The point is outliers in a dataset do not negate a trend line. They are already included in the trend line.

You haven’t established that the trend line exists or is applicable in this context. I don’t expect you to admit this.