Well you would probably know better than me, but my understanding is that probability of survival is logrithmically better with both low speed and also with low impact angle. Any time both can't be assured in my opinion would present a strong case to at least consider the 'chute handle. Whether the issue is preventable/likely doesn't mean it's not possible and this is where the risk tolerance bit comes in IMO. I don't think the decision is clear-cut, but perhaps rather, which is less bad. Loss of SA in IMC, Loss of engine low IMC or inhospitable terrain, flight control jam/failure, flutter/important part of airplane breaks off, etc are what first comes to mind when I would at least consider the 'chute. Each situation would be different of course...
If you can easily calculate the deployment envelope more accurately, that would be great! (I know this is in the RV-10 section), but it would be awesome if you can PM with what you think for an 1800lb GW RV that I'm building.
To your first point, what is the ratio of chute equipped aircraft to all others? Something to think about. Several of your reasons to deploy were mentioned. In RVs, how many flight control jam or failures have there been, where a chute would deploy successfully? How many flutter events or "important parts break off" events have there been in RVs?
I am certainly NOT saying that a chute is necessarily a bad idea but the failures you are citing are very rare, if even existent, in the RV community.
As far as the deployment altitude, pick a failure, let's say complete engine failure. You will have a rate of descent, call it around 1000 fpm. How long will it take you to troubleshoot the problem, AFTER the startle factor. Don't use the "3 second" startle factor; after observing EAA's simulator scenario, it isn't realistic, and the time is usually much longer. How much time are you burning in making the decision to deploy, realizing you are descending the whole time? Now that you have made the decision to deploy and pull the handle, physics take over. BRS has the numbers for the time it takes for the rocket to fire and pull the chute. Then there is the required time for the chute to actually open and hopefully stabilize the descent rate. BRS also has the numbers for the stabilized rate of descent. If you research it, you may find that it is actually more than you would think. So, add up all the times and, using your initial descent rate, calculate the altitude that you have lost in that time. You will then have an idea of the minimum successful deployment altitude. I can guarantee you that it will be higher than what a test pilot, knowing that it is going to happen, with the decision to deploy already made, will post.
This is also one of the best cases; do the same analysis with your example of losing SA in IMC. What do you think the descent rate will be in a steep spiral? Guarantee it will be more than 1000 fpm!
These are just some things to consider. Again, NOT saying to disregard the chute if it makes you more comfortable...
Like I said, this topic always generates some very interesting conversations but I think, more importantly, it causes people to think more about it.