Interpreting Linear Models — I can interpret y=mx+b as a model and identify when extrapolation may be unreasonable.

People have been looking for patterns in numbers for thousands of years. Farmers tracked how much grain they grew each season. Merchants recorded how prices changed over time. The idea of drawing a straight line through data to predict what might happen next is actually ancient.

Over the centuries, mathematicians developed better tools for finding and using these patterns. The equation y = mx + b became the go-to formula for describing straight-line relationships. But scientists also learned an important lesson: a pattern that works for some values can give ridiculous answers if you push it too far.

So here's the big question this lesson answers: How do we use y = mx + b to describe real-world data, and how do we know when our predictions stop making sense?

Before we dive in, let's get clear on the key vocabulary. A linear model is a straight-line equation that represents real-world data. Think of it as a simplified version of reality — not perfect, but useful.

Let's look at a picture of what interpolation and extrapolation really mean. Imagine you collected data on how tall a plant grows over several weeks. The diagram below shows the data points, the best-fit line, and the danger zone where extrapolation gets unreasonable.

Notice how the real data points cluster near the line inside the green zone. That means our model fits the data well there. But look at the red zone on the right. The line keeps going up, but do plants really grow forever at the same rate? Of course not! At some point, growth slows down or stops. That's why extrapolation can be unreasonable.

The equation of a linear model is written in slope-intercept form. Let's break it down piece by piece.

Here's a real example. Suppose a streaming service charges a $3 base fee plus $2 for every movie you rent. We can model the total cost like this:

Now think about the slope's meaning in context. The slope m = 2 tells a story: "For every 1 additional movie you rent, your total cost goes up by $2." The y-intercept b = 3 also tells a story: "Even before renting any movies, you already owe $3."

Let's compare interpolation and extrapolation side by side. Imagine you tracked daily high temperatures for the first 10 days of June and found the model y = 1.5x + 72, where x is the day number and y is the temperature in °F.

A lemonade stand tracks sales over 6 Saturdays. They find that the linear model y = 12x + 10 describes their data, where x is the number of Saturdays since opening and y is the total number of cups sold that day. The data was collected for weeks 1 through 6.

Linear models are incredibly useful, but they aren't perfect. Here's an honest look at what they do well and where they fall short.

You've learned that linear models have limits. In future math classes, you'll meet other types of models that handle curves, not just straight lines. Here's a quick preview.

For now, the most important skill is knowing when a linear model fits your situation and when it doesn't. If data curves or levels off, a straight line won't tell the full story. You'll learn those tools in later courses — but the critical-thinking skill of checking whether your model makes sense? That starts right here.