On February 22nd, 2022, I played in my first USCF-rated tournament – it was hosted at the Seattle Chess Club, and it was the U1200 Novice event. I had a decent bit of experience playing over the board (OTB) while using a chess clock, but I remember still being quite nervous – surprising because I had played decades of professional Magic: The Gathering events, mid-high limit poker games, and a few World Series of Poker events. I guess there’s nothing like your first bout of competitive play in a new game!

In this blog post, I want to recap how the year has gone, what things I’ve learned, what bad habits I’ve picked up, and how I approach not just the game of chess, but the lore of it. I’m mostly cataloguing this for myself, but I know there are quite a few adult improvers (the whole #chesspunks Twitter community is one such example) and my former friends in the Magic and gambling communities who may find it interesting and useful, so on we go!

U1200 Novice @ Seattle Chess Club (2-26-2022)

Coming into this tournament, I had very little practical experience studying the game. I had studied a few opening lines on Chessable and messed around with the analysis engine on lichess, but that was about it. A cursory look at my chess.com ratings shows an 822 rating in Blitz, with no games played in Rapid at that time.

My first round opponent was Chad Foster, who was very nice. He was provisionally rated 747 coming into the tournament. I had the white pieces and started off with the Accelerated London, which is basically all I was playing at this time. He went into an opening I had never seen before and still really don’t see, which is a Queen’s Indian Defense / Hippo-type shell.

I am playing quite fast, blitzing out the obvious London moves. A big reason I started playing the London is that I’m something of a notoriously slow player in Magic: The Gathering, and with a timed clock, I wanted to be able to play a system-like opening so I wouldn’t flag.

On move 11, Chad blunders a central pawn, which I nabbed after a brief think:

He then played Nf5, and I played the immediate and obvious Be5.

Oh, wait:

Without thinking, I take his knight.

With my pawn. Shortly thereafter….

Not great. It’s my first USCF rated game and I’ve gone from winning to basically dead lost inside 14 moves. Such is life as a novice player. Had I thought at any point inside these moves, I would have clearly found the right continuation, but I didn’t, so, well, I didn’t.

What follows from there is probably best summed up by the lichess computer analysis tab:

Somehow we play nearly 50 moves of blunder-free chess from there on out, with Chad finding ways to trade down into a simplified endgame and press his advantage. In this position, it’s not looking good for your boy, and the engine announces mate – though you don’t need Stockfish to tell you to push the c pawn to victory, I’m guessing:

Fortunately for me, Chad plays Rf1+ and my king gets off the back rank. I queen first, we trade down to a bishop v. pawn endgame, and it’s a draw.

Sheesh.

The second game is not nearly as exciting – I am playing the Seattle Chess Club’s president, John Selsky, who is rated 811 at the time of the tournament. I blunder fewer times than him and go on to win up two knights in the endgame, though it took far too long to figure it out, as multiple children helpfully pointed out to me after the game.

It was supposed to be a four round tournament, but I hadn’t prepared well regarding what to eat, and I was absolutely exhausted with both games finishing last. Making matters worse, I had a splitting headache from insufficient hydration and, well, food. I withdrew with a record of 1.5/2.0 and went home.

Side Note: Chess is Exhausting

I thought for sure that a lifetime of playing Magic: The Gathering at premier events, 12+ hour long poker slugfests, and overnight sessions at the blackjack tables would have prepared me to play a few chess games in one day, but it absolutely did not.

Pure focus in chess is nothing like poker, blackjack, or Magic – it is constant calculation. While I’ve adapted some over the last year, it hasn’t been more than a 20% relief in exhaustion, and after speaking to lifelong players, they say I shouldn’t expect much significant improvement in the future.

The Next Few Months OTB: Misunderstanding Game Selection

At this point, it’s customary to talk about all the terrible study habits you had, what you’re vowing to do better, and so forth. All of this is certainly true, and we’ll get to that, but those are not the largest mistakes I’ve made over the last 12 months which stagnated my growth – no, the largest mistake I made over the last year was applying the very wrong form of Game Selection.

Don’t know what Game Selection is? Here’s a quick definition I searched up from Microgrinder:

In poker games, if we are always playing against players who have equal ability to us or superior ability, then long term, we cannot expect to be a profitable player.

Theoretically, if we played a table of players who were of exactly the same ability as us and utilized the same strategy, then we would all lose money to the rake in the long term, so despite none of the players being better than us, we would still lose money.

This is why it is important to find games with inferior “weak” players to us. These players provide the “dead money” that covers the rake and provide profitable opportunities. It is our skill edge over these weak players that account for a majority of our profitability. At the poker table, the regulars duke it out for the fish’s money.

Microgrinder, Poker Game Selection

While this doesn’t exactly describe how I chose events, it unfortunately rhymed quite closely with it. A lifetime of choosing good gambling games (see also: Becoming a split-pot and mixed games expert in poker, learning how to play uncommon blackjack variants, etc) caused me to unconsciously think that playing people at or below my skill level was a good idea.

