Complexity isn't difficulty

The game I spend the most hours in over the last few years is Timberborn. It's a colony management sim where you control a settlement of beavers. Because it's a colony a beavers the main focus of the game is water management, you'd have to build dams and storage to preserve water during droughts and redirect entire rivers during a 'bad tide' when toxic liquid replaces all water.

How beavers manage

The thing I noticed while playing the game is how the difficulty curve has a clear downward slope. At first you struggle to survive at first, using limited tools to keep your crops watered and beaver's thirst quenched. Your small colony has to prioritize between gathering resources, like food and water, and building infrastructure, like dams and storage, to build a stockpile for the inevitable drought or bad tide. And the game gives you all the tools you need for this: a gatherer flag to gather berries, a farm to grow carrots, a woodcutter flag to chop wood, a dam to dam of water, a log pump to product water, and storage for everything you need.

But those are only your short term goals, for your short term goal you have to sacrifice precious labor for 'science' in research huts. So you have beavers not actually produce anything directly tied to survival, but investing in the future.

Evolving complexity

In the first few hours the game is a constant struggle, choosing short term survival and long term prosperity. During the game it quickly becomes clear that you need science. Without science your beaver are able to cut down trees, but there are only a limited supply available to your colony. To get more you have two choices: Either you build stairs to be able to climb to new trees, or you build a forester. Both of these require you to make planks. To make planks you need a woodcutter, which you don't have access to from the start. So this is where the science and researchers become important. Then you can build the woodcutter, power it using a waterwheel or 'hamster' wheel and start producing planks.

With the forester you colony can (almost) become self sufficient, and you're more than ready for the next drought. But the one after that will be longer, and after that even longer. Soon the water you can safe won't be enough to last a drought and you'd need larger dams and stores. This leads to more research and building. Then, when you're feeling safe, the first bad tide hits. And none of the infrastructure you build was prepared for this. Your crops die out, beavers get sick, and you run out of water. Just dams aren't enough, you need to invest in sluices or floodgates to survive.

This is the core gameplay loop of the game, you have a set of tools, and by the time you've used them you need the next. Each time survival does become easier the more tools you unlock. But as the difficulty drops the complexity grows. You're not only managing some water pumps, farms, and woodcutter. You'll have an industrial district which requires power and entire supply chains of food production.

Changing the challenge

During a regular game of Timberborn the challenge slowly shifts. At first the challenge is difficulty, with a limited set of tool you struggle to survive. You juggle different needs, both short term and long term, to ensure having plenty of food and water for your colony.

Once you master the difficulty and you have a thriving beaver colony, survival becomes almost trivial. But this is where the complexity takes over, instead of a limited set of tool you've now unlocked a wide selection of connected tools you have to manage. The challenge isn't as much gathering resources but the management of those resources.

Estimating challenge

This leads to the point I'm trying to make: when estimating the challenge of a task, there is a difference between the 'difficulty' and the 'complexity' of a task. When estimating 'story points' for a task we have to, as a team, decide what those represent. And no matter how we estimate a given task, hours spend on it are what's tracked. We always reiterate that complexity doesn't equal time, we can somehow arbitrarily decide how many points of complexity can fit in a duration.

To use a more facetious than a video game, let's take the task "throw a basketball through the hoop from the center of the court". When estimating this as a 'complex' task, we can see that it isn't complex at all. We have a basketball, a court, and a hoop. No variables or hidden conditions. But if we take the difficulty of the task in account the story changes a lot. At least for me, somebody who hasn't even touched a basketball in the past 8 years. Just because the task isn't complex, that doesn't mean it will be quick or easy.

In the end it doesn't actually matter which metric you use to estimate a story, as long as it works for the team as a whole. Do you prefer hours so it's easier to communicate? Go for it! Or if you prefer something more abstract like complexity? Be my guest! Do you feel the need to devise your own mathematical formula that takes the cups of coffee per task in account? Sure, if your team and stakeholder agree!

Powered by wisp