I’ve taken up solving Rubik’s cubes in my spare time (and when I need a little break throughout the day). Here’s a wonderful guide I’m following that lays out all the algorithms needed to go from start to finish. Disclaimer: I’m not claiming I have ever solved a cube completely on my own, and I think that misses the point. It’s a very, very difficult task, and, to be frank, not one that I have the time or mental stamina for. I do think there is still tons of merit in memorizing and understanding algorithms other people have discovered. It’s kind of like research; stand on the shoulders of giants, don’t try to reinvent the wheel. Knowing which algorithm to apply at what time and practicing enough to do it quickly builds mental dexterity.

The beginner’s guide is a nice starting point, as you build some intuition for solving the first layer, then depend on one algorithm for the second and five for the third. After practicing this for a week or two, I was able to solve a cube without looking at any guides. For fun, I’ll give a rundown of my speed at this point.

As many competitions do, I use a trimmed mean of five to record my results. I randomly scramble and solve the cube 5 times, remove the best and worst times, and finally average the remaining 3 for a final score. Additionally, upon finishing the scramble, I get 15 seconds of inspection time before the timer starts (I’m never too strict about this, though). A good goal I have each time is to mentally map out the first few moves to get 2 or 3 pieces of the first cross together.

On 4/16/26, here are my results using the beginner method:

  1. 2:19
  2. 1:57
  3. 2:00
  4. 2:15
  5. 1:52

Average: 2:04

Pretty fun, but I knew I could do much better. I moved on to the speedcubing page on the same website, where we start putting the pedal to the metal. This now involves solving the first two layers at once, then applying what’s called 2-look OLL and then 2-look PLL for the final layer. The first two layers rely on a lot of intuition and some algorithm memorization, and I delegate that to later. What I focused on was the final layer, which involved memorizing around 13 new algorithms.

After spending a week doing this, here are some results on 4/23/26:

  1. 1:39
  2. 1:13
  3. 1:23
  4. 1:46
  5. 1:23

Average: 1:28

That’s a 29% time reduction, or a 41% increase in speed! Now, memorizing the algorithms wasn’t trivial, and it involved a lot of repetition and practice, but honestly they weren’t that bad. And they built some intuition for how the cube works, which was really cool to feel. As a mathematician, I like rigor, but intuition is just as important.

So, this is where I leave off on 4/23/26. My next goal is to improve my speed on the first two layers, and maybe I can get close to solving the cube in a minute.

As a final note, I’m sure there are people who have spent less time practicing and can already solve it faster than I can. And that’s pretty cool. It’s pretty easy to feel bad when things don’t come as easily to you as they do to others. But I’m finding this hobby fun, and I’m competing against myself. I tend to like hobbies like that. There’s a beauty and simplicity in taking time and effort to hone a skill, even if I’m not the best out there. End of soapbox.