If one had complete clairvoyance and absolute understanding of every element in a system then there would be no estimates, no probabilities, merely outcomes foretold in advance.
For an estimate one needs to gather up what evidence and information is available; make a judgment on how well understood the system is, and its components; determine how predictable the behavior of the system has been in the past; figure out how close what is attempting to be done is to any previous testing or operations; and so on. Then take all of that and make an informed estimate of the likelihood of something occurring.
That's what an estimate is, a statement about how well a system seems to be understood and the probability of some event occurring. An estimate can build in uncertainty in the understanding of the system quite easily, by simply making a more cautious or pessimistic assumption.
SpaceX designed and built these rockets. They've flown them numerous times. They've done test flights exploring landing operations. They've done re-entry flights and landing attempts multiple times. All of these things increase confidence in understanding how the rocket operates under different conditions during landing, making it possible to come up with a reasonably informed estimate of the probability of a successful landing over a given number of attempts.
Perhaps they are wrong, perhaps they have made a key error in their modeling, who knows. That's why these are estimates, because there's always the possibility for unknown variables to affect outcomes. But to say that there is no basis for SpaceX's estimates is patently ridiculous.
I feel like a truer estimate would be "100% chance of success unless some unknown factor causes failure." It's saying that the unknown factor subtracts 20% that makes me go hmmm.
For an estimate one needs to gather up what evidence and information is available; make a judgment on how well understood the system is, and its components; determine how predictable the behavior of the system has been in the past; figure out how close what is attempting to be done is to any previous testing or operations; and so on. Then take all of that and make an informed estimate of the likelihood of something occurring.
That's what an estimate is, a statement about how well a system seems to be understood and the probability of some event occurring. An estimate can build in uncertainty in the understanding of the system quite easily, by simply making a more cautious or pessimistic assumption.
SpaceX designed and built these rockets. They've flown them numerous times. They've done test flights exploring landing operations. They've done re-entry flights and landing attempts multiple times. All of these things increase confidence in understanding how the rocket operates under different conditions during landing, making it possible to come up with a reasonably informed estimate of the probability of a successful landing over a given number of attempts.
Perhaps they are wrong, perhaps they have made a key error in their modeling, who knows. That's why these are estimates, because there's always the possibility for unknown variables to affect outcomes. But to say that there is no basis for SpaceX's estimates is patently ridiculous.