The thing to remember about this topic is that the term "unstable" gets thrown around a lot with minimal qualification as to how it's really being used. The term "unstable" should be really defined as causing effects such as keyholing or oblong holes in the target. Bryan Litz used to recommend a minimum gyroscopic stability coefficient (Sg) of 1.40 as being sufficient to garner the full intrinsic BC of a given bullet. He later increased that recommendation to an Sg of 1.50. Some have interpreted achieving an Sg sufficient to obtain the full intrinsic BC of a bullet as "stability", when it's not necessarily the same thing in practical terms. Obviously, a twist rate necessary to achieve an Sg of 1.50 would be faster than that necessary to achieve an Sg of 1.40. However, running a given bullet with an Sg as low as 1.25 or 1.20 doesn't necessarily mean the bullet will fly erratically, potentially keyholing at the target. It generally does mean the bullet will behave as though it had a lower BC than the advertised value. However, someone running a bullet with an Sg of 1.25 may never actually observe oblong holes or keyholes at the target. So there is a gyroscopic stability coefficient window in which the bullet will still fly with sufficient stability to reach the target without having an erratic trajectory, but will likely have an effective BC slightly lower than expected. That does not mean it is "unstable". When the Sg gets down around the 1.10 range (or less), that is when major stability problems such as oblong holes or keyholing usually start to show up. Nonetheless, you can often run a barrel twist rate with a given bullet that generates an Sg in the 1.20 to 1.30 range, and the bullet will remain stable; however, it will behave as though the BC is lower than it really is.
As elevation increases, the twist rate necessary to stabilize a given bullet to the target will decrease slightly. Along the same line, less elevation travel will be required to zero at a given distance, again due atmospheric effects. The best thing to do is run a given bullet at the known elevation (altitude) and expected temperature, together with the known velocity and barrel twist rate, using a program such as Berger's Twist Rate Calculator (
https://bergerbullets.com/twist-rate-calculator/). Although such programs may not be "perfect", they usually give a pretty good idea of the relative Sg values that will be generated for a given bullet/rifle setup in the given conditions. The good part is that it can be done entirely
in silico, only requiring the necessary input values. One can play around with differing elevation, temperature, twist rates, etc., then look at the predicted effect on the Sg and BC (outputs). Even if a bullet is predicted to have an Sg of of 1.20 to 1.30 from a certain twist rate barrel under specific conditions, that does not mean it will necessarily be "unstable". Certainly values closer to 1.50 are better, but the bullet will likely still fly true, merely behaving as though its BC was slightly lower than expected. Thus, a Twist Rate Calculator is a useful tool to compare various possible scenarios and come up with a pretty good idea of whether a given bullet will really be unstable (i.e. having a predicted Sg of less than 1.10 to 1.20), or merely behave as though the BC was a little lower than expected.