If you do that then there will be reflections causing the displayed voltage to be frequency dependent. The frequency of the nulls depends on the length of the piece of coax, shorter->higher frequency of first null. Use an online VSWR calculator to find those frequencies; better simulate it with a Spice or any of the online equivalents.
Didn't thought of this to be honest. I just compared it with the ground spring and lead, and the results was more of the range of the spring, but without needing a third arm to hold everything in place
But I will have a look into that, thanks.
The other, obvious, point is that coax will load your circuit. For low frequencies that will appear to be a capacitance, for high frequencies it will be the characteristic impedance, 50ohms.
For a standard "high" impedance *10 probe, it is worth calculating the tip impedance (i.e. R//C) at relevant frequencies. Relevant for a digital signal is determined solely by risetime.
Then calculate the same for a "low" impedance *10 resistive divider probe. Note the quotation marks.
Have a look at the references in
https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/