One of the more baffling phenomena in higher-level amateur chess is a curse that seems to befall many National Master contenders, sometime in the expert stage but particularly in the high 2100s. To be sure, there are always many players plateauing at all rating levels. But the horror stories of master hopefuls stagnating or crashing down after reaching the cusp of 2200 have stood out the most to me. Hibiki Sakai of UPitt tumbled down the 2100s shortly after reaching 2198 USCF in early 2013, almost 3 years before he broke master. Franklin Chen became a master after winning the 2016 Pittsburgh Chess Club Championship, but not before he’d broken 2190 twice and spent nearly 9 straight years in the high 2100s.
I haven’t been a member of the 2100+ club for too long and recently enjoyed a rapid swing into the 2150s, but a tournament I played last Saturday (in which I gained a whole rating point!) has me convinced that I have a long task ahead of me – more on this below!
So what’s taking us so long?
Soundness and Practical Skills
Due to my relatively little experience with serious competition and formal study, I don’t have the theoretical background that many other experts exhibit. Upon returning to regular competition in 2014, I wanted to adopt a playing style that would hold up as long as possible; since then my priority has been to play as soundly as possible according to basic tactics and strategic principles. It’s definitely not the only way to go; for example, some players (like Isaac) are drawn to the deeper positional side of chess early on, and some specialize in channelling their speculative energy into some very dynamic chess. But attention to fundamentals (despite the occasional lapses) has proven critical to my recent success, and as IM Alex Katz has confirmed, makes its mark in higher-level competition as well.
I’d say most typical chess troubles can be explained by not playing soundly enough in some way. Most serious players have likely heard the above many times, but judging from my conversations and games with lower-rated players (particularly in the Class B-C range), the overemphasis of extraneous theory in chess learning culture and failure to apply basic tactics (yes, you shouldn’t hang rook pawns at will!) and basic positional principles during games are often to blame. However, chances are that if you consistently execute a sound strategy in most “normal” situations (not an easy task by any means), you are well on your way to at least 1800 USCF.
At this stage, it can be more difficult to pinpoint what to improve. Most Class A players certainly have the game to give experts a hard time. Some Class A players do exhibit some prominent weaknesses (I’ve seen a decent amount of atrocious positional action at the 1900 level, often compensated by otherwise tricky play) but the difference between Class A players and experts is often shown most clearly in critical positions when one has to get the job done (for example, at 1950, I was being consistently outplayed by 2100s in pressure situations even when up material and time). I’ve noticed some players too readily accept time trouble blunders as acceptable losses. It’s not necessarily a fair reflection on Class A players, but does indicate the importance of maintaining quality play in critical positions is often underemphasized when discussing improvement.
As before, awareness of basic principles should always be a priority, but “practical” skills such as time management and mental strength are often underemphasized yet hugely important, and or me, were a major factor in breaking 2000 and beyond.
Consistency and More Consistency
In the previous section, I didn’t talk about consistency against lower-rated players (relative to oneself) as it’s a direct consequence of improving everything else I mentioned. However, it’s certainly important at any rating level – it’s hard to gain rating if you’re losing more than occasionally to opponents rated 200 points lower!
Most experts have the skills and experience to fend off a lot of trouble from Class A/B players without having to pull an Alex Katz-style swindle. However, it’s this increased tactical astuteness that makes the consistency task so difficult when climbing up the expert range. For example, now rated 2157 USCF, I’d have to score 3/4 against four 2050s to gain any rating at all. Given that I was rated 2037 merely 4 months ago, this isn’t the easiest task to accept. I’ve had consistently good games against other experts, but I make mistakes like they do, and as I learned back in my 1950 days, outplaying someone in the first 90% of a game isn’t a ticket to the win.
The other new challenge that comes with gearing toward master is… having to score against masters! This is pretty obvious, as one shouldn’t become a master without holding their own against masters, but somehow having to improve on my 50% score against masters in 2016 (unfortunately, drawing 2250s [or equivalent] won’t get me to master) was a little daunting to accept.
Overall, these realizations have created some personally unprecedented expectations; I’m certainly ready to tackle the challenges, but objectively I’ve got a difficult (yet hopefully rewarding) task ahead.
Finally, Some Chess!
Hardly the most egregious example, but it’s probably not a good idea to lose positions like the following, from my second round game in a local G/30 tournament last weekend, against Eastern PA master Christopher Yang:
Black has sacrificed a pawn to reach this mutual time-trouble position. Both kings are in some danger; Black’s somewhat more so due to White’s ability to open the kingside at will. However, Black’s powerful knight and f-pawn are not to be ignored, since …f3 would seriously restrict any major piece action on White’s part.
25…Kh7 26. g5 Nf5! With 26…f3 stopped, Black quickly prepares …Ne3, seemingly to prevent White from using the g-file. The second reason is a little less obvious. 27. Qe2 Ne3 28. Qh5?!
