Thursday, September 27, 2018

The Relationship Between Playtime and Skill

Disclaimer: I'm not going super deep into all the factors that might cause variation in how fast players gain skill. I'm just taking it as a given that you'll agree there is variation in progress. Here, the variation is simulated with RNG.

Sometimes players will ask things like, "for [X] hours, what rank should I be?" or "how long does it usually take to get champion?"

The common reply to this, if it's asked on Reddit, is this:






This prompts a bot to make a post with a link to a video of a pro (CJCJ) explaining how people all make progress at different and varying rates. The video does a good job with its explanation, but it's wrong. The only factors he holds constant are time and skill, which is leaving a lot of the picture out.

In almost all cases, when someone is learning a skill, their progress will follow what we can just call a sort of "natural law", and this is how it works:


  • Progress is always fastest when beginning to learn a skill. This is pretty self-evident if you take a moment to consider any of your learning experiences. For example, take typing. If you're typing at 10 words per minute when you start, it's not hard to double that in a day. Conversely, when you've been practicing for a while you can't just double your typing speed--even in a week. There's just not as much room for growth once you've started.


This omission is why CJCJ's advice falls a little short: an incredibly common pattern is not taken into account.

However, we can use a bit of math (which is available if anyone really wants it) to simulate how progress might actually look over time. Here is a graph of a single function which uses randomly generated data, and which also takes the variable rate of progress into account. These numbers are fictitious, but the point is not.

With this picture of how things work, we can actually say another thing about progress over time. Looking at the entire graph, the line of progress is more or less smooth. However, if we take only a small section of time, there are obvious trajectories--things are more line-like than curved. In the big picture it's clear that these don't mean anything (they're just part of the big line that slowly curves), but up close they can be easily misinterpreted. A spike might seem to indicate that the rate of progress is increasing, but it could really just be a hiccup.

Hopefully this helps those currently frustrated with slumps for now, but when it's out, my new video will go much more in depth on the issue.

Getting Started

I've never used a blogging site before, so I'm stuck figuring things out as I go.

The reason I'm starting this here is that as head coach of Team Meteor (TM), I need a platform to organize my thoughts and content. I had been using Google Docs, but there were pervasive formatting errors for those that don't have the app on mobile. It's possible that when the TM website is complete, I will move everything over and reorganize it, but that's a problem for another day.