< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> 
Jun0203   drukenknight: that's okay bears. Last night I got into a dragon sicilian against pretty good player. We swapped the N on c6 so I had isolated his a pawn. I played h4 as white, he played h5, so it was a close positional game but all along I knew I had the extra connected pawn. I parried every move and after awhile the pawn formations finally broke apart and there were a couple of passed pawns on both sides. Found all the right moves and there it was! I had my d pawn all alone vs the K. I pushed it straight down the file. RIght into a stalemate on d8! 

Jun0303
  patzer2: Drunkenknight after 17...g3, 18 fxg3 Qd6, 19 Kf2 and white has saved his major piece, regained his pawn, secured his king, connected his rooks to both join the attack and maintained the initiative. Capturing the Knight at black's 17th move was not a mistake on the part of GM Kuzman. However, missing 18...Qd7 may have been a mistake that threw away a potential win for black. And if it is a refutation of the line starting with white's 16 g4, then 17...gxNf3 was clearly the best move. GMs are not superhuman, but they can usually quickly see combinations that are a bust after only three or four moves. So, I find it unusual that both GMs would have both missed this move. Perhaps they were in time trouble, or the game is not properly recorded, or there is yet a combination that was obvious to both of them that we are still missing. It is not unusual for amateurs to miss combinations, even after much study of a position, which GMs will see almost instantly. And that may well be the case here with the suggested 18 ...Qd7 in this game. Whether or not there is a refutation, Bears 092 deserves a lot of credit for finding this move. Keep up your study Bears, as you may be well on your way to an expert and master rating if you can consistently find these kinds of moves. Good luck! 

Jun0303   drukenknight: Boy patzer, you sure are not at a loss for words. What will you say after 17...g3, 18 fxg3 Qd6, 19 Kf2 Rxe3? ****
"I find it unusual that both GMs would have both missed this move." I dont get it. What do you mean BOTH? Only one guy is playing each side. 

Jun0303   Cyphelium: After 18. Qd7, maybe 19. gxf7+ will be fine after all. A/ 19. Kxf7 loses to 20. Rh7+ Bg7 21. Rxg7+ Kxg7 22. Qh7+ Kf8 23. Qxd7. B/ So black must play 19. Qxf7,
B1: and then my first thought was 20. Rh7 Bg7 (other moves seem to lose, 20. Qe6 21. Qg6+ Kf8 22. Bf5), but after 21. Kd2!? with the simple idea of Rdg1, I noticed that black is alive after 21. Re4!, for example 22. Nxe4 dxe4 23. Rdh1 and black doesn't seem to be worse, though it might be unclear. B2: So it might be better to play (after 19. gxf7+ Qxf7) 20. Kd2!. Now on 20. Re4, white can play 21. Rdg1+ Kf8 22. Qd1 threatening to bring in the queen by taking on f3. 22. Bg7 23. Rh7 looks bad and 22. Be7 23. Rh3 hardly looks better. Of course there other moves, but right now I think white still has a dangerous attack. 20. Kf8 (instead of 20.Re4) 21. Rdg1 looks pretty awful for black. But I'm sure you all have opinions on this. =) 

Jun0503
  patzer2: Cyphelium, thanks for the analysis. However, I am still leaning toward the opinion that 18...Qd7 is a win for black and refutes white's play beginning with 16 g4 in this game. You are correct that the best line of play for white after 18...Qd7 is 19. gxf7+,Qxf7 as 19 ...Kxf7 loses for black and other lines give white far less initiative for the sacrificed knight. I found the position so complicated that I went over to chesslabs.com and had their computer program analyze the position. Black according to this analysis has a winning advantage in all lines of this variation. In your B1 variation, the win for black found by the chesslabs.com computer analysis is 18. ..Qd7 19. gxf7+ Qxf7 20. Rh7 Bg7 21. Kd2 NC4+ 22. Kc1 Qe6 23. Rg1 Re7 at which point white's initiative has run out and black sits in a better position, with a secure king, more mobility and a solid piece up. In your B2 line, after the suggested 20. Kd2, the computer analysis gives 20.... Kf8 21. Rh7 (not sure why the computer passed up your suggested 21. Rdg1, unless 21...Qe6 followed by 22...Re7 holds the position here) Nc4+ 22. Kc1 Qe6 23. Rxc7 Re7 24. Nb5 Rxc7 (24...a6 instead seems better as it holds the piece advantage without giving up the exchange, but the computer passed it up) 25 Nxc7 leading to an advantage of a queen and three active pieces against a more restricted queen, bishop and rook. The computer gives black the advantage (2.16) equivalent of two pawns plus in this position. I'll take a look at your B2 line after 18...Qd7 19. gxf7 Qxf7 20. Kd2 Kf8 21. Rdg1 in more detail and post my results later. In any event, thanks for the detailed analysis. You seem to be on the right track in finding survival chances for white if there are any. 

