+1 I think we have covered the how they work/ how the OP protects to the downside. We haven't really covered the cost portion of the post because I/we can't quantify cost without knowing the underlying position. However, the above message in quotes definitely is getting into the weeds on the dynamics at play with option pricing.
How we doin OP?
Metoo
Selling puts will certainly provide some downside protection that the OP is looking for. However, when looking at this for a longer period of time (5 years or so), maybe selling most of it isn't such a bad idea? Mainly due to the costs.
Of course, at 95, we may unfortunately not be looking to a long term, but if in good health otherwise, five or six years is certainly possible ( about 30% for female, average health, non-smoker per
https://www.longevityillustrator.org ).
So while paying the 15% LTCG tax (on the gains, which is probably almost all gains?) when you could see the stepped up basis sounds painful, there is that looming fear of a drop in the stock/market. And the 15% rate holds for up to $445,850 taxable for single, so she could convert much of it in 5 years @ 15%.
As mentioned, the cost of the puts must be considered (I'd avoid a collar - you have to be careful, those can be assigned to you unexpectedly, then you have the cap gains anyhow). We don't know the name of the stock, but I'll assume it's an old "Blue Chip", probably similar to P&G? Using P&G as an example:
PG is currently ~ $143/share. To protect against a ~15% drop with a 120 put would cost ~ $3.40 (well, June, so 10 months, but close enough). That's ~ 2.4% per year, so 12% spent over 5 years. And you still take the first 15% loss. Let's say PG has dropped 15% at the end of 5 years - you've actually realized a 30% drop, 15% from the stock, and 15% from the cost of puts. So you would have limited your loss to 30%, as the puts would then increase pretty much dollar-for-dollar as the stock drops. But what if you need to buy them again at the end of that year? OK, that's kind of worst case, but it might help to understand where the limit edges are.
This is where "analysis paralysis" sets in for me, and I end up doing nothing. Which is probably the best course of (in)action.
-ERD50