The past constrains the future. And then what isn’t constrained will happen freely.

So yeah. There is no absolute determinism. But also, no absolute spontaneity. Given how much past has already accumulated, the world is highly organised and very restricted in the accidents that can occur.

You say there are futures where the laws can break down. You would have to be more clear about where physics might think this.

For a start, as the Universe gets ever more cold and dispersed, the chance of any spontaneous fluctuations dwindle accordingly. At infinite time, they would also be infinitely unlikely.

You could argue we are in a false vacuum state and so may plunge through to a lower true bottom level for some reason. But even that would only mean we didn’t understand the background story when we were describing the physical constraints producing a stable universe when writing down its “laws”.

If one is a pessimist then the future is easy to predict. If an optimist then it's really hard.

Whether this is due to the human nature (self included) or divine power is a question that remains to be answered.

You examine all patterns. See how the patterns influence other things and see how everything is connected. Have your mind overflowing with an endless stream of data which will periodically send you into an exsistional depression. Plan for all and every possible outcome and expect everything. Which might also lead to sleep problems. Examine every person and thing into traits and behaviors. Have everyone think you are a genius despite internally dealing with a feeling of being inadequate and imposter syndrome.

The only way to predict the future with certainty is to determine the possible futures. The only possible futures are those that are logically possible. This gives you too many possibilities because everything that doesn't entail a contradiction is logically possible. Given that there are so many possible futures you can't say that any one outcome is destined to happen.[....] — Purple Pond

It seems reasonable to suppose it's beyond our power to predict the future with certainty -- especially if we allow it's doubtful that we know the present or the past with certainty.

We can and do more or less reliably predict possible futures in terms of our estimation of what's "likely". We have a natural tendency to form reasonable expectations about the course of events in light of past experience, a knack we share with nonhuman animals like dogs. We refine (and confuse) our human power of prediction along with the rest of our peculiarly human conceptual capacities through the medium of culture.

In general, where we may speak of the mathematical probability of future scenarios, it seems there must be a historical record of similar past events, or some other means of analyzing the range of outcomes and their connection to the present. Short of such evidence, we may argue about the "reasonableness" of expectations, but in such cases our claims about the "reasonableness" of expectations cannot be evaluated in terms of mathematical probability. I suppose the term "likely" may be enlisted as an indeterminate middle between such uses of the terms "reasonable" and "probable".

I wonder whether what's called ancient "probabilism" is better construed as taking the reasonable as its standard, or the probable, or as riding roughshod over this distinction between two criteria of the likely.

This is quite a deep subject actually, it's one of the main topics that Aristotle and the Scholastics spent a lot of time thinking over.

The short answer is that the totality of logical possibilities is always constrained in some way (e.g. bullets don't turn into soap bubbles like in Infinity War), and that constraint represents things' nature or essence, which is basically a thing's set of potentialities when it encounters various contexts.

This gets rid of the problem of induction (which is probably why Aristotle said "induction is easy" and unproblematic). When you induct that something will happen in the future, given certain conditions, you're not basing your induction on past instances, you're basing it on the fact that you're pegging the entity with having a particular nature in terms of which only some actions are possible (and, in the concept, in the essence, logically necessary) under those conditions - and that would have been true in the past, is true now, and will be true in the future.

The difference I think we see with hindsight and a few hundred years of modern philosophy (i.e. where the old school essentialists were wrong), is that it was a mistake to think that essences are perceived (as inhering in the object, and then somehow we "read them off" from the object). Rather we punt them, and see what fits with eventuating behaviour (scientific method - science being "instructed common sense"). If a proposed essence fits the consequent behaviour of an object consistently, then it's permissible to rest our case and say "well, it looks like this thing is an x." (Although of course the case is kept on archive, should an anomaly crop up in the future.)

IOW, the old school essentialists were correct to say that things have natures and essences, particular ways they are and are not, particular sets of potentiality, and they were correct to say that what we're doing with cognition is trying to ken those essences. But they were wrong to say that we can "read off" those essences from the objects (the same plain way we'd see the colour of something). There is no necessary connection between what we say and how things are, it's always a guess that never leaves the logical status of conjecture, even if strongly corroborated.

What does it mean when someone says "the laws of physics break down"? When a physical law is abandoned in favor of a new one, we don't say that the laws of physics are breaking. What is happening is that our understanding is evolving.

Possibilities exist only in our minds as the result of uncertainty. There are no possibilities out there only deterministic events of cause and effect. Randomness, uncertainty and possibilities go hand-in-hand as the effect of our own ignorance.

Logic only fails when you don't put all relevant information into the logical system. When we use logic and get the wrong answer, it isn't the fault of logic. It is our fault for not inputting the correct and relevant information into a logical system.