Black prepared for an ending! In time trouble, it’s easy to forget that not all pawn-up, time-up endings are winning. It certainly takes some effort to untangle the situation on the kingside. It turns out White can’t avoid the ending, but had I been more aware of the challenges ahead, I would have chosen something more constructive like 28. Re1 instead. With 7 minutes to Black’s 4, the game continued 28…Qg4+ 29. Qxg4 Nxg4 30. Rf3 hxg5 31. Nxg5+ Kg6 32. Ne4 Rh8 33. h3 Ne3 34. R1f2 Rag8.
Black has made some scary progress, but amazingly has no immediate threats yet! White has enough time to try an Exchange sacrifice on e3 with 35. Re2, which leads to a strange equality. Unfortunately, time trouble does some weird things to the human chess brain and despite still being up time, I panicked with 35. d4??, which lost immediately to 35…Kf5+.
My choice of G/30 games to describe the difficulty of reaching master might seem odd. I haven’t much choice in the matter, as it’s part of recent changes to the Pittsburgh Chess Club tournament program. However, I’ve become more appreciative of quicker games in the last few weeks, partially because I’ve performed better than I ever have, but also because of how they mirror critical situations in longer games. Obviously they aren’t a perfect simulation of long games, but even playing longer games that don’t spill into time trouble bears some resemblance to playing under pressure, since most people aren’t quite as alert 3 hours into a game as they are at the beginning.
Fortunately, my game against Chris was my only loss of the day. Unfortunately, it was soon to be overshadowed my last-round draw.
Thompson (2047) – Li (2156)
Zack is a friend who is locally known for playing 1. b4 and a trickiness that compensates for his self-declared shaky hold on the chess fundamentals. In a rematch of our G/30 meeting from last month, the game began:
1. b4 e5 2. Bb2 d6 3. e3 g6 4. d4 Bg7 5. dxe5 Nc6!
I’d remembered this from analyzing our last game; an interesting idea to avoid having to take on e5 in an awkward fashion. Aside from playing 1. b4, White hasn’t done much wrong yet, but his position is looking kind of ridiculous already. 6. Bb5 Bd7 7. Bc3?!
I’m usually wary of these kinds of single-purpose moves unless the reason is particularly good. Even if White intends Nd2 from the start, Bc3 isn’t exactly the most robust way to defend b4 and doesn’t contribute to holding e5 (interestingly, 7. Nf3!? does hold the e5 pawn for the moment, due to 7…Nxb4 8. Bxd7+ Qxd7 9. Bc3! followed by 10. exd6, which is quite different from the Bc3 position above). Instead, the game continued 7…dxe5 8. Nf3 Nge7 9. O-O a6 10. Bc4 O-O 11. Nbd2 Nf5 12. Rb1?
This probably belongs in the same single-purpose category, but more pertinently it’s not really consistent with 1. b4. As long as White is playing it, it makes more sense to press the queenside now that he’s developed, before b4 starts sticking out too much. In fact, the immediate 12…Nd6 13. Bb3 Qe7 stops b5 and threatens to cause trouble on b4, and it’s not clear what White is doing on b1.
Things went downhill after 14. Nc4? Nb5 15. Ba1 and although Black can’t take on e5 just yet, 15…Bg4! causes too many problems for White.
Rather than deal with the upcoming …e4 White opted to jettison b4 with 16. h3 Bxf3 17. Qxf3 e4 18. Qg4 Bxa1 19. Bxa1 Nxb4 leaving Black up a clear pawn.
I later won another pawn, but in time trouble opted to simplify into a pawn-up major piece endgame with 7 minutes to White’s 3 (sound familiar?). The problem with craving simplicity in a dominating position is that some pretty big wins are likely to be missed. The bigger problem is that the actual positions may not be won in the end.
White played the seemingly obvious 33. cxb5 and I replied with 33…axb5?? The question marks aren’t for the resulting position, which is still clearly winning, but for missing 33…Rd2! 34. Qc1 Qe2 (several other moves win in a similar fashion) 35. Rg1 Qe5+ 36. Kh1 Qd5.
White is completely busted as 37…Re2 and 37…Re3 (depending on if White bothers to prevent Re2) cannot both be stopped.
Both of us stopped notating shortly after the skirmish at move 33, but I distinctly remember the last chance Zack gave me:
Perhaps the miss at move 33 was forgivable. Certainly not …Rd2?? (threatening mate but hanging f5; instead, …f4+ wins as Black will force the king to the first rank and either be able to trade rooks or win the h3 pawn easily).
My point wasn’t to complain about messing up so much as it was to confirm that experts do suffer from these kinds of mistakes (one of the points of Chess^Summit was to produce content from a more mortal perspective than that of grandmasters playing near-perfect games). In this case my opponent suffered quite a bit from his offbeat opening play and personally, overcoming seemingly simple mistakes like the ones above is going to be important for reaching the next level.
However, it’s possible that the Round 4 swindle was deserved due to the way I won my previous game…
Li (2156) – Yaskolko (1972)
My opponent, Maxim Yaskolko, is a strong local 13-year old who’s been sitting on the cusp of 2000 for a few months. I’d broken even with him in three games from 2014 and had played blitz with him during the lunch break to try to confuse him; however, if anyone was confused, it was probably me.
