GitHub

purplesyringa

Things I wish I knew before starting speedrunning

I started speedrunning Minecraft a while ago. It’s an interesting niche with many tutorials and a helpful community, so all the information you need is freely available. But I used to struggle a lot, because it wasn’t clear to me how to learn to execute this stuff – partially because my hand-eye coordination and reaction time suck, but also because most of the advice on improving mechanics is literally “play more”, and for me personally familiarity doesn’t easily translate into improvements.

I think I found an approach that works for me, and I want to share it here. Since it focuses on how to deal with stuff mentally rather than on MCSR specifically, I think it should be applicable elsewhere in case traditional approaches fail for you, so hopefully it’s going to be useful to more than one person.

"Speed"running

This a common novice mistake, but I never saw anyone talking about it and it took me a long while to figure out myself.

The purpose of speedrunning is to beat the game as fast as possible. From here, you might infer that you need to push yourself out of the comfort zone bit by bit, and as you do so, your skills will grow and you’ll get faster. This doesn’t work.

My mental model for this is that your brain has a fixed bandwidth at which it can process information. When you panic or push harder than usual, you’re increasing the input without increasing your processing capabilities, so your brain starts to lag behind. If you try to build skills this way, you will condition yourself to never quite understand what’s going on at the moment, and any skills you’ll obtain will be bound to this situation. Your entire modus operandi will revolve around praying your muscle memory can resolve everything without including your consciousness in the process.

Instead, only do the things you’re confident in, at the speed at which you’re comfortable. Not technically achievable, but comfortable. And then, calmly, experiment, learn tips and tricks, and inspect your mistakes. This will grant your confidence to do things faster, and that’s something you can integrate into your runs.

Knowing stuff

I can’t improve simply by doing stuff. Every time I improved, it was because I was missing something, and then I had an “aha!” moment and realized what it was.

If I had to explain the point in one sentence, I’d say this: if you can’t achieve something consciously, there is no point in training your muscle memory, because it won’t succeed either. Instead, try to improve until you can do the thing while focusing on it, and then delegate that to your muscle memory.

Here’s an example on how to apply this to dealing with mobs.

Each mob has an attack distance, and there is a certain distance at which you can attack the mob, but it can’t attack you. Learn how this distance looks visually and maintain it. It may be difficult to both keep distance and attack at the same time, so you can focus on the former first – it’s more important! This keeps you safe and should give you confidence so that you don’t panic and make more mistakes.

After that, you can add attacks. Each mob type has a fixed health that can be translated into the number of attacks. For example, a creeper can be killed with two iron axe crits and one normal attack (let’s call this 2.5), and a blaze can be killed in 1.5 hits. Instead of thinking about the goal as “kill a mob”, think about it as “deal X damage”. It’s simpler mechanically and it’s more short-term, which allows muscle memory to grasp it more easily. Once you do that and free up your consciousness, you can start counting the number of hits dealt to each mob if you’re facing multiple, and once you’re comfortable start relegating this to your muscle memory, and so on.

That’s a lot to learn for such a simple action! And this applies to everything – you can’t wing it and hope for the best, you have to understand it. But the benefit is that you know how to make things work out, so you stop panicking and making rushed decisions, and have more mental effort to allocate towards trickier tasks. The less stuff you have to do on the fly, the better – your end goal is to execute a perfect algorithm and reserve on-the-fly thinking to things that are really specific to the run.

List

Here are some other useful details, with shorter descriptions.

I’m probably missing a lot! I might add a couple more points if I think of something.

I learned most of this stuff from listening to top runners. They usually don’t explain stuff (save for Feinberg, I guess), but you can find some gems in training sessions for Midoffs and such; and for bastions specifically, Silverr’s tutorials have lots of useful details as well. Good luck :)