The amount an individual point departs from the trend line is the important feature, not that the number above and below should be balanced.
A point that is much further away from the trend line is less likely to represent the process being plotted.
Each point is of course quite real, but each are influenced by measurement error, random processes within the experiment, recording (transcription) error, and so on.
If most of your points lie on a smooth curve, it is more likely that an outlier is subject to error.
I apologize, but I cannot see images or scatterplots. To determine the line of best fit, you typically look for the line that minimizes the distance between itself and all the points in the scatterplot, often using methods like least squares regression. If you can describe the scatterplot or provide data points, I can help you understand how to find the line of best fit.
7
above, below, and over the rainbow
below, it looks better
Above it
a line on a map joining points of equal height above or below sea level
True. In a data plot, the line of best fit represents the average trend of the data. Therefore, approximately half of the data points should lie below the line of best fit and half should lie above it if the data is evenly distributed.
Bisect two arcs above and below the given points or line and the perpendicular of these arcs cuts through the midpoint.
A line on a map joining points of equal height above or below sea level.
they should be below the shoulders, but above the waist :D
They should be above grade, below grade or filled with mortar.
If there is just one dot above or below any note it means the note is staccato which menas short and dettached.