Compare two scatters and fit both regression directions

The eleven paired points have a positive but visibly scattered linear trend. They are not all close to a perfect line, matching a coefficient between 0 and 1. The source prints $r=0.693$; the scatter comment and plot are authored from the exact pairs. The unrounded check is $0.6926559702\ldots$.

Use the first coordinates as explanatory and the second as response: Xlist contains $x$, Ylist contains $y$. The printed coefficients agree with independently calculated intercept $4.504442925\ldots$ and slope $0.6390977444\ldots$, rounded to 3 s.f.

Swap the explanatory and response lists. The printed reverse coefficients agree with intercept $4.421918908\ldots$ and slope $0.7507025291\ldots$. This is a new least-squares fit, not an algebraic inversion of the preceding equation.

The ten paired observations show almost no overall linear trend. The coefficient is very slightly negative and close to zero; this does not prove statistical independence or rule out every nonlinear relationship. Preserve all three distinct observations with $x=4$. The source prints $-0.0234$; the scatter explanation and plot are authored. The unrounded check is $-0.02338016784\ldots$.

The printed line agrees with independently calculated intercept $8.470628230\ldots$ and slope $-0.07478923035\ldots$. Its very shallow negative gradient matches the near-zero linear correlation.

The printed reverse line agrees with independently calculated intercept $3.010299003\ldots$ and slope $-0.007308970100\ldots$. Both negative slopes have the same sign as $r$, but neither is a measure of correlation strength on its own.