Four of the first six tournaments I played were U1200 Novice events, deliberately skipping the free Friday league matches at Seattle Chess Club when I found out that they were paired Swiss-style with no regard to rating. I didn’t want to show up at 8 PM on a Friday night with a provisional rating of 800 and get killed by someone rated 1450 or higher – that seemed like a waste of 2-3 hours and would be demoralizing to boot.

This was, to put it mildly, incredibly stupid. I should have known better, too – how people approached professional Magic in the days of Elo-based ratings is exactly how you should approach playing chess: You want to play the highest events possible that are reasonably within your skill level. In Magic, it’s even more pronounced due to the variance and luck inherent in the game – Elo already does not work correctly when you have large disparities in rating skill, and worse yet in Magic, there’s a huge element of randomness to any given match between two players! Thus the lower-rated player is heavily incentivized to play the best people they can find, exploiting this weakness in the rating system.

Somehow, I knew all that and didn’t apply it to my chess improvement path, which is quite silly. It’s compounded in stupidity by the fact that in chess, it is customary for the opponents to analyze with you post-game to give their opinions about positions, tactics, and overall feel of the game, and getting free lessons from players much better than you is an opportunity you shouldn’t turn down.

While I’ll briefly come back to this theme later in the post, the long story short is this: Play people better than you and don’t worry about your expected win/loss records. Loss aversion is no way to approach improvement.

Now, the Flawed Study Habits

In Magic, there’s a huge emphasis on deckbuilding – you need to be good at constructing your deck that you bring to the event, since you cannot change it once the tournament begins. Furthermore, knowing the metagame of what the other players are likely to bring to the tournament plays a big role in your success or failure as a player.

In chess… this is not true. And unfortunately, I fell into this trap quite early.

Behold, the site that has befallen a great many adult improver:

I partitioned out time per day to grind Chessable opening lines, memorizing as many variations as I could in what I usually play, thinking this was a great use of resources. Never mind my good friends and literally every master-level player saying this sucks and I should study tactics until my eyes bleed, if I could just memorize 18 moves of the Caro-Kann, Exchange Variation, I would have a huge advantage!

Instead of parroting the others that I ignored (which seems useless), let me illustrate why this idea is stupid via a very recent game I just played. This was in my weekly Friday match at the Seattle Chess Club against Liam Sarwas (rated 1300+). I had white, and six moves into the game, we had this position:

I have had this position or a position like it many times, and I was blitzing out my moves. Time control in these games is G/75 with an extra 60 minutes when you hit 40 moves, but there’s no increment or delay. Liam had been playing deliberately ever since I played 2. Nc3, and in the position you see above, he spent over 12 minutes before he played 6. … Bxc3 (which is fine, not great, but certainly not a mistake).

Liam had less than an hour on his clock and spent basically a quarter of his time in the opening, while I had spent nearly none, since I have these lines and their common variations down pat (please ignore that h3 is not a good move, thank you). I felt pretty good about the position and the time situation.

I lost this game after 48 moves.

The time advantage was wiped out by move 29, and unsurprisingly, Liam is simply better at chess than me, so he ended up winning a game where I had a small advantage the entire game.

While I no longer crush opening lines for 15-20 minutes per day on Chessable, I still use the site quite a bit to review tactical ideas from my online games and refresh my knowledge. The latter part has become a staple of my training, and I have Todd Bryant (@thestrongchess) on Twitter to thank for the modifications:

For learning huge Chessable courses, lately I like the settings All Moves/Quiz Immediately/1 Rep.

All Moves feels mandatory to me if the course is big. The other settings mitigate the tedious task of entering the same 15 move stem over and over again 3x per line.

— Strong Chess (@thestrongchess) January 18, 2023

If you use Chessable yourself, I recommend the changes above. Has helped me quite a bit.

The Probably Still-Flawed Habits, But They’re New

So what am I doing now that is so much better than what I did for the first three-quarters of 2022? Well, I’m doing more tactics and puzzles per everyone’s advice, but I can’t shake Bobby Fischer’s quote from my mind:

I spoke to the closest thing I have to a “coach” in my life, IM Marc Esserman, who said that he wasn’t sure that doing mindless puzzles and tactical analysis was all that helpful, but he had done it a lot as a juniors player himself. He recommended a two-pronged approach for novices who like studying:

1. Study key classical games of the old masters – Morphy, Rubenstein, Alekhine, et al
2. Complete thematic puzzles from positions you’re likely to get

The first piece of advice was one I’d heard before, but lost in a sea of games, I had no idea where to start. Fortunately, Marc has a bunch of free ones on his YouTube channel, and his paid Patreon has a curated list with his commentary that I have found immensely helpful as well. (If you join up at the $50 level, you’ll see him tearing apart a Semi-Slav game I played.)

Thematic puzzles was something I hadn’t heard of until someone pointed out that Lichess recently released a feature where you can study puzzles by themes and openings!

So I’ve been making my way through Marc’s Patreon videos, dedicating one day per week to study a video, and completing Lichess thematic puzzles in the openings I play three to four times per week for 15-20 minutes per day. This has been a significant improvement.

