I’m missing something: White gives the pawn back and keeps four rooks on, with attack (and here, the split pawns are a disadvantage)
But this was not played! Now should be drawn easily
Look at board 4 in Russia - Argentina: Jakovenko in trouble?
It probably is a draw. Nakamura decided to give up his h6 pawn instead of sitting on the position. I hope he succeeds, but I doubt that he will. So, it may be a 2 1/2 win for China.
The Rh8 was not put on this earth to babysit the h6 pawn. Houdini suggests both players missed stuff around moves 42-43. But the general decision to sacrifice for activity is usually correct.
Congratulations to Nakamura for saving the game. I didn’t think he would. 2-2 looks like it will happen soon.
Nah Onishuk is lost.
The official site is reporting that Onischuk won. If so…how? Did Ding have a cell phone go off or something? Because Onischuk looks dead lost to me in the final position.
I’m glad it’s not just me this time.
Loss on time, maybe? Seems strange among grandmasters with so few pieces on the board and 30 second increment.
Or–scorekeeper wrote down wrong result? I suppose we’ll see.
I am watching chessbomb.com, and official site lists as a win for Onishuk?? Is there a transmission error, or did Ding’s cell phone go off at the most (in)opportune time?? It most certainly looked like a certain 2.5 - 1.5 defeat for the USA. Is this result correct?
David A. Cole
I’m thinking that they placed the kings on the wrong squares at the end of the game (which is how the dgt boards auto-detect win/lose/draw).
Must be an input error.
I find it hard to believe that Onischuk lost that position. Artichoke’s comment that we gave back the Grischuk gift from yesterday is appropriate.
Ouch ouch ouch: congratulations to China and MVP Ding Liren.
Hey masters: what’s the best drawing technique for Black? I would have played h5xg4. R + f&g pawns vs R + h pawn is generally a draw (but not when the g-pawn is passed, as in this game), so I’d think R + g&h pawns vs. R + f pawn is an easier draw.
I can understand Onischuk not wanting to get king cut off after 55… fxg5 56.Rd6+ Kf5 57.Rd5+ Kg6 58.Rxg5+ Kh6 (drawish, but still) Or 55…Ra6 56.f4.
It’s now corrected to 1-0.
OK, Mig Greengard’s Twitter feed indicates that Ding did indeed win that game. He’s having a monster Olympiad. China could sweep the men’s and women’s team events.
Basically, the US has to win the last round to have an outside shot at bronze, it appears. Hard spot to be in, but as the #5 seed coming in, it’s about where one would have expected them to be.
We may get paired with Armenia anyway: on the top boards, Nakamura-Aronian and Movsesian-Kamsky, I like our chances. Armenia is so strong on the lower boards, however.
Wouldn’t pretend to know what’s best, but one way I can think of is to keep the Rook on the 5th rank after g4. Then if White puts his Rook on the 6th rank, preparing g5, then drop the King to the 7th, so that I can capture on g5. If White plays f4, before g5, then play hxg4. Might you, this is still too complicated for me to be confident I won’t lose it against a Grandmaster with a Sudden Death TC.
I believe, Onischuk blundered with his Rh1 (in response to g5). He simply missed the Rd6 Rxh4 f4 sequence 
Michael Langer
I was also surprised by Onishuk’s approach - the h pawn, as played, is essentially useless. The key is to prevent connected passed pawns, which generally win. In that respect, keeping the f-pawn alows it to be used as a lever, when appropriate, even to be sacrified for an f and h pawn drawn rook ending.
That said, it’s not a trivially easy draw.
That seems reliable enough. I’m sure that Onischuk didn’t crack, he just went wrong. Ding Liren is relentless: hard to believe he’s only on board 3.
I had lunch with a NM who hadn’t followed the game. I described the position to him, and he too cited the rule of thumb that it’s easier to draw with f & h pawns than with connected pawns. “But you wouldn’t be telling me this if Onischuk had drawn,” he added.
Onischuk was short on time, which must have contributed to his failure to calculate the somewhat simple 55.fg5. The US team has burned too much clock time, particularly Akobian, Robson, and Onischuk.
I remember others suggesting it was easier if the pawns were split as well, but it’s hard for me to understand why. If black had, for example, pawns on h5 and g6, put white’s f,g and h pawns in any placement on the 3rd or 4th ranks (with reasonable placement of the kings and rooks), and I’d be confident of drawing the position against anybody. It’s very difficult for white to generate anything other than a R+P vs. R endgame.
Here’s a combination of Michael’s method (defending rook stays on same rank as h-pawn) and my suggestion (play hxg)
The defender kept the tension, then when the attacking rook dropped to the same rank as the f-pawn, the defending rook checked the king back, and only then did the defender play hxg.
After 39…Kf5 (see game below)
[Event “London+”]
[Site “London”]
[Date “1883.??.??”]
[Round “?”]
[White “Englisch, Berthold”]
[Black “Rosenthal, Samuel”]
[Result “1/2-1/2”]
[ECO “C48”]
[PlyCount “92”]
[EventDate “1883.??.??”]
[EventType “tourn”]
[EventRounds “26”]
[EventCountry “ENG”]
[Source “ChessBase”]
[SourceDate “1999.07.01”]
- e4 e5 2. Nf3 Nc6 3. Bb5 Nf6 4. d3 Bc5 5. O-O d6 6. Nc3 O-O 7. Be3 Bb6 8.
Bxc6 bxc6 9. d4 Ng4 10. Re1 Nxe3 11. fxe3 exd4 12. exd4 Bg4 13. Qd3 Qf6 14.
Rad1 Bxf3 15. gxf3 c5 16. Nd5 Qxd4+ 17. Qxd4 cxd4 18. a3 d3+ 19. Nxb6 dxc2 20.
Nxa8 cxd1=Q 21. Rxd1 Rxa8 22. Rc1 c5 23. b4 cxb4 24. axb4 Kf8 25. Rc7 Rb8 26.
Rxa7 Rxb4 27. Kf2 g6 28. Rd7 Rb6 29. Kg3 Kg7 30. Re7 h5 31. e5 dxe5 32. Rxe5
Kf6 33. Ra5 Rb4 34. h4 Rd4 35. Rb5 Ke6 36. Ra5 Rd5 37. Ra7 f6 38. Ra4 g5 39.
Rb4 Kf5 40. Ra4 Rd3 41. Ra5+ Kg6 42. hxg5 fxg5 43. Ra6+ Kf5 44. Ra5+ Kf6 45.
Ra6+ Ke5 46. Rh6 Rd6 1/2-1/2