Jun0503
  patzer2: DrunkenKnight, I do not wish to use this forum to play out a game continuation. If you have some specific analysis to justify 17...g3 then I would appreciate seeing it. My point about your 17...g3 is not that it is that bad a move, but that 17...gxf3 taking the piece is much better as white's game plan of 18. hxg6 is busted after Bear's suggested 18...Qd7. If you want to justify 17...g3 as the best move, then you must first prove that 17...gxf3 is not the far stronger move. And to do that, you need to prove that Bear's suggested 18...Qd2 is not a win for black. Otherwise, you are suggesting getting into complications with no sure advantage in order to pass up a winning move. So, give the refutation of 18...Qd2 a try. And if you can find a win with 17...g3, then post the analysis. However, I don't want to take the time, especially if 17...gxf3 followed by 18...Qd7 is a winning advantage for black. 

Jun0503
  patzer2: DrunkenKnight, if 18...Qd7 wins, then both GMs did indeed overlook it. White missed it when he initiated the 16 g4 sequence, overlooking the fact that 18...Qd7 would bust his combination and lose the game. Black overlooked it when he played the losing 18...Re7 instead of 18...Qd7. In most continuations GMs are looking about six or more moves ahead. Obviously, it didn't happen here. However, it is possible that white didn't overlook 18...Qd7 and that white still has a winning or equalizing continuation. If there is a winning white continuation after 18...Qd7, then it is possible that white and perhaps black did not miss it. Maybe we're missing it. So, the challenge is on. Can anyone refute Bear's 18...Qd7 as a winning advantage for black? 

Jun0503   drukenknight: patzer you sure are funny. you subjected no less than two (count them two!) of Bears/Cyph's line to a computer analysis. Happily black can stay alive. I am glad for that yes indeed. I was worried about having to survive the Magneto Bortzoff line but now I have no fear. THen you say you do not want to get into a game continuation. Okay neither do I. How about a few moves? Hey! Wait a minute! why dont you just put 17...g3 into the computer and see what happens? Thats the nice thing about computers we can at least see a few moves from chess nirvana and see what the good and bad pts are. I think I can hold a draw after 17....g3. if you take KxR I think my queen will come to c3 before the other rook give check. somethign line that. Is that not significant, patzer? you attach great significance to the 18th move. what if both lines are draw? what does it all mean? 

Jun0603   Cyphelium: patzer2> However, I am still leaning toward the opinion that 18...Qd7 is a win for black and refutes white's play beginning with 16 g4 in this game.
Yes, I have taken a closer look and white's opening play looks quite dubious, as you say. The problem is that the knight on c3 and the bishop on b1 are so passive. I'm still not totally convinced, but I didn't see that Qe6Re7 plan, which looks strong. <In your B2 line, after the suggested 20. Kd2, the computer analysis gives 20.... Kf8 21. Rh7 (not sure why the computer passed up your suggested 21. Rdg1, unless 21...Qe6 followed by 22...Re7 holds the position here) As I tried to show above, I believe the 21. Rh7 move to be a waste of tempo. So if you want to refute my analysis, you should try to refute the 21. Rdg1line =) But maybe that refutation isn't to far away. I haven't found one yet though. Back to analysis. 18. Qd7 19. gxf7+ Qxf7 20. Kd2 Kf8 21. Rdg1 Qe6 22. Qg6 I'm thinking that more pieces are needed here. That awful bishop on b1 needs to get a life. 22. Re7
There might be better moves, I just want to show why this plan looks a bit out of place here. 23. Bf5 Qf7 24. Qg4 with the threat of 25. Rh7. One sample line is 24. Ke8?! 25. Rh7 Qf8 26. Rxe7+ Kxe7 (only move; 26.Qxe7 27. Qg8+ etc, 26. Bxe7 27. Qh5+ Kd8 28. Qh7) 27. Rh1 with compensation. Or something. In practical game, both sides have winning chances after 21. Rdg1. Objectively speaking though, white may have pushed to hard. 

