I Beat My All-Time-High Rating of 2716

Here's what I learned along the way

It came down to the wire, but I completed my challenge of beating my all-time high blitz rating of 2716 on demand in a month. To be honest, it was harder than I thought it would be, but I made it in the end. Here are my three biggest takeaways from the challenge:

Consistency is key

For the first week of the challenge, my rating sank like a rock. I hadn’t played much blitz in a long time and that first week was all about knocking the rust off. After that, I gradually started to climb back up.

This is typical for many forms of chess training: the initial phase is just about getting oriented, and the real value doesn’t kick in until you’ve stuck with it for a significant amount of time. The problem is, a lot of chess players change their plan if they don’t see tangible results immediately. When was the last time you stuck with a plan for a month?

My plan for this challenge was to play and review four blitz games every day. As far as I know, the only other person who stuck out the challenge and played every day, Evan Seghers, also set an all-time high rating in Lichess rapid. So I’m calling it 2/2 with the method!

To paraphrase Lord Chesterton, “Consistency has not been tried and found wanting. It has been found difficult, and left untried.”

But what is it that makes consistency so difficult? Four blitz games per day is not a very heavy workload. In terms of total time commitment it’s far less than what many chess improvers do. It seems to me the hard part is not the work itself, but the expectations that creep up around it. When your rating goes up, you get excited that you might reach your goal. When it goes down, you become dejected and consider abandoning the plan.

There was one very strange moment in the course of the challenge. I seemed to be on the verge of reaching my goal about a week ago. I was in the middle of a game where, if I won, I would make it. I had a winning position on the board and 15 seconds vs. 5 seconds on the clock. It was at that moment that the power went out in my house. It came back on almost immediately, but I ran out of time while my internet router was restarting.

I had the sense that some greater power had stepped in my path and said, “You shall not pass.” But at the same time, it’s only one game. I could recover those rating points in a game or two. Even with such a bizarre occurrence, it’s really only a problem in your own head.

In fact, this process might work even better without a rating goal. Just play and review four blitz games per day without any expectations other than enjoying the games and learning from them. Maybe I’ll try this next month.

Look for patterns

One of the benefits of playing a lot of games is you then have the chance to look for patterns over many games. In a great post about how he went from the 1500s to 1800s in his 40s, Dan Bock said one turning point was when he resolved to “analyze every game like a detective trying to answer the question: Why am I so bad at chess?”

It can be painful to review your mistakes, but as poker player Phil Galfond says, “mistakes are the best possible clues on your path to improvement.” You’ll know you’re on the right track when you start getting excited to find mistakes, and you get really excited to find patterns of mistakes. As a detective, that’s how you break the case wide open:

“Oh my god, all the victims went to the same drug store!!!”

The biggest pattern I noticed in my own games is that I’m often unaware of my opponent’s intentions. The most pointed example of this was my game in Titled Tuesday against Anna Cramling. I had just played Qb6 when Anna thought for two+ minutes (the time control is 3+1, so this is really long!).

During those two minutes I was trying to figure out if I had gotten disconnected from the internet. Not for a moment did I consider that she might be calculating a knight sacrifice. I was shocked – SHOCKED!!! – when Nxd5 appeared on the board.

But of course, if you’re in the habit of looking for your opponent’s ideas, Nxd5 is totally obvious. What else could she be thinking about?

Sometimes, degeneracy works

Up until now I’ve made it seem like pretty smooth sailing. Be consistent, play a little every day, don’t worry about the results. Seems very easy, very responsible. But this isn’t quite the whole story.

I had been hovering around 2700 for a few days when I decided I would just go all out to finish the challenge. I played about four hours of blitz in one day, not reviewing my games, and finally made it.

I don’t know if I would recommend this as a strategy, but it worked for me this time. And maybe it’s not so dumb after all. Popular streamer Tyler1 achieved a Chess.com rapid rating over 1500 after just 90 days of playing chess. His secret? Playing a disgusting, unreasonable amount of games. He played over 3000 games in that period. That’s over 30 per day.

That seems completely insane, and yet, people with more nuanced study plans can struggle for years to crack 1500. In fact, in an article about which factors lead to the biggest improvement, Matt Jensen identified a group of “speed runners” who improved by an average of 400 points in a year almost exclusively through playing.

A few caveats: obviously, playing this amount of volume requires a significant time investment that is not possible for many people. And the all-play no-study strategy seems to work best when you’re just starting out. Once you reach an intermediate level and beyond, you might need to get more creative to continue to improve.

Still, many learning experts say that the best form of practice is the one that most closely resembles that activity you’re training for. Well, if you want to get better at chess, what could be better than playing chess?

Maybe we’re all getting too fancy with our training plans and we should just be playing a lot more chess.