1. e4 c5 2. Nc3 d6 3. g3 Nf6 4. Bg2 Nc6 5. d3 g6 6. h3 Bg7 7. Be3 a6 8. Qd2 Bd7. The insertion of …a6 and …Bd7 is a little unusual; more normal is …Rb8 followed by rushing the a- and b-pawns. However, the opening isn’t so time critical and committing White to an early Qd2 does limit the ways White can deal with an attack on the b2-pawn (see later).
9. f4 O-O 10. Nf3 b5 11. O-O b4 12. Ne2 Ne8. Here’s the little nuance (intended or not) to Black’s earlier move order; White often plays Qc1 in these situations, but usually White hasn’t committed to Qd2 yet.
Unfortunately, 13. Rb1 gets awkward after 13…Qa5 so I settled on 13. c3 and preparing for the opening of the b-file. Instead after 13…Rb8 14. g4 Qb6 it didn’t occur to me that 15. c4 and 15. d4 were very legitimate options. 15. c4 completely closes off the queenside, giving White all the time necessary to pursue the usual kingside pawn storm, but I was a little wary of the long-term consequences, since White’s bishop isn’t the best ever. 15. d4 looked a little weakening and I didn’t want to risk getting blown apart in the center in a G/30 game. Instead, I played 15. Rac1 and faced the dilemma of how to deal with a2 after 15…Be6.
16. c4 is still an option, but I had doubts that Black wanted to keep a bishop stuck on the queenside. This wasn’t a very good assumption since there wasn’t really any other point to 15…Be6, and it turns out Maxim had calculated out …Bxa2 more than I’d thought. But I gambled with 16. f5?! Bxa2 17. Ra1 bxc3 18. bxc3 Bb3.
In situations where intuition plays a lot more of a role, there’s likely to be more inconsistency in a nontrivial series of moves, whether that’s due to tactical oversight or just poor planning (something that I am personally not good at). Indeed, over the next few moves I straddled between harassing the light-squared bishop and making gains on the kingside. 19. Rfb1 a5 20. Bh6 Qa7 21. Nc1 a4? 22. Kh1?
Far too slow, and Black takes control of the game. Both of us missed 22. Nxb3! Rxb3 (22…Bxh6 23. Qxh6 is more critical but with fxg6 and Rf1 to follow, White should gain the advantage with proper play) 23. Rxb3 Bxh6 (23…c4+?? 24. Be3) 24. Qf2!
In the game, I finally had to come to terms with not being able to take advantage of Black’s awkward bishop. After 22. Kh1? we continued 22…Qc7 23. Bxg7 Kxg7 24. h4 Nf6 25. Ne2!?.
Objectively this is kind of silly (Stockfish evaluates the positions over the next few moves as around -2.5), but I was down a clear pawn and had 7 minutes to Maxim’s 14. A rushed attempt at a kingside attack was, practically speaking, the only real try. 25…Nxg4 26. Nf4 Nge5 27. fxg6 hxg6 28. Ng5 Qd7?!.
This spoils nothing, but it’s an overreaction to White’s shallow Rxb3/Ne6+ threat. It’s important (as I’ve learned before and relearned in my last-round game) not to flock to defensive-looking moves to convert a possibly dangerous position. In this position, Black had the more active 28…Qa7! daring the king to step back to g1; with …Rh8 coming, Black’s king safety problem is just an illusion, compared to White’s.
The real mistake came after 28…Qd7?! 29. Bh3 e6??.
Black’s best was 29…Qa7 as similarly discussed last move. In one move, due to trying to keep the queen closer to the defense of the king, Black has created enough counterplay opportunities for White to equalize. After 30. d4! Nc4 31. Qg2 Qe7? (31…Qe8 was relatively best, but White has some tricks on f7 that I won’t go into) 32. Ngxe6+! Black finally fell apart with 32…fxe6??
Black felt the need to make me prove my time trouble moves, which proved fatal. As in, checkmate-level fatal. After 33. Qxg6+ Kh8 34. Qh6+ Qh7 35. Ng6+ Kg8 36. Bxe6+ Rf7 37. Bxf7+, 37…Qxf7 fails to 38. Qh8# so Black had to jettison the queen and got mated in a few more moves.
(To Maxim’s credit, he bounced back in the last round with a surprising upset against NM Yang and broke 2000 for the first time.)
Though I was happy to turn the game around in such an exciting way, it wasn’t comforting that it was necessary. In Round 1, due to some very dubious pawn grabbing, I found myself in a losing position on move 17 against a 2007-rated opponent but managed to pull off a win due to his severe time trouble. So I ended up winning two games I absolutely shouldn’t have won and failing to win two games I probably should have won (fittingly, I broke nearly even in rating, gaining a point in the process).
The issue wasn’t one of the particular games. It’s true that I had chances in the two games I didn’t win, but it was also possible that the two games I “won” would have given me 0.5/4 for the day. Ouch! If you’re looking for consistency, my performance isn’t the best place to start.
So the most important lessons weren’t shown by the ratings. While everyone is expected to have good/bad days, the bad days are less forgivable in the master contender pool. It’s clear I have room to improve with regards to the consistency necessary to become a master, but it will certainly be a worthwhile journey.