Week 1: Reward structures in games, and which ones a course can use

Six reward structures that games actually use, and an honest argument about which two of them could survive being pointed at a 130-hour course

22 February 2027

This course exists because of one lopsided fact, so we start there.

Putting 130 hours into a game is easy. Nobody plans it, nobody forces themselves, nobody finishes a ranked season by gritting their teeth — you queue again, and the hours go by. Putting the same 130 hours into a course is miserable, and most people cannot do it at all without a deadline standing over them.

Same hours. Same person. Often the same week. The only thing that differs is how quickly the activity pays you back, and that is a design property, not a fact about your character. Which means it can be built.

That is the entire course: build it, and finish a course the way you finish a game — a step at a time, absorbed, and without it costing everything you have.

So: what exactly is a game paying you, and how much of it can you take?

Why the obvious version fails#

“Gamify your studies” almost always means points stapled to the outside of work that has not changed. You still sit down to the same undifferentiated three hours; there is just now a streak counter judging you afterwards. It does not work, and the reason it does not work is worth naming precisely: the reward is about the work rather than inside it.

A game does not reward you for having played. It rewards you inside the playing, continuously, on rules you could recite.

The six families#

• Score. Arcade lineage. A number goes up as you act. Needs a machine that can tell, instantly and without argument, whether the thing you just did was good.
• The ranked ladder. MOBA and shooter lineage. Many short contests, a rating that survives between them. Needs a win condition per contest and a rating rule you trust.
• Boss gates. Souls and Metroidvania lineage. Progress is a sequence of discrete obstacles, each of which can be beaten and is then over. Needs every obstacle to have an end somebody can check. In a game they are placed for you; finding your own is week 8, and you find them by attacking rather than by listing them first.
• Collection. Achievement hunting, completionism. Needs somebody to have authored the list.
• Streaks and dailies. Free-to-play mobile lineage. Needs a threat — the streak’s whole power is that it can be broken.
• Execution mastery. Fighting games, rhythm games. Needs a long apprenticeship before the first payoff, and therefore needs you to already love it.

Which two survive#

Four of these die on contact with your situation.

Score needs an instant impartial judge, and for most study there isn’t one. Collection needs an author, and if you write the list yourself you will write one you can finish. Streaks work, but they work by threat, and a mechanism that punishes you for a bad week will be abandoned in the first bad week — which is exactly the week you needed it. Execution mastery is the trap this course exists to avoid: it is the model that requires you to be the kind of person who already puts in the hours.

That leaves the ranked ladder and the boss gate. The rest of the semester is those two, in that order, and the last three weeks are you running both at once on the course you stalled on and still need — the one you started, stopped, restarted and stopped again. Not a subject you dislike: that would be a cheap target, and getting through thirty hours of dull material proves much less than getting past a wall you have already failed at.

Three systems that failed, and the number that ended them#

I will spend some of this lecture on my own record, because it is the only evidence I have that the obvious approaches do not work.

I tried an effort currency — a point per hard idea got through — because an hour means different things on different days. It inflated. By the fourth month things that had not been worth a point were worth one, my daily totals were climbing, and my output was flat.

I tried decomposition, and got genuinely better at seeing what a problem was made of, and no better at doing any of it.

Then I tried timing everything, honestly, for nine months. Sixteen waking hours, six at the desk on a good day. I had guessed four, so the number was better than I feared, and it still changed nothing — because a timer measures how long you sat there.

All three measured what went in. That is the comfortable half, and it is the half that cannot tell you whether the semester is going well.

Before the lecture#

Work out, roughly, the feedback loop in one course you are taking right now: how many hours of your work pass between doing something and finding out whether it was any good. Bring the number in hours. Most people’s first honest answer is somewhere between forty and the whole semester.