When wins slip away, the loser is rarely given much credit for saving the game. This may be justified, as the definition of a “lost position” implies that the winner must have messed up somewhere. But this does not mean the losing player must sit by idly, knowing the game is most likely lost, hoping for the other to mess up. There are specific ways to facilitate that kind of situation.
Of course, it may not be prudent to try too hard to save a truly lost position, and no matter how well one plays, lost positions are – well, lost, and should probably not be saved too often. But even if this only happens once in a while, the half (or full) point gained in the tournament is often significant, as are the rating points one saves from the better result.
More so if you are somehow able to pull this off three times in a row, as I managed to do about two weeks ago. Although the chaos of settling into the Bay Area has mostly pushed chess to the side recently, the area is, put mildly, overflowing with chess tournaments. So for my first weekend here, wanting to take a break from furniture, housing, and job setup, I went up to Berkeley to check out their monthly $50, 5-round weekend FIDE event. And despite being completely lost in three of my four games (I took a half-bye in Round 1 – it’s too hard to drive out of Silicon Valley on a Friday night), I emerged with a score 4/5 for clear first, with FIDE gains to match.
It must be stressed that, as always the case with swindles, there was an element of luck involved. My opponents (2100s USCF) were fairly strong, so I have no doubt that they would have converted these positions in most situations.
In two of my games, I was basically saved by running my opponents down on the clock (maybe if I had spent more time earlier, I could have played better!). In this situation, the conventional strategy is to hope for a more complex position.
In this decisive last-round game (in which I needed a win for first place), I’d spoiled a clear extra pawn and found myself in a losing position. Fortunately, the position was not so simple for Black (who had about 3 minutes to my 19 in this G/90;+30 time control) as the e-pawn dropping would be disastrous.
The upside of this situation is that the winning side (but with lower time) will often veer toward simple and direct moves, potentially overlooking… everything else. That’s not to say I (with a fair amount of time) was immune to this. After the forced 32…Qh4 33. Nd3 e2 34. Nh2, Black played 34…Ne3?! giving me the chance to sacrifice the Exchange with 35. Rxe2 Nxd5 36. Rxe8+ Bxe8.
Although that is objectively much best, I thought Black should be able to convert his slight extra material in that simpler position. Instead, I opted for 35. Rd4?? not realizing 35…Qxe1+! 36. Qxe1 (36. Nxe1 Rf1+ mates) Nxc2 wins. Although perhaps not such a complicated tactic, it’s not the first thing that came to Black’s mind with a minute on the clock. In the end, Black ducked with 35…Qh5? 36. Qd2 Bxd3? 37. Qxd3 and I was soon able to round up the e2-pawn with an easy win.
In Round 2, where my opponent also failed to convert a win in time trouble, it was enough for me to get to an ending that was comfortable, but not quite better for me.
After some earlier miscalculations, I found myself getting steamrolled in the center and on the queenside. At the same time, I had some counterplay, with …g5-g4 to come. Would it be enough? Probably not, I figured, but with 30 minutes to my opponent’s 10, there was no reason not to try.
The situation changed dramatically over the next few moves. With my opponent having only 2 minutes left, I was hopeful that my threat of 34…Rh8! 35. Bxh8 Qh2+ would keep me in the game. But had White found 34. Rd1! threatening 35. Nxa7#, I would have been a step behind. Instead, White hastily plowed ahead with 34. Nxa7+ Kd7 35. Rd1+ Ke8 36. Bf4, which is of course completely winning, but led to 36…e5 37. Bxe5 Bg5.
38. Bf4+ wins rather easily. But at long last, White finally cracked with 38. Bf6+?? Be3+ overlooking check. White doesn’t quite lose his queen for nothing: 39. Qxe3 Nxe3 40. Rd8+ Kf7 41. Rd7+ Kg6 42. Rxh7 Kxh7 wins it back and leaves White with some pawns to compensate for the Exchange, leaving a dynamically equal (according to the engines) ending. But practically speaking, White’s fragile queenside pawns are not easy to hold, and with one minute left, White could not hold against my extra Exchange.
But those games were nothing compared to my lone draw, in Round 2. If completely lost wasn’t bad enough, try completely lost down time. A simple oversight early on left me in a tough position and scrambling for time for the remainder of the game.
I hastily played 17…g6, thinking I would follow up soon enough with …Nf3+ with favorable trades. Until I realized 18. Qh6 threatened my rook on f8, and left me in a rather terrible position after 18…Rd8 19. Kh1 Qg7 20. Qh4.
In a position with so many issues (the pin, the dark squares in general, a looming f4-f5, and time), it’s important to know when to jettison material for relief or counterplay. After 20…Be6 21. f4 Kg8 22. f5 I was not in the mood to bunker down and hope White didn’t crash through, so I decided to sacrifice the Exchange with 22…Nxf5. White could accept with 23. Qxd8+ Rxd8 24. Bxg7 Nxg3+ 25. hxg3 Rxd3, but Black’s bishop pair will provide good compensation in the ending.
Understandably, White saw a brighter future in 23. Bxf5 Qxb2 24. Bxe6 fxe5 25. Ne4. Such a dangerous position would have been a good time for lots of concrete calculations, but I figured that with 20 minutes left, I might regret that decision later. So I went with my intuition, which was to finally take care of the systemic dark-square weaknesses on the kingside, 25…Bd4.
Matters improved greatly over the next few moves until I tried to be clever (a terrible idea in a dangerous position with little time). This culminated in 26. Ng5 Rd7 27. Rab1 Qxa3 28. Rf3 Bc3? threatening the cute 29…Qxb4. Never mind that I was on the run again after 29. Qc4 Rad8 30. Qxe6+ Kg7.
I was very lucky that my opponent eschewed 31. Qxd7+! Rxd7 32. Ne6+ with mate next move (it happens!). Unfortunately, his continuation of 31. Rf7+ Rxf7 32. Qxf7+ Kh6 was still quite winning. However, instead of the simple 33. h4 threatening mate next move, he gave me an opening with 33. Ne6 Qxb4!. I would say that it is again important to have a feel for when to sacrifice material – White continued with 34. Qf1 with the dual threat of Rxb4 and Nxd8. But in this case everything else simply got mated – not much choice there.
However, the resulting ending turned out to be far from simple for White to win. The ending was undoubtedly scary to hold as Black with little time (by now about 5 minutes), but White’s bank rank proved surprisingly problematic.
Ordinarily, Black should worry about losing the c3-bishop to forks, but White’s checks are largely useless, as any scenario winning the bishop (unless with check) allows mate on the back rank! g3 and h3 don’t help – the former gets bombarded with checks, and the latter gets hit with …Be5 when White’s queen and rook definitely should not leave the back rank. White played on for about 15 more moves, but burned a lot of time in the process and eventually agreed to a draw.
It must be reemphasized that a losing position is a losing position, and there is always some luck involved to execute these swindles. But it doesn’t have to be a passive effort, and if you take the time and energy to save a few games, you might just surprise yourself!