The last thing we will cover about Excel is how to select a trendline type and what that actually means. It would be better to phrase that as “how to select the correct mathematical model”. Mathematical modeling is the process of using math to represent real-world phenomena. The reason the second terminology is better is because it correctly expresses the thinking that you must do to select a correct model. Before we jump into the different types of mathematical models (and associated trendlines) that you are expected to learn, we need to have a quick word on thinking.
There is an old saying in carpentry that you should measure twice and cut once. The idea is that you can’t un-cut a piece of wood so you better be sure you are cutting it correctly. While this analogy doesn’t perfectly translate to mathematical modeling (you could technically re-run the model), the reality is that once you choose your model and think it is good enough, it is unlikely you will re-scrutinize it.
The danger with mathematical modeling is that sometimes your data may look like they fit a certain type of model so you select that trendline type and move on. It is easier that way and uses less brain power. Humans are all lazy, and in fact, our laziness has led to some of our greatest inventions! However, your laziness can get the best of you when it comes to mathematical modeling.
You need to think about what you would expect the model would be off on your understanding of the underlying physics before you model it. What I am trying to convey is that you think, then choose your trendline type later. Think twice, model once will be our trendline selection mantra! Now that we have that out of the way, let’s look at the different model types that we will consider: linear, power, and exponential.
