Roses are red,

Violets are blue,

This experiment

Got out of hand.

I'm removing the poetry guidelines for now. Clearly there is public demand for this, so it may return periodically on individual posts which I deem appropriate. Keep your eyes peeled, budding bards!

Times in past of greatest growth

when through clear thought I tossed the yoke

Came when I set clocks for 3 hours

and cross multiple days worked, achieving power

Time blocking plus some deep work make

A rationalist able to take the cake

Separate from the previous one

was asking only "What needs to be done?"

through scouts I learned to craft of doing

No blame or evasion, but an end I was pursuing

(I don’t know if this qualifies as a full-fledged technique.)

One of the things I always found valuable was the use of "motivational" videos for mental bootstrapping. You may be saying, "TurnTrout, those videos are just platitudes told by people trying to build a brand", and you’d be totally right. However, the videos still worked for me when not much else did - when I was in a slump, I’d just click a few videos, and soon I’d be fired back up and able to motivate myself to start doing other good things for myself.

At this point, it’s important to stop watching, as you run the danger of bathing in the nice warm motivational bath and getting nothing done.

These days, I don’t need this as much (mostly due to having something to protect), but I found it to be a high reward, easy-to-use technique.

A rationality technique I am experimenting with now is using exercise to boost my energy during the middle of the day. It turns out energy boosts are more useful for me mid-day, because in the morning I am already energized. This is something that I usually do through exercise, but adding caffeine might be useful as well.

Listing all of the times I've improved rapidly in the past was a very interesting exercise. Many of the times where I made a really big improvement stemmed from making a decision that was hard to make in the moment, but locked me into a path toward something good. That's very much in line with the basic principle is "take advantage of willpower now to put yourself on a course in which you don't need willpower to do the right thing later."

Another idea that seems common to a number of the examples I can come up with is an idea that I might call "seize the moment." It's very similar to the first one but perhaps just slightly distinct. The basic idea is if you do one hard thing and it works, you may find that lots of other hard things suddenly seem easier, at least momentarily. I've had times where I was able to get on a roll of compounding improvements that way, and I think it can be surprisingly successful at allowing you to make big changes. 

Not a full technique but the seed of one - maybe it's an obvious idea, but most of the times I made major leaps forward were when I stopped wasting time on tangential things and just "stared at the sun". This is probably just a different name for deliberate practice. I need to remind myself that 70% of 100 is much bigger than 100% of 10 (doing an important thing passably vs. doing something unimportant perfectly).

I think the main generalization I could come up with from past successes is that if I want to make rapid improvement in any area, it helps to focus singlemindedly on just one thing for a while.

However, I haven't learned how to do that and then sustain any gains made as I shift focus to something else.

Thinking through my history of improvements, what seems to have worked for me:

1. Realizing I am dissatisfied with something.
2. Describe the problem and the desired outcome.
3. Search for solutions, compare multiple options.
Most techniques might require some tinkering/experimenting before they actually become useful. The longer feedback loop makes it is easier for snake oil salesman to enter such fields. To avoid these, check if the people promoting the solution are actually far above average in the ability the solution is supposed to grant, and if it is actually coming from the system and not from other factors. (e.g., if someone is selling you a way of making money, but the majority of his income is from selling seminars or running podcasts about it, that's a red flag)
4. Try implementing several versions, and keep using the one that seems to work best for my situation.
5. Practice the method, it should be running mostly on system one if you want to apply it effectively.

Example:

1. There was a period when I was dissatisfied with my grades during uni.
2. I thought about the issue and realized most of it was due to having started a second major and additional outside activities, so I had less time and more material to learn. I realized I needed to get more efficient at studying.
3. I searched for solutions on the net, and came across multiple memory enhancing tricks, mostly based on association and/or visualization of spatial environments. As these were usually promoted by memory world champions, whose records were verified and recorded (and who claimed to have been average before they started using the promoted material, for which there was no counter evidence), the techniques looked plausible enough to put in the required effort.
4. I tried two main versions: one using simply associations and one using mainly spatial locations (memory palace). At the start the first one seemed to work better, but with a bit more practice I realized the second one is the stronger technique that I need.
5. Through the rest of uni I kept practicing it even when it was not needed, making it an always ready to use skill of mine, cutting time required for cramming drastically while also improving the results.

Another important aspect is that may happen when searching for solutions, is realizing that it is possible to have it better, you can (and could already have) create a world, counterfactual to the current one, where you are way more satisfied. 

One rationality technique that I can infer from my past experiences is one I'm not really sure how to name; possibilities include "path divergence analysis," "counterfactual defaults," "adjacent life heuristic," "near-miss solutions," and "reality branch mining." The idea is to look at what common actions for you would be if your life had gone slightly differently (e.g. you went to a different school, were born in a different country, etc.), see what sort of actions you would commonly take under these conditions, and see if these actions have value in your current life circumstances. One example that I applied was in terms of realizing that if my college major were slightly different, I'd have learned to code in C++ by now, realizing that learning C++ has significant advantages for me even with my major being what it is, and finally learning C++.