I also find time once per week to do deep study in other ways. Currently, I’m making my way through Evaluate Like a Grandmaster by FM Nate Solon and GM Eugene Perelshteyn, which I am really enjoying:

Like most adult improvers (and I would guess kids as well), abuse of Stockfish and engines was a significant weakness of mine, never learning how to evaluate a position while also training my visualization skills. This inexpensive book ($10 at the time of writing this blog post) helps you do both without boring the hell out of you with hundreds of pages of pontification.

A Better Approach to Playing Games

As I noted above, I now play in the once-avoided Friday night league at Seattle Chess Club. My first game was against Makarand Lahane, rated 1450. We played a Caro-Kann, Exchange Variation / Delayed Panov game, and it got extremely sharp very fast.

Just 24 moves into the game, we’re in this position, which is dead lost for black despite equal material:

Nxf6 is a free minor piece at best, because …gxf6 is mate in two, while …Qxf6 more obviously loses a full rook. Fortunately, Makarand didn’t calculate this, and I ended up pressing for the win with a passed pawn; Makarand found a series of only-moves, blockaded the pawn, traded it off, and offered me a draw, which I accepted.

This game made me realize that while yeah, the rating delta between Makarand and myself was 600 Elo points, and I was definitely losing for many parts of the game, the true delta in skill might not be as large as I thought. Again, this was something I already knew to be true in Magic but simply didn’t apply to chess until literally last month (Feb 2023).

The four games I’ve played in the Swiss-Style Friday night league have gone as follows:

• Lahane-Boddy, Caro-Kann (Delayed Panov) – Draw, rating 1450
• Boddy-Sarwas, Jobava London – Loss, rating 1270
• Song-Boddy, Semi-Slav (Meran) – Win, unrated player
• Adams-Boddy, Caro-Kann Advance (Botvinnik-Carls) – Loss, rating 1650

Not only will I obviously gain ratings points from this 1.5/4.0 performance, but the postgame analysis with my opponents in both losses were incredibly instructive.

Getting Better: Playing Above Expectation

I have loosely-held ratings-based dreams, but I primarily care about playing above-expectation. Studying engines like Leela Chess Zero (lc0), examining how the Maia chess bots work, and breaking down Lichess/Chess.com game ratings has helped me create a system where I take my current rating and calculate my expected performance against my opponents.

There will be games that I play like shit (round one of the Eastside Open comes to mind), but I don’t mind blips. I can handle variance over the long run, it’s one of the traits any successful gambler learns to develop the hard way. However, if I’m consistently turning in actual performances in games and tournaments that is above my expected performance (as I calculate it), then I’m pleased. When I lost to Henry Adams in my favorite line with black, I reviewed the game, put the moves into my hacked-together system, and was pleased with the result.

I got absolutely dismantled by Henry, but I made zero mistakes and zero blunders up until the end where I deliberately went for a low-chance desperado attack to simply increase the variance of the game, which was also a decision I stand behind. Henry played a single inaccuracy with no mistakes or blunders; when he is 800 Elo points higher than me, he should win that game. But what I can take from it – beyond his gracious postgame analysis – is how I played relative to expectations, and in that regard, I did quite well.

To me, this is no different than analyzing poker hands, blackjack bets, and sports bets of expected vs. actual, something all good gamblers catalogue and review. In a profession where you can run worse than your imagination can ever cook up, doubt and fear will begin to cloud your mind, and the only way out is the cold hard truth of the mathematics and statistics underlying your play. I’ve taken this approach to chess – perhaps later than I should have – and it’s helped me develop a game plan for the future.

And Yet, Getting Better is Not the Only Goal

What I’ve also learned from studying openings and talking to master-level players is that I truly enjoy the lore of the game nearly as much as I do the competitive side of it. Studying the openings of players is now something I do because I really enjoy exploring the history of the game through analysis of how the game came to be, not because I want to memorize a ton of moves in the Najdorf Sicilian or the Nimzo-Indian.

This realization has been quite the surprise, as I’ve not been one to care that much about the history of games I play; I’m also not an avid student of world history, either. But for some reason, chess is that interesting to me, which is pretty cool.

Someone Hit the Wrap it Up Button, Please

Alright, alright.

Let’s summarize this blog post and get out of here.

• Play all kinds of competition, especially regular league matches with players better than me
• Solve thematic puzzles, not random ones
• Spend deep study time on evaluation, visualization, and historical games of my own
• Study select games of the old masters who have similar styles (currently studying a lot of Harry Pillsbury)
• Limit hours spent on opening study and preparation (main counterexample: Transitioning to studying new lines like 1. e4 and other approaches to combating 1. d4 as black)

If you’ve made it this far, I left my Classical Chess Play Log for last. Here’s where I am cataloguing my classical games with various pieces of information that go into my analyses for playing above/below expectation.

Let me know what you think in the comments! Hopefully we’ll be back in early 2024 with the same zeal for improvement and some new ideas.