 also quite likes 18...Ne8 19.d4 Ne4 20.Qxf8+ Kxf8 21.Nd7+ Ke7 22.Nxb6 Rxc3 23.Nxc3 Rb8 24.Nbxd5+ exd5 25.f3 Kd7 26.fxe4 dxe4 27.Re2 f5, when White is a pawn up but Black's pieces are more active.
19.Rec1?
19.Qxf8+ Kxf8 20.Nc7 is strong.
19...Qxa3?!
Better, according to the engines, is 19...Rc5, but with White keeping an edge.
20.Rxc8+?!
20.Nxa3 Rxc3 21.Rxc3 may be enough for a slight edge.
20...Rxc8 21.Rxc8+ Qf8 22.Rxf8+ Kxf8 23.Nd6 Nxe5 24.fxe5 Nd7 25.d4 f5?!
Gaining space, but weakening e6. It was probably better to immediately seek counterplay with 25...Nb8.
26.f4 Nb8
The knight starts a powerful journey, but Black can hardly sit still as the white king will invade the queenside on the light squares.
27.Kf2 Nc6 28.Ke3 Nb4 29.a3 Nc2+ 30.Kd2
30.Kd3?! Ne1+ looks difficult for White, but may be holdable.
30...Nxd4 31.Kd3
Simpler equality is to be had by 31.Nc8 b5 32.Na7 b4 33.axb4 axb4 34.Kd3 Nb3 (only move) 35.Nc6 Nc5+ 36.Kd4 Na6 37.Nd8
31...Nc6
31...Nb3 32.Nc8 Nc5+ 33.Kd4 Nd7 seems equal.
32.Nc8 b5 33.Nd6 b4 34.a4 Ke7 35.b3 g5 36.g3 g4 37.Nb7 h6 38.Nc5
White offered a draw.
38...h5 39.Nb7 Kf7 40.Nc5 Ne7 41.Nb7 Nc6 ½–½

Upset Kitty

MY round-two game from the Bad Wörishofen U2000.
Spanton (1837) - Domenico Natoli (1628)
Sicilian Bb5(+)
1.e4 c5 2.Nf3 d6 3.Bb5+ Nd7 4.d4 cxd4 5.Qxd4 Ngf6
The main line goes 5...a6 6.Bxd7+ Bxd7 7.c4 with a Maroczy Bind in which White has exchanged his bad light-square bishop. But Black has the bishop-pair, and my main analysis engines Stockfish10 and Komodo10 reckon the position is equal.
6.Bg5 e5?!
Again the main move is ...a6, but the text has been tried by strong players. The big problem with it, as far as I can see, is that Black will almost certainly never be in a position to follow-up with ...d5
7.Qd3 Be7 8.Nc3 0-0 9.Bxd7!
Giving up the bishop-pair without even being provoked by ...a6, but White's lead in development and the backward d pawn cause Black problems.
9...Qxd7 10.0-0-0 Rd8 11.Kb1 a6?
Ironically, having passed up several good opportunities to play ...a6, the move is played at a time when it is a mistake.
12.Bxf6 Bxf6 13.Nd5
The threat of Nb6 is difficult to meet.
13...Qc6?
Better is 13...Qb5, when White's advantage my be manageable.

How should White proceed?

*****
*****
*****
*****
*****
14.Nxe5!
This cannot be successfully met.
14...Qc5
Obviously losing are 14...Bxe5?? 15.Ne7+, and 14...dxe5?? 15.Nxf6+ gxf6 16.Qxd8+.
15.Nxf6+ gxf6 16.Qg3+ Kf8 17.Nd3 Qc6 18.Qh4 Kg7 19.Nf4
The game continued another 22 moves, but the result was never in serious doubt. White's advantage is much larger than a pure material count would indicate.
But there was bad news from elsewhere in the tournament when, on bottom board, Ray Kearsley mated an unrated player in 17 moves, becoming the red-hot favourite to win our 40-euro shortest-win prize.