Jun0603   drukenknight: cyph on the 18...Qd7 line can white do 19 Rg1? I havent looked too much at this line. 

Jun0603
  Honza Cervenka: After 18...Qd7 19.gxf7+ Qxf7 20.Kd2 Kf8 21.Rh7 Nc4+ 22.Kc1 Qe6 white can play 23.Nb5!? which seems to be much better than 23.Rxc7. Now if 23...Re7, then 24.Nxc7. 

Jun0603   Cyphelium: Honza Cervenka> I agree that 23. Nb5 looks much better. So perhaps black shouldn't play 20. Kf8 but rather 20. Nc4+ 21. Kc1 Nd6. drukenknight> Yes, I gave that a brief look before posting my first kibitz to this game. I couldn't make it work after 18. Qd7 19. Rg1 Kf8 and now either 20. g7 Bxg7 21. Qh7 Bf6 when white seems to unable to get anywhere or 20. gxf7 Qxf7 when white has problems activating his rooks. (Rg1 cannot reach the seventh rank and is sort of in the way for the rook on d1.) 

Jun0603
  Honza Cervenka: <Cyphelium> After 18...Qd7 19.gxf7+ Qxf7 20.Kd2 Nc4+ 21.Kc1 Nd6 the move 22.Rh7 works perfectly: 22...Bg7 (22...Qe6?? 23.Qg6+ Kf8 24.Qh6+ Kg8 25.Rg1+ ) 23.Rg1 etc. 

Jun0603   Cyphelium: <Honza Cervenka> Yes, that is true. I thought that after 18.Qd7 19.gxf7+ Qxf7 20.Kd2 Nc4+ 21.Kc1 Nd6 22.Rh7 Bg7 23.Rg1 black could play 23. Ne4 24. Rhxg7+ Qxg7 25. Rxg7+ Kxg7 but I see now that this runs into 26. Nxe4 dxe4 27. Qxc7+. So right now it seems black has two promising lines to defend against, before he can claim being better after 18.Qd7. 

Jun0603
  patzer2: <Honza Cervenka> Glad to see you weighing in on this one. Having a former correspondence master analyze this will help. After 18...Qd7 19.gxf7+Qxf7 20.Kd2Kf8 21.Rh7Nc4+ 22.Kc1Qe6 your improvement of 23.Nb5!? initially looked strong enough to ressurect white's game. However, I put it to the computer for analysis and although the position gets real tactical, the complications seem to favor black (assuming the computer analysis is correct). After 18...Qd7 19.gxf7+Qxf7 20.Kd2Kf8 21.Rh7Nc4+ 22.Kc1Qe6 23.Nb5!?, the chesslab.com computer gives 23...Nxe3 24.Qg6Re7 25. Qh6+Ke8 26.fxe3Rxh7 27.Qxh7Qxe3+ 28.Rd2Qe1+ 29.Kc2f2 30.Nxc7+Kd8 with advantage for black. One possible continuation here is 31. Qg8+Kxc7 32. Qg3+ to capture the passed pawn at f2, playing on a piece down but hoping for drawing chances in an apparently lost game. However, Honza, I have no doubt you will find some improvements on the computer analysis in this complicated sequence. Good luck and have a safe and enjoyable weekend. 

Jun0603
  Honza Cervenka: <patzer2> Why 27.Qxh7 in your line? Why not 27.Bxh7? 

