... cluster.¹

I say semi-random because we do not want to take a control field that happens to fall within another stellar cluster region.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... test²

Numerical Recipes in C by William Press, 2nd Edition, pg. 623-625.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... field.³

Note that when we shift the main sequence by shifting the distance to the farthest star in the field, the lines are not being shifted in the J-K direction at all. This is because the magnitudes in J and K are both dependent on the $\log$ of the distance but independent of wavelength, so both will get shifted by an equal amount.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.