Here's our editorial take on this.
When your trendline equation doesn't match the data, the instinct is to blame the tool. In this case, Excel appears to be the culprit. But the real problem isn't Excel's math, it's the assumption that a simple right-click will give you a useful equation without understanding what the chart is actually showing you. The user expected y = -1.37x + 102. Excel returned y = -27.433x + 129.83. That is not a rounding error. That is a fundamental mismatch, and it points to a common misunderstanding of how trendline equations work in practice.
The most likely explanation is that Excel is treating the x-axis as categorical, not numerical. When you plot data with text-based labels or non-contiguous numeric values, Excel assigns each point an integer index starting at 1, regardless of the actual x-values. So the equation it calculates is based on those indices, not the real numbers you intended. If your x-axis runs from 1 to 100, but Excel sees it as 1 to 5, the slope will be wildly different. The user's expected slope of -1.37 suggests their actual x-values are in a reasonable range. Excel's -27.43 suggests it is counting points, not values. That is the disconnect. The tool did what it was told, but it was told the wrong story.
For anyone relying on trendline equations for further calculations, this is a practical warning. The equation displayed on the chart is only as trustworthy as the chart's axis configuration. If you paste that equation into a cell and get nonsense, check whether your x-axis is truly numeric. A quick fix is to use the LINEST function directly on your data, which bypasses the chart entirely and gives you the coefficients based on the actual numbers. Or, ensure your chart uses a scatter plot rather than a line chart, because scatter plots treat x-values as continuous data by default. The line chart is designed for category comparisons, not mathematical modeling.
Our opinion is straightforward: do not trust a trendline equation without verifying its axis. The feature is powerful, but it assumes you have set up the chart correctly. This user's frustration is valid, but the solution is not a better tool, it is a better understanding of the tool they already have. If you need equations that actually reflect your data, skip the chart and go straight to the worksheet functions. That is where the math lives, without the visual layer getting in the way.