does the least square method equally devide the points in two? in general i don't think so... (altough i can't prove it

) (my feeling is that, since the LSM is about the deviation of the line you draw, it doesn't say anything about the number of points)
what i think is the correct way is this:
- devide the set with a vertical line exactly in 2.
- Compute for each set the mean point (<x>,<y>)(that is: <x> = (x1+x2+x3....+xn)/n), idem for y)
- the straight line between the 2 mean points divides the points exactly in 2.
i'm quite sure this is the right way; a formal proof of why this works is tricky... (maybe later more

)
Bo