Jun0603
  patzer2: <Honza Cervenka> It's the computer's line, not mine, but you certainly do seem to have taken the sting out of the black attack in this continuation with 27 Bxh7 instead of the computer's anticipated 27. Qxh7.
If then 27...Kd7, then 28. Qf4 is strong for white. If 27...Qc6+, then 28.Kd2 seems to hold the position nicely for white.
If 27...Kd8, then 28. Rg1 also looks like it might offer an advantage to white. Can't get into the computer program now, but will look at it later to see if it has an answer for 27. Bxh7 in this line. I might mention that the computer had three other winning (according to the computer) alternatives it offered for a response to your 23 Nb5!?, but I want to analyze them in more detail before posting them, as I'm realizing you can't always trust computer analysis  even in miuddle games. And with computer chess programs assisting, you still have to supplement them with your own study and analysis to understand them. For now, I have to admit that white at least seems to have equality if not an advantage in the two lines (your's and Cyphelium's) found for white so far in response to the suggested 18...Qd2 in this game. Appreciate both of you taking the time to do the analysis of bears092's suggested move to find a surviving and perhaps winning continuation for white. I learned a lot from both your's and Cyphelium's analysis. 

Jun0603
  patzer2: <drukenknight> Given that 18...Qd2 may at best offer black only even chances against best white play, let's take a second look at 17...g3. Perhaps I dismissed it as inferior to 17...gxf3 a bit too hastily. Certainly, it seems to offer black active counterchances in a very tactical game with lot's of chances for white to go wrong. We may have both missed the best moves after 17...g3 in our initial continuation. After 17...g3, I'm now thinking that 18. hxg6 may be preferable to 18. fxg3 because of the immediate 18...Rxe6+ weakening white's pawn structure. One line (computer analysis) for white, with only a very small advantage is 17...g3 18.hxg6Nc4 19.gxf7+Kxf7 20.Qh7+Ke6 21. Bg6gxf2 22. Kxf2Rf8. White has nice play and mobility against a fairly exposed, but defensible, black position. However, I prefer white's chances in practical play here. In our continuation of 17...g3, 18.fxg3Qd6 19. Kf2Rxe3, I took your suggestion and submitted it for computer analysis (which has not always been 100% accurate as I've found) and came up with the following continuations (all favoring the white side): 20.Nb5Qe7 21.hxg6Nc4 22.gxf7+Qxf7 23.Qh7+Qxh7 24. Bxh7+Kf7 25.Nxc7Rf8 (0.46 plus score) 20.Nb5Rxf3 21.Kxf3Qd7 22.Nxc7Rc8 23. Rc1gxh5
(0.36 plus score)
20.Kxe3Qxg3 21.Rdf1c5 22.hxg6cxd4+ 23.Kd3 dxc3 24.gxf7+Kxf7 25.bxc3 (1.63 plus score) 20.Kxe3Qxg3 21.Rdf1 Bg5+ 22.Ke2Ba6+ 23.Kd1Bxf1 24.Rxf1 (1.70 plus score) 20.Kxe3Qxg3 21.hxg6Re8+ 22.Kd2Qxf3 23. gxf7+Kxf7 (1.80 plus score) I find it interesting that the computer thinks the king can survive taking the rook at e3. If white tries to hold the Knight with 21. Rdf1, I'm thinking 21...Rd8+ followed by 22...Nc4+ doesn't look good for white's survival. So, I would be inclined to play the 20. Nb5 lines above unless it was a correspondence game with lot's of time to analyze the complications following 20. Kxe3. In any event, I'm sure you and others will find analyzing the 20. Kxe3 lines, to see if black really can win after grabbing the rook in this wild continuation, interesting and instructive. Again, I'm sorry I dismissed your 17...g3 try so readily as inferior to 17...gxf3. Against best play, it may well offer as good a chance as 17...gxf3 for black's survival in this game. Enjoy the weekend and stay safe. 

