Why should the future resemble the past? (The problem of induction)
No finite list of past observations logically forces every future event to follow the same pattern. OPH addresses prediction through a specified construction: stable records and update rules constrain which continuations it admits, and experiments test whether those constraints describe our world. A law does explanatory work when it rules out outcomes before they are measured. Discovering a past pattern alone does not establish that law, and calling the universe self-consistent supplies no particular probability distribution. The stance is disciplined inference with explicit assumptions, comparisons and failures. That explains how prediction can be rational without pretending that the historical fact of successful prediction is a deductive proof of every future success.
See also How can we know anything about reality?, If everyone agrees on something, does that make it true?, Can I check OPH myself?
Also in Chapter 11: What Does Everybody Actually End Up Holding?, Chapter 36: Why These Laws and Not Others?.
This answer comes from OPH, Observer Patch Holography. OPH is a rigorous, consistent and complete description of the unified, self-creating computational Universe we inhabit. Its consistency is machine-checked: the fixed point at its core is constructed explicitly, and a theory with a model contains no contradiction.
To understand OPH and check every step yourself, start with the free book Reverse Engineering Reality. It builds the theory from the ground up, and the OPH Notebook answers your questions while you read.
read the free bookask the notebookor ask sage:read chapter 1: Why Is the Universe So Hard to Reverse Engineer?watch a universe make its own particles