All models are wrong, but some are useful. Thinking of evolved features as having a purpose is wrong, but it is also incredibly useful.

Why do we have eyes? In some sense, there is no reason, just a sequence of random coincidences, combined with a slightly non-randon bias refered to as "survival of the fittest" (itself an incorrect model).

However, saying that we have eyes to see has incredible explanatory power, which makes it a useful model. Just like Newton's law of Universal gravity. We've known it that is wrong for a century at this point, but most of the time still talk as if it's true, because it is useful.

I hear what you're saying, and you're 100% correct, but I think most people will realize it's a figure of speech, and easier to say than "Via the process of gene mutation trial and error over many, many generations of tigers, spots have developed on their ears that look like eyes, resulting in predation from behind being discourged."

Yes, they did though. That's the purpose of this evolutionary trait. I see what you're getting at, but you seem to be implying this was a concidence

The spots might be helpful for baby tigers?