Jun0803
  patzer2: <Honza Cervenka> Going back to Bears092's recommended 18...Qd2 improvement in the game and your recommended 19.gxf7+ Qxf7 20.Kd2 Kf8 21.Rh7 Nc4+ 22.Kc1 Qe6 23. Nb51? (your improvement over the computer's anticipated 23. Rxc7), I thought you might be interested to know the chesslab.com program has still tried to analyzed this to advantage for black with a small end game plus. The second recommended improvement you found in the computer analysis at the end of the sequence 23...Nxe3 24.Qg6Re7 25. Qh6+Ke8 26. fxe3Rxh7 was 27. Bxh7!? (your recommended impvovement over the computer's 27. Qxh7). Now after your 27. Bxh7!?, black cannot play 27...Qc6+?? (hoping to force the retreat N(b5)c3 or capture the Knight on b5) because of 28. Kb1!Ke7 29.Bd3 Rh8 30.Qf4 a6 31. Rc1 Qxc1+ 32. Kxc1 Rh1+ 32. Kd2axb5 33.Qxc7+Ke6 34. Bxb5 with an easy win for white. NOTE: After 27...Qc6+?? in this sequence, white would go wrong with 28. Kd2?? (instead of 28.Kb1!) as 28.....Qxb5 29. QxB(f6) Qxb7+ Kc3 30. Ba6# follows for black. So after 27.Bxh7 the computer analysis gives best play as 27...Qe7 28.Bf5 Kd8 29.Rg1 Bc6 30.Rg8+ Be8 31.Bg6 Kd7 32. Nxa7 (if 33. Nxc7 then 33...Kxc7 holds for black) Bxg6 33.Rxg6 (surprisingly the rook at a8 is poison as 33. Rxa8 f2 34. Qh3+Qe6 35. Qf3 Bh4 36. e4 dxe4 37. Qe2 Qxa2 results in advantage for black) 33...Rxa7 34. Rxf6 Qe4 35. Rf7+Kc8 36. Qh3+ Kb7 37. Qxf3 Qxf3 38. Rxf3 Rxa2. Black has only a miniscule advantage if any in this position and it's beginning to look drawish. However the position that began this analysis is so complicated, one would think the better player will win after your recommended 23. Nb5!? or 27. Bxh7!?. I'd appreciate it if you can find another hole in this computer analysis to try for a white advantage here. I wonder if Fritz 8, Crafty, Shredder or any of the other programs could do a better job of analysis. I guess even computers make mistakes on long analysis, or else they would never lose to humans. In any event, I've found analyzing GM games with the assistance of computers and fellow Kibitzer's on this site very helpful. Thanks again for the analysis and instruction. 

Jun0903   Cyphelium: <patzer2/Honza Cervenka> I've given the position after 27. Bxh7 a look. It seems ok for white. 19.gxf7+ Qxf7 20.Kd2 Kf8 21.Rh7 Nc4+ 22.Kc1 Qe6 23. Nb5 Nxe3 24.Qg6 Re7 25. Qh6+ Ke8 26. fxe3 Rxh7 27. Bxh7 Qe7 as indicated above and now 28. Kb1!? is worth a try, creating the threat of 29. Nxc7. After 29. Kd8 30. Rc1!? c6 31. Nc3 we have an interesting position. Though e3e4 won't be immediatly possible due to Bf6g5xc1, this way of opening the centre is clearly on the agenda. The point with 30. Rc1 can be seen in variations such as 31. Bg5 32. Qg6 Bxe3? 33. Qg8+ when 33. Kc7 34. Nxd5+, 33. Qe8 34. Qxe8+ Kxe8 35. Re1 and 33. Kd7 34. Bf5+ Kd6 35. Qg6+ all lose. So black has problems moving his king out of the way for his rook, giving white some time to prepare e3e4. I think white has compensation for the pawn, since black's king is awkward. 31. f2 32. Bd3 still preparing e3e4. 

Jun1003
  patzer2: <Cyphelium> Good analysis! I believe what you found beats the computer analysis in this line. After 18...Qd7 (Bears092's suggestion) 19.gxf7+ Qxf7 20.Kd2 Kf8 21.Rh7 Nc4+ 22.Kc1 Qe6 23. Nb5 Nxe3 24.Qg6 Re7 25. Qh6+ Ke8 26. fxe3 Rxh7 27. Bxh7 Qe7 as indicated above, you came up with the interesting improvement 28. Kb1!? which indeed does seem to offer white advantage after 29...Kd8 30. Rc1!? Over the board, I much prefer white's chances in this position.
The computer analysis I ran pretty much analyzed this position to a decisive white winning advantage after your 30. Rc1!? If black has any salvation here it might be (after your 28.Kb1!?) in playing 28...c6 29. Qg6+Kd7 30.Qf5+Kd8 31. Nc3, but this also seems to lead to much the better chances for white. Of course, the position is so complicated, chances are improvements could be found for both sides. In any event, white seems to have equal, if not better, chances in your's and Honza's recommended lines after 18...Qd2. So, I must tentatively conclude that, in this game, GM Movsesian was justified in playing the interesting 16. g4!? followed by 17. h5!?, especially if he was ready in advance to spring these kinds of complications against an unprepared opponent. Again, thanks for taking the time to look at the analysis and recommend improvements. I've gained a greater appreciation for the kind of preparations professional players must undertake when springing surprise moves in serious tournament play. I've also learned a lot from yourself, Honza Cervenka and our fellow kibitzers in looking at Bears092's suggested 18 Q...d7 in this game. 

Jun1003   drukenknight: patzer: thank you for your kind remarks re: g3; I was simply playing according to theory which says that since black is ahead he should not try for more material but rather to exchange. This theory has never failed me, and although I still lose games, it is not the theory that is wrong it is my moves! ANyhow going back to the position on black's 17th, perhaps 17...g3 is not in accordance with theory after all. Looking again at the board (admittedly I dont have much time today) it appears that the most effective way to encourage exchanges would be 17...Bg5 since if the B can take on e3 (followed by Rxe3) that would break up a triplet of connected pawns. A B for 3 connected pawns? Sounds good to me, I think it's equal. Now it's possible that I have missed some positional problem, and there are a lot of positional problems in this game, because of blocked bishops, exposed KIngs, etc. But w/o spending too much time on analysis (got of lot of catching up to do today) I have a strong hunch here. Anyhow, I thank you for taking time to plug 17...g3 into the computer and I wonder if it's not too much trouble to do the same w/ ...Bg5? You never know what will come out, but Ive seen a lot of interesting things happen when I play moves strictly in accordance with theory and not worry too much about complicated lines. Do you remember that spassky/suetin game where the gang had this tremendous debate on the 26th move? lots of examples to give. bye for now. 

Jun1003
  patzer2: <drukenknight> You might want to try accessing the computer yourself. Just go to chesslab.com to access their java chess board. Then, follow the instructions to setup a position and have the computer analyze it. I also noticed Amazon is selling Fritz 8 for under $40 as of June 19. For now, I've had all I can handle of this game's analysis. 17. Bg5 looks interesting and I may plug it in later, but I think I'll pass at this time. Your theory of not taking material when material ahead is interesting, but I'd recommend you be flexible with this rule. It all depends on the position. Sometimes grabbing material and holding on to beat back an unsound attack is the only winning method. Other times, refusing the offer of material, especially when it gives your opponent advantage, is the best way to go. You might want to try playing both sides of a gambit, such as the sicilian smithmora gambit or evans gambit to get a flavor of both sides of this approach. Much can be learned both in executing and in defending against gambit play. 

Jun1003   drukenknight: Oh there is a way to do that? Okay I'll have to go and try that. thanks for the tip. As far as being wedded to the rule about taking not taking material: the only time to violate this rule is when a blunder has been made. And much of the time, there are already blunders that have been made so you have to really think. But the rule I stated assumes that no blunders have been made. If that is true then following the theory is at least prudent and oftentimes mandatory to avoid destruction. 

Feb0118   TheBish: Nice finish, easy tactic to miss... on both sides! Of course Black blundered with 19...Bxg7?? (he's winning after 19...Kxg7). This hasn't been a POTD since 2003! Maybe time to bring it back? It was a Monday puzzle before, which it looks like it should be, but could pass for a Tuesday puzzle too